ArticlePDF Available

Robotic Warfare, Human Rights & the Rhetorics of Ethical Machines



Killing with robots is no more a future scenario but became a reality in the first decade of the 21 st century. The U.S. and Israel forces are using uninhabited combat aerial vehicles (UCAVs) in their so-called wars on terror, especially for targeted killing missions in Iraq, Pakistan, Afghanistan as well as in Lebanon and the Palestinian occupied territories (for example in Israel's recent war on Gaza). In the last years, the number of UCAV air attacks is rising significantly as well as the number of killed civilians. Nevertheless, the automation of warfare is envisioned by the US government and military for 2032 at the latest and military robots are increasingly used in civilian contexts. In the face of these developments, discussions on robotic warfare as well as security technology from a science and technology studies and technoethical perspective are highly needed. Important questions are how robotic warfare and security applications may find their way into society on a broad scale and whether this might lead to a new global arms race, violation of the international law of warfare, an increasing endangerment of civilians transporting racist and sexist implications, and the blurring of boundaries between military, police and civil society.
Robotic Warfare, Human Rights & the Rhetorics of Ethical Machines
Jutta Weber
to appear in:
Rafael Capurro / Michael Nagenborg / Giugelmo Tamburinni (eds.): Ethics and Robotics.
Deutscher Akademieverlag (in preparation)
Killing with robots is no more a future scenario but became a reality in the first decade of the
21st century. The U.S. and Israel forces are using uninhabited combat aerial vehicles
(UCAVs) in their so-called wars on terror, especially for targeted killing missions in Iraq,
Pakistan, Afghanistan as well as in Lebanon and the Palestinian occupied territories (for
example in Israel’s recent war on Gaza). In the last years, the number of UCAV air attacks is
rising significantly as well as the number of killed civilians. Nevertheless, the automation of
warfare is envisioned by the US government and military for 2032 at the latest and military
robots are increasingly used in civilian contexts. In the face of these developments,
discussions on robotic warfare as well as security technology from a science and technology
studies and technoethical perspective are highly needed. Important questions are how robotic
warfare and security applications may find their way into society on a broad scale and
whether this might lead to a new global arms race, violation of the international law of
warfare, an increasing endangerment of civilians transporting racist and sexist implications,
and the blurring of boundaries between military, police and civil society.
The Iraq and Afghan wars can be seen as a test bed for the development of U.S. military
robots as well as robotic warfare. While ground and water combat robots are still under
development, unmanned combat aerial vehicles are already used widely. For example, today
tele-operated uninhabited combat aerial vehicles (UCAVs) are regularly and in increasing
numbers deployed by the US and NATO forces in the Iraq and Afghanistan wars (Cordesman
2008). The numbers of civilians killed in these wars, probably also because of the use of these
devices, are rising (Fischer 2008)1. At the same time, research on autonomous uninhabited
systems that can start, land, monitor and kill people without a human in the loop are
generously funded by many states including USA, Germany, France and other EU states as
well as Israel (see for example Department of Defense 2007).
These developments raise serious questions concerning international law, that is disarmament
agreements, law of armed conflict, and human rights. For example, experts point out that the
spread of robotic weapon systems might lead to a new global arms race as well as to the
lowering of the threshold for entering into war (see Sparrow 2007, Altmann in this volume).
Another issue is the blurring of boundaries between the military and the police by new and
emerging technologies deployed in both contexts (for example, UCAVs for the surveillance of
national borders). The bi-directional use of military / security robots opens up critical
juridical, political, and social questions.
It is quite astonishing that in the face of these developments there are up to now only rarely
discussions on robotic warfare as well as robot security technology. We need a close look
from science and technology studies as well as (techno)ethics perspectives2 to see whether
robotic warfare and security applications may find their way into society on a broad scale –
for example by causing a new global arms race, by violating international law of war by
heightening the endangerment of civilians (Boes 2005, Rötzer 2007a, 2007b, Sparrow 2007),
and blurring of the boundaries between military, police and civilian tasks or opening up
opportunities to use killer robots for crimes (Miasnikov 2004, 2007; Altmann 2006).
This paper will sketch some recent UCAV developments and deployments by US, NATO,
Israel and European forces and their ethical, political, and sociotechnical implications.
Problems of future war scenarios are outlined with regard to human rights and international
law issues. Technophilic imaginaries linked to the ‘Robowar Dreams’ (Graham 2007),
‘humane’ warfare as well as rhetorics of a possible ethics of future autonomous robotic
systems are discussed and recommendations are given.
Uninhabited Aerial Vehicles – Forerunner of Future Robotic Weapons
Uninhabited aerial vehicles (UAVs) have been used for surveillance since the Vietnam War,
some nations are now developing and deploying combat UAVs. Especially the U.S. and
Israeli3 forces are using uninhabited combat aerial vehicles (UCAVs) for so-called ‘targeted
killing’ missions. Most of them were executed in Iraq, Pakistan and Afghanistan respectively
1 See also the appendix with examples of UCAV air strikes and ‘collateral damages’ respectively casualities in Afghanistan, Iraq and
2 See for example Cerqui et al. 2006, von Schomberg 2006
3 Khalifa 2008
in the Palestinian occupied territories or in Lebanon. Especially in Iraq, Pakistan and
Afghanistan, the number of UCAVs air attacks is significantly rising4 and – despite the
rhetoric of ‘precision strikes’ – the number of killed civilians as well. Lately, many little
villages in Southern and Northern Waziristan - an area in the North of Pakistan close to the
Afghan border – have been destroyed by US and NATO UCAVs, weddings have been
bombarded and school and other civilian houses destroyed.
Between 2004 und 2007 the number of US air strikes rose from 285 to 1119 per year in Iraq
and from 6495 auf 12.775 in Afghanistan. At the same time, the number of flying hours of
uninhabited combat aerial vehicles (UCAVs) tripled between 2003 und 2007, while the
number of surveillance flights in both countries rose only very slightly. Therefore it is very
likely that air attacks by uninhabited combat aerial vehicles massively increased lately (see
also Cordesman 2008, Rötzer 2008). And the numbers of so-called ‘collateral damages’ are
very high5. The BBC reported that in Afghanistan “civilian casualties caused by pro-
government forces are rising – 577 so far this year, compared with 477 over the same period
last year. Over two-thirds were caused by air strikes and the UN is calling for an independent
assessment of damage, so that survivors and relatives can be compensated.”
( Not all air strikes are undertaken by
UCAVs but – as I said before – the numbers of UCAV strikes are rising as well as those of
civilian casualities, so that a causal connection between these developments seems quite likely
(Boes 2005, Rötzer 2007a, 2007b, Sparrow 2007).
Robot Wars and UCAVs
Despite the increase of killed civilian victims by robotic warfare, armed forces and politicians
are pushing the development of military robots in general and UCAVs in particular. The USA
military today spends two thirds of the global expenditure for military R&D (Brzoska 2006,
Altmann in this volume). It is no surprise that it is also the leading force in the development of
combat robots. In 2001, the US Congress decided that the armed forces should implement
“remotely controlled technology such that (1), in 2010 one-third of the aircraft in the
operational deep strike force aircraft fleet are unmanned; and (2) by 2015, one-third of the
4 See below
5 To give one example, Anthony Lloyd of ‘The Times’, reported from Kandahar on May 24, 2007 about the near obliteration of the village of
Gurmaw on the night of May 8, 2007: “ Mr Lalai’s village, a settlement in the Sarwan Qala valley north of Sangin, which is patrolled by
British troops, was bombed by aircraft on the night of May 8 after fighting between the Taliban and foreign soldiers. Crawling wounded
from the wreckage of his home, Mr Lalai discovered that his grandfather, grandmother, wife, father, three brothers and four sisters had died
in the bombing. The youngest victim was 8, the oldest 80. Only Mr Lalai’s mother and two sons, aged 5 and 3, survived. Both boys were
wounded. Yet the forces that wiped out his family were not British, nor those of any other Nato unit. The airstrikes were called in by
American Special Forces operating with their own rules of engagement on a mission totally devolved from Nato command in Afghanistan.
At least 21 Afghan civilians died in the bombing of Gurmaw.” ( In the
New York Times Carlotta Gall reported that according to phone calls she had with local residents the toll was much higher involving 56-80
operational ground combat vehicles are unmanned. (US Congress 2000, 38). An
outcome of this decision was the largest technology project in history, the U.S. Future Combat
Systems (FCS) - a $127-billion project – which includes uninhabited aerial and ground
vehicles, inhabited vehicles, unattended sensors, new munitions, launchers, and a network for
communication and data-sharing between all FCS elements (Marte / Szabo 2007). This
program was mostly substituted by the Joint Robotics Program Master Plan in 2005. In
December 2007 the ‘Unmanned Systems Roadmap 2007-2032’ was published by the US
Department of Defence, which frames the development of robotic systems for the next 25
years. Until 2013, 21 billion dollars are planned for research, development, supply and
deployment of uninhabited systems (air, water and ground) But not only the US forces are
pushing the development of military robot systems. Today, more than 50 countries all over
the world are working on the development of uninhabited systems (Warren 2007, Jane’s
Uninhabited Air Vehicles are the most deployed military robots today. These aircraft can be
operated remotely controlled and (partly) autonomously. They – and especially the future
autonomous UCAVs – are predicted to be the future of military aircraft (Department of
Defence 2007, Sparrow 2007). Ground combat unmanned vehicles are not yet systematically
deployed and still need further research & development.
UCAVs have three components: an airplane with sensors and (partly) with weapon systems, a
ground control station from which the plane is tele-operated and a communication
infrastructure such as radio communication, laser or satellite link. As Altmann points out,
‘flight control is done by on-board processing, but general directions and in particular attack
decisions are given by remote control, often via satellite link from hundreds to many
thousands of kilometres away.” (Altmann in this volume) Following the dominant logic of
information warfare, the video images produced by UCAVs can be transferred to ground
troops, helicopters or ground vehicles. Some of the best known UCAVs which are already in
production are the MQ-1 Predator and the MQ-9 Reaper of the US Air Force. In the 1990s,
UAVs such as the MQ-1 Predator – then RQ-1 – were primarily used for surveillance. In
2001 they were retrofitted with missiles (for example, air-to-ground AGM-114 Hellfire or
AIM-92-Stinger air-to-air-missiles).
Uninhabited aerial vehicles for surveillance were extensively used in NATO military
operations in Kosovo and were and are regularly deployed and used also for combat by the
U.S. Forces in the Afghanistan and Iraq wars (Barry/Zimet 2001; Sparrow 2007).
In May 2007, the U.S. Forces formed their first uninhabited combat aircraft wing. The 432nd
Wing of the Air Force consists of six operations squadrons and a maintenance squadron of 60
MQ-1 Predator and six MQ-9 Reaper – a wing with huge bombing power up to 1.7 t. The
MQ-9 Reaper is an up-graded version of the UCAV MQ-1 Predator, with 11 meters length
and 20 meters wingspan. Possible payload mass is 1702 kg. The MQ-9 Reaper is capable of
14 hours non-stop flying – the F-16 is capable of 2 hours flying but at much faster speed. MQ-
9’s maximum speed is 400 km/h, service ceiling is 15.000 meters. Most of these UCAVs are
flown from bases in the United States – about 12.000 kilometers away; only take-off and
landing is operated from Afghan or Iraq bases. The tactical aim of UCAVs is to hold a huge
amount of ammunition on call for short-notice strikes – especially for targeted killing
missions and ‘precision attacks’ and thereby to combine surveillance and combat tasks. One
Reaper system (a ground station and 4 planes) costs about 69 million dollars.
The ‘Unmanned Systems Roadmap’ of the U.S. Department of Defence states that the latest
US wars have been a most welcome test bed for the weapon technologies, not only for
engineers and military strategists but also for the development and fund raising: “For defense-
related unmanned systems, the series of regional conflicts in which the United States has been
engaged since the end of the Cold War has served to introduce and expand the capabilities of
unmanned systems technology to war fighters. This conflict-driven demand has ensured the
technology’s evolution and continued funding, with each new conflict reinforcing the interest
in such systems. Global Hawk owes its appearance over Afghanistan to the performance of
Predator over Bosnia and Kosovo, which in turn owes its start to the record establishes by
Pioneer in the Persian Gulf War.” (Department of Defence 2007, 47).
High-Tech Military Robots for Europe
The UK has ordered three MQ-9 Reapers from the USA for its Royal Air Force (Hanley
2007). Since 2004, Predators are used by the Italian Air Force and since 2006 by the Royal
Air Force. In August 2008 Germany made a request at the US Department of Defense for 5
MQ-9 Reapers ( At
least one Predator is also used by the Pakistan Air Force (Rötzer 2007b).
The air forces of the UK, Italy, Germany and some other European countries also deploy
uninhabited aerial vehicles and develop first prototypes – technology demonstrators – of
uninhabited combat aerial vehicles6. In 2006 France, Greece, Italy, Sweden, Spain and
Switzerland started to build an uninhabited combat aerial vehicle called ‘Neuron’ to be
6 The early UAVs were controlled by remote control. Full autonomy of the aerial vehicles was developed later. It is probably the
case today that aerial vehicles can easily be switched from the remote control mode to one of full autonomy.
finalized in 2011 (Johansen 2007). In Germany, the UCAV demonstrator Barracuda –
developed in 2006 – crashed soon after its public presentation into the Mediterranean Sea
because of software problems. The project was stopped thereafter. In December 2007, the
European Aeronautic Defence and Space Company (EADS) announced the project Barracuda
II as part of the project "Agile UAV in Network-Centric Environments" initiated by German
Ministry of Defence. At the same time, EADS was awarded a 60-million contract by the
German, French and Spanish governments to develop a concept for a ‚network centric warfare
network’. This network is supposed to be a common platform with mobile ground stations for
new modular family of German, French and Spanish UAVs that cooperate in swarms, learn
and transfer their information to further systems.
Hermes kills in Lebanon and Palestine
The second biggest developer of UCAVs is Israel. Israel deployed UCAVs from the Hermes
series (Elbit Systems Ltd.) in 2006 in the war against Lebanon, but also for surveillance,
targeted killing and war operations in the West Bank and the Gaza strip. Between 2000 and
2006 three hundred people characterized as terrorists were killed – together with 129 civilians
(Case 2008) Israeli Human Rights Groups filed a lawsuit against the Israeli Government.
They claimed that according to Israeli as well as international law ‘targeted killing’ is an
illegal use of force. This behaviour could be compared with those of policemen who kill the
suspect instead of arresting him or her. “In December 2006, the Israeli Supreme Court issued
a landmark decision in the case. While the court stopped short of an outright ban on Israel's
assassinations program, it ruled that international law constrains the targeting of terror
suspects. Currently, in order to justify a strike, Israel must have reliable information that the
suspect is actively engaged in hostilities (such as planning a terrorist attack) and must rule out
an arrest as being too risky. The court also requires that there be an independent investigation
after each strike.” (Case 2008) US Human Right Groups did not go to court because they say
that U.S. courts rarely dare to challenge the president’s national security policy.
The killing of civilians was not an issue in this lawsuit: According to international law
standards it is necessary to distinguish clearly between combatants and non-combatants,
between military and civilian targets, and there has to be a proportionality of force with regard
to so-called collateral damage. The intentional killing of civilians or the non-proportional
injury or killing of civilians is regarded as a war crime. Regrettably, proportionality between
the military aims and the injury and death of civilians in this context is defined only vaguely.
The Price of New Warfare Scenarios: On Racism, Sexism, & Cost-Efficiency
Given this background, military forces proceed to rely increasingly on UCAVs in their ‘war
on terror’ in Afghanistan, Pakistan, Iraq and Gaza (Mellenthin 2009). With nationalist rhetoric
these systems are praised as the remedy to save the lives of one’s own soldiers. For example,
Lin et al. 2009 (in this volume) write: “Instead of our soldiers returning home in flag-draped
caskets to heartbroken parents, autonomous robots […] can replace the human soldier in an
increasing range of dangerous missions.”
This approach relies on a problematic ontological stance. Obviously, the priority is to save the
lives of one’s own soldiers. There is less or no concern for the humanitarian costs of these
new technologies with regard to the non-combatants of other (low-tech) nations from the
South. Despite the common rhetoric of ‘precision air strikes’ by the military and media, the
deployment of UCAVs using bombs and missiles for targeted killing costs the lives of a
growing number of civilians. There seems to be an underlying racism and partially sexism
that takes it as obvious that US (or NATO) soldiers are of much higher value than Afghan or
Iraqi civilians – which means women, children or elderly people. Despite the fact that the
attacks affect also some men and boys as well and US militaries include female soldiers, the
sexism of this warfare politics lies in the structural effects of military politics and the politics
of international relations (Tickner 2004). Both ignore to a wide extent the different situation
and needs of women and children which leads to much more severe effects of war and
conflict on women (Moore 2007) There are still astonishingly few discussions of the racist
and sexist implications of the different valuation of the lives of Western troops and non-
Westerns combatants as well as civilians (Butler 2008, Herold 2008). Sometimes, also the
illusion is evoked that the coming wars will be robot wars only.
Efforts to overcome legal and techno-ethical limitations are also already under way. For
example, John Canning from the Naval Surface Warfare Center proposes to use armed
autonomous systems without a human-in-the loop – who is in his view always a
“’performance- and cost-killer’ – when considering the employment of large numbers of
armed uninhabited systems” (Canning 2007, 11). He recommends that autonomous machines
should only target machines, while men target men thereby overcoming political and legal
ramifications of the use of armed autonomous systems. Autonomous systems should be built
with a switch between an autonomy mode and a remote-control mode. “An enemy would then
have a choice of abandoning his weapon and living, or continue using it, and dying.”
(Canning 2007, 31) This seems to be a quite unrealistic proposal. The impression is evoked
that warfare with autonomous weapons will be mostly a robot war only – machines only
fighting machines. Canning also proposes to equip autonomous weapons with video cameras
in case the system is hacked by the enemy and used to kill the wrong people. That way one
could give direct evidence of the guilt of hostile forces (Canning 2007, 30). I do not think that
such a high risk is tolerable with regard to civilians.
What is mostly overlooked in this context is the fact that the military also is intrigued by the
cost-efficiency of military robots. UCAVs are much cheaper than jet bombers and the training
of bomber pilots is much more expensive than that of U(C)AV pilots and operators of ground
stations. UCAVs are also regarded as a key technology for the future market. The USA has
already sold and still sells their MQ UCAVs to France, Italy and other countries. The USA
spend several billions every year on drones. For example, one of the mentioned M-Q 9 Reaper
systems (with four aircraft) costs about 70 million dollars. Experts estimate that from 2015 on
UCAVs sales will amount to five billion dollars every year (Nikolei 2005). With regard to the
huge techno-ethical problems Europe should engage in preventive arms control to regulate the
development of this market and to prevent an arms race in the near future.
At the same time, the new warfare scenarios build on the features of new military robot
technologies. For example, the UCAVs have a different tactical aim than traditional jet
bombers: Predators or Reapers are supposed to function as a permanent threat to the enemy:
"You've got a lot of ammo circling overhead on call for short-notice strikes," said John Pike,
director of the military think tank, Globalsecurity. "It seems like a good idea."
At the same time, it is much more difficult for counter-combatants from non high-tech
countries to destroy UCAVs: They are only endangered by – relatively expensive – surface-
to-air or air-to-air missiles. At the same time, some propose to equip UCAVs with signatures
from inhabited systems to provoke an attack to discover the enemy’s positions (Boes 2005).
Especially autonomous UCAVs are regarded as an important part of new technoscientific
warfare scenarios. Together with inhabited systems integrated in a complex network of air,
water and ground agents, new techniques of warfare are developed “… toward a vision of a
strategic and tactical battlespace filled with networked manned and unmanned air, ground,
and maritime systems and the technologies needed for navigation and fighting. Unmanned
systems provide autonomous and semi-autonomous capabilities that free warfighters from the
dull, dirty, and dangerous missions that might now be better executed robotically and enable
entirely new design concepts unlimited by the endurance and performance of human crews.
The use of UAVs in Afghanistan and Iraq is the first step in demonstrating the
transformational potential of such an approach.” (Department of Defense 2007, 34) The US
forces dream of a high-tech transformation that makes them invincible. Autonomous robotic
warfare systems equipped with artificial intelligence and learning capability are supposed to
act – at least in the long run –more precisely, more quickly and to process more data than any
system operated by a human soldier.
Old Dreams of Techno-Supremacy & the Production of Asymmetric Warfare
Dreams of perfect robotic and information warfare networks resonate with the older dream of
power supremacy by nuclear weapons – a dream that led to a dangerous and expensive global
arms race. At the same time, the dream of almighty robotic warfare seems to be undermined
in the very moment of its rise. The contemporary experiences of U.S. and NATO forces in the
Iraq and Afghanistan wars show – like the Vietnam War before – that high-tech supremacy
does not lead automatically to supremacy in warfare. While the first Gulf war – the first one
waged under the paradigm of information warfare – was a quite successful high-tech war
because of its novel approach, information and cyberwar scenarios in the Afghanistan and
Iraq wars did not function in the same way and led to asymmetric warfare which means here
counter-insurgency warfare within cities. These asymmetric wars cost more lives (especially
of US combatants) than the traditional ‘Air-Land’ warfare did before. And it seems to be a
war that cannot be won by U.S. and NATO forces. Given these experiences, many Western
military militaries do not rethink their military strategies but concentrate on high-tech
solutions for ‘military operations on urban terrain’ (MOUT). The latter build on the idea of
installing high-tech panopticons in ‘oriental’ cities with their labyrinth streets and rich and
complex (infra)structures. These high-tech scenarios include anticipatory tracking and risk
profiling systems which use digital sensors, persistent surveillance, with “continuous,
anticipatory, ‘armed vision’” (Graham 2006). Parallel to the idea of autonomous combat
robots, the US military follows its dreams of autonomous electronic surveillance, tracking and
targeting systems that do not only identify ‘targets’ but automate the destruction of the targets.
This development demonstrates the endless spiral of the race for high-tech ‘solutions’ of war.
Following this logic, low-tech insurgents will find means against panoptical, automated
warfare (Boes 2005) but it is very likely that these new ways of warfare will endanger
civilians more than traditional warfare ever did. In this sense, new high-tech war is everything
else anything but a ‘clean’, ‘precision-guided’ war saving the lives of one’s own troops as
well as those of civilians.
Starting up a new global arms race with high-tech warfare is one problematic aspect. Another
one is the tremendous humanitarian consequence of urban warfare. Disarmament experts,
political scientists as well as many philosophers point out that it is highly probable that
automated warfare will not lead to effective deterrence and thereby to the avoidance or
shortening of wars. On the contrary, automated warfare will lead to a lowering of the
threshold for warfare (Altmann in this volume, Sparrow 2007)7. It is also unclear whether
robot weapon systems are counted with regard to certain disarmament agreements such as the
Conventional Forces in Europe Treaty (CFE) (Botturi et al. 2008). Altmann (in this volumes)
points out that some robot systems would fall under the Treaty definitions, for others the
Treaty provides mechanisms to include them The question would be whether robot systems
fall under the criteria of preventive arms control such as the Conventional Forces in Europe
Treaty (CFE) which sets limits to the armed forces.
Hacking Drones
Before I will analyze more deeply the political and ethical problems of robotic warfare and
the deployment of UCAVs, I would like to point to an additional problem of robotic warfare
with UCAVs: The high risks of hacking their communication structures. Until now high-
volume data transfers are vulnerable to wiretap and noise. It is highly probable that hostile
forces will engage in disabling the robot systems by hacking their communication
infrastructure. The latter is the weak point in UCAVs (see Altmann 2003, ISIS 2006, Sparrow
2007). Hacked UCAVs would be highly dangerous not only to the soldiers of one’s own
troops but also to anybody and especially civilians if UCAVs fall – for example –into the
hands of counter-insurgents. As the military is also aware of this great danger, it is very likely
that either a robot war between two high-tech nations will escalate into a space war to secure
one’s satellites or autonomous weapon systems (AWS) will be deployed in the near future
because they are less dependent on communication systems. The latter is highly problematic
with regard to the question of international law and questions of responsibility, but also with
regard to the heightening speed of warfare where wrong decisions can no more be cancelled
or changed (see i.e. Sparrow 2009a).
Another problem is that UCAVs can be reconstructed quite easily. The availability, stability
and low cost of hard and software components as well as the modularity of robot systems
make it possible to build surveillance drones with only little know-how – and these drones can
theoretically also be equipped with bombs (ISIS Europe 2006, Miasnikov 2007).
7 See also below
Bi-Directional Use, Distancing & Online-War
Autonomous robot systems can be easily copied and remade. Because of the modularity and
universality of today’s robot systems, relevant parts can be bought from the civilian industry
without any obstacles (Miasnikov 2004, 2007; Boes 2005, 6; ISIS 2006). Therefore robot
weapon systems can be built and used by criminals easily.
At the same time, robots developed for warfare are increasingly invading civil society. For
example, UAVs are already deployed for the observation of the Californian-Mexican border
by U.S. Homeland Security and the border of Switzerland8 by the Swiss police. The German
company Rheinmetall Defence has already installed an own economic sector called Homeland
Security and applies for the commission to ‘secure’ the borders of the European Union. Up to
now there are only few surveys how the bi-directional[dual?] use of military robots might also
violate rights of privacy and data protection.
Robot manufacturers like Foster-Miller, maker of the armed SWORDS robot for ground
combat, are actively promoting robots equipped with tasers for U.S. police forces9. As robots
are seen as cost-savers, it is likely that robots will be increasingly used even for sensitive
police tasks. Pressing social aspects and questions are often made invisible. There are already
studies pointing out that the use of tasers increases the amount of violence in police-civilian
encounters (LIT). It seems highly probable that this will become even more problematic with
tele-operated robots equipped with tasers. Mediating interaction via technology does not
always contribute to the reduction of violence.
The problematic aspects of distancing oneself from the relevant encounters have been
discussed widely with regard to warfare. Weapons of mass destruction often rely on
technological artefacts such as planes, missiles etc. which allow the distancing of the
commanding and responsible officer. This becomes also evident with regard to UCAVs.
Some remote-controlled robot combat systems – for example, the MQ-1 and MQ-9 deployed
in Iraq – are controlled from Nevada, which is about 7000 km away. This might pose a real
challenge for soldiers to execute their responsibility because of the hyperreal character of their
actions. It becomes close to the experience of a computer game to command a robot drone for
destruction which is thousands of kilometres away and to monitor the result solely via video.
The question is whether a reliable experience of the consequences of one’s actions can be
made with regard to these remote-controlled (or even autonomous) weapon systems.
Technology design should be aware of this problem and think about how to avoid these
effects (Sparrow 2009a).
8 The EU agency Frontex is planning to use them in the near future.
Uninhabited Systems and Jus in Bello
Many ethicists – whether they argue from a deontological or from a consequentialist
perspective – have pointed out that responsibility for the killing of human beings is a main
condition for jus in bello: “If the nature of a weapon, or other means of war fighting, is such
that it is typically impossible to identify or hold individuals responsible for the casualities that
it causes then it is contrary to this important requirement of jus in bello.” (Sparrow 2007;
emphasis given). If responsibility is no more a critical issue, this might have severe
consequences for the way wars with autonomous weapon systems will be fought in the near
To avoid the discussion on moral problems and autonomous robots, the military often claims
that autonomous systems will only be deployed under the supervision of human (military)
operators (Marsiske 2007, Sparrow 2007). On the other hand, the U.S. military invests heavily
in the development of fully autonomous systems: „DARPA is expanding the level of
autonomy and robustness of robotic systems. Progress is measured in how well unmanned
systems can handle increasingly complex missions in ever more complex environments […].
Autonomy and robustness are improved by networking manned and unmanned systems in a
more tightly coupled combat system that will improve our knowledge of the battlespace,
enhance our targeting speed and accuracy, increase survivability, and allow greater mission
flexibility.” (Department of Defense 2007, 34)
There is an internal tension between the claim of the responsibility of military operators and
the intense development of AWS: Why should one want to build fully autonomous systems
and only use them as more or less remote-controlled systems? One of the main reasons for
building autonomous systems is to increase the speed on the battlefield. Human operators are
slowing down the velocity of the battle. On the other hand, it is very likely, that from the
moment an enemy will deploy totally autonomous systems, the other side will also use them.
In this case, the battle could get very easily out of control. And then there is also the
‘temptation’ to use fully autonomous UCAVs to avoid the communication infrastructure
which might be threatened by the enemy.
There are other propositions on how to ensure responsibility with autonomous combat
systems: Either responsibility is addressed towards the programmer, the machine or the
commanding officer. As autonomous systems will show unpredictable behaviour, some argue
that the responsibility lies by the programmer and / or manufacturer. But if the manufacturer
gave appropriate information about the risks of autonomous weapons, the manufacturer
cannot be held responsible for a machine failure (Nagenborg et al. 2008). Think for example
of the destruction of the wrong target as an outcome of the autonomous behaviour of the
system. If a system is supposed to act increasingly autonomous and the system does so, the
programmer cannot be held responsible for the negative outcome of the unpredictable
behaviour of an autonomous system. A programmer could be held responsible – beside those
who are deploying the machines – only in a legal sense, if AWS (Autonomous Weapon
Systems) would be banned internationally and he or she further participates in programming
such machines.
To hold an autonomous machine responsible does not seem to be reasonable regarding their
serious cognitive limitations of these systems, think for example of the state of art in object
recognition, the incapability of machines to adapt the (right) rules to complex situations
adequately, etc. (see Tamburrini in this volume, Botturi et al. 2008).
The preferred approach of the military is to attribute the responsibility to the officer – as it is
the case with long range weapons. This seems to be a non-satisfying and possible incorrect
solution of the problem because AWS have – at least theoretically – the ability to choose their
own targets: Then officers would be held responsible for weapons which they do not control.
(Sparrow 2007, 71)
Push-Button Wars on Low-Tech Nations?
As I already mentioned, one of the most pressing concerns about autonomous combat systems
is that they might make going to war quite easy. Until now, in democracies a basic agreement
in the population about going to war has to be achieved – or at least an disagreement has to be
avoided. How will this change if war is conceived as a matter of pushing buttons from a
remote place, without risk to one’s own soldiers? And what chances are taken and people
killed if there is no one responsible for the killing of civilians or surrendering combatants?
Also disobeying inhumane orders will no more happen in robot wars. This is (or was?) a
crucial part of at least a bit more humane way of warfare. We know – for example – that
human soldiers often point and shoot their guns in the air and not at the combatants. But
robots will always do what they are programmed for. As autonomy of weapon systems on the
one hand and responsibility of the soldiers for their own deeds on the other hand is
contradictory in itself, robot wars could endanger the international law of war (Geneva
Conventions etc.). The general introduction of robot weapons will possibly lead to a
“destabilization of the military situation between potential opponents, arms races, and
proliferation, and would endanger the international law of warfare.” (Altmann 2006) And at
the same time, it is quite likely that advanced and capable autonomous killing systems will be
deployed anyhow because one is afraid that they might be used by one’s enemy. Therefore it
is highly necessary to ban autonomous weapon systems. Bans are not new. We have a ban of
biological and of chemical weapons as well as of anti-personal mines in most European
countries. If there could be an agreement that autonomous systems are in contradiction with
the Geneva Conventions, further development, and deployment would have to be stopped.
Since 2002, violations of the international laws of warfare, such as the Geneva Conventions,
can also be prosecuted by the International Criminal Court in The Hague, Netherlands. In
Germany, for example, there is a code of law, which enables the State Attorney? to open a
lawsuit against suspects of war crimes or crimes against humanity in the cause of the
international law of war.
The Rhetoric of Moral Machines
In the face of the serious juridical and ethical problems regarding military robots and a
possible ban of military robots, research on ethics and especially so-called ‘ethical’ software
for military robots is sponsored by the U.S. Army Research Office and the U.S. Office of
Naval Research (Arkin 2007, Arkin 2008, Moshkina/Arkin 2008, Lin et al. in this volume).
Ronald Arkin, a roboticist at the Georgia Institute of Technology, is one of its best known
figures. His approach is to embed ethics resp. software for ethical behaviour in robots’
architectures. In his paper ‘Governing Lethal Behaviour: Embedding Ethics in a
Hybrid/Deliberative/Reactive Robot Architecture’ (2008) he even proposes that future robots
will be more ethical than humans because they can be programmed in such a way that they do
not have emotions or a drive for self-preservation. He suggests that these future robots might
have a better sensory to destine whether a target is legitimate. He argues that robots do not
have the human psychological problem of ‘scenario fulfilment’ which means the tendency to
interpret all input according to fixed expectations and a pre-given frame of thought – a daring
perspective given the fact that robots are not capable of questioning their own programming
while humans have the potential to reflect on the grounding of their decisions.
According to Arkin, robots also compute information from more sources much faster, so that
they have more time for reasoning about lethal decisions. He proposes that robots could even
supervise the decision of the human teams they work with. (Arkin 2007, 6f). Arkin does not
argue that so-called ethical robots would be perfect but that they could perform better than
humans with regard to the massive violations of the laws of war during ‘Operation Iraqi
Freedom’. In the face of advanced weapon technology and the failure of the U.S. forces to
train their troops properly with regard to ethical, intercultural and social issues, automatic
ethics is invented as the only possible solution.
This reductionist and technophilic logic follows the well-known pattern of ‘solving’ socio-
political and ethical problems via technology. But very rarely the underlying epistemological
as well as ontological foundations of such a proposal are discussed. Let us see what features
of his future robots Arkin takes for granted on which he builds the formalization of ethics in
1. Robot systems have at least as much information as soldiers. He does not discuss the
meaning of ‘information’ – whether this term is used in its everyday sense (linked to
understanding and meaning) or in a technological sense (Shannon/Weaver) which
does explicitly exclude these dimensions (Hayles 1999).
2. The proposed advanced systems do not only have the capability to resist the
performance of an unethical act but to explain the reasons of their resistance. If the
resistance is overridden by the commanding officer, the latter is responsible for the
system’s actions. This approach either suggests highly intelligent systems that will not
become reality in the next decades (Tamburrini in this volume) or this mechanism of
resistance works on a very reductionist level.
3. A central assumption of Arkin’s approach to realize the formalization of ethics in
robot systems is one that will not be solved in the near future: The capability of
autonomous weapon systems to distinguish between soldiers, surrendering soldiers
and civilians.
A central claim of Arkin’s which he does not make explicit is that every possible complex
situation can be formalized correctly and computed in real time. This is an old and never
realized claim of Artificial Intelligence. But also the problems of navigation, object
recognition as well as the scaling-up problem (parallel computing of many behaviours in one
system) in real and complex worlds have not been solved and will not be solved in the near
future in a satisfying manner.
In face of these difficulties, even Arkin’s ‘Closed World Assumption’ does not help. The
latter states that a system is never allowed to make lethal decisions in situations which are not
covered by ethical prescriptions. But the unsolvable question is: How can one make sure that
a system is applying rules adequately to a certain situation and that the system decides
correctly that it is allowed to apply its rule to this specific situation?
Arkin also avoids the problem of formal verification. Above other things, to use autonomous
lethal systems – even if they have ‘ethical’ software – one has to make sure that there are no
bugs at all in the software. But formal verification of software for systems as complex as
combat robots is not possible in a reasonable amount of time – if at all. So how can one think
of ‘ethical’ warbots?
Analyzing this approach, one gets the impression that the development of ‘ethical’ software
for robot systems is not driven by the intent to solve humanitarian problems in the first place
but to raise acceptance for a new, emerging and highly problematic technology that ends up in
highly automatized wars with unforeseeable atrocities.
Other roboticists already point out the enormous danger of the proposed autonomous weapon
systems. Given the overestimation of Artifical Intelligence, the British roboticist Noel
Sharkey appeals to the ethical consciousness of computer scientists and engineers to discuss
the unsolved and profound technical problems of military robotic systems in public.
Responsible roboticists are asked to withstand generous offers from military and civilian
funding agencies as well as the perpetuation of the old and well-known salvation stories of
AI. Sharkey states: “Computer professionals and engineers have a duty to ensure that the
funding bodies , policy makers and – if possible – end users know the current limitations of
AI technology, including potential mishaps in the complexity of unpredictable real-world
events. Do not be tempted to express your opinions or future predictions of AI as if the
technology were already in place or just around the corner. The consequences of playing the
funding game are too serious. Ultimately, we must ask if we are ready to leave life-or-death-
decisions to robots too dim to be called stupid.” (Sharkey 2007, 123; my emphasis).
Given the massive ethical and socio-political problems that come with robot weapon systems,
their worldwide development and supply, as well as the increasing deployment of UCAVs,
esp. for (partial illegal) targeted killing in so-called ‘wars on terror’, we need a broad
international and public debate on robotic warfare technologies.
It is quite obvious that military robots such as UCAVs support the escalation of asymmetric
warfare and lead to an increase in the numbers of killed civilians. At the same time,
autonomous weapon systems are contradicting the existing law of warfare. On the one hand,
they are not capable of distinguishing between soldiers, surrendering soldiers and civilians –
also not in the near future; on the other hand it becomes impossible to hold soldiers
responsible for their actions. As I stated above, the dominant attitude to save the lives of
soldiers while endangering those of civilians in ‘wars on terror’ also relies on an unspoken
racism and sexism – valuing the lives of U.S. or NATO soldiers higher than those of civilian
Afghan, Pakistan or Iraqui women, children and elderly people disregarding their structurally
different situation and exclusion of relevant decisions in international security and warfare
We need to dismantle the dream of the revolution in military affairs in terms of a perfect and
clean information warfare based on the Global Information Grid (GIG) and Joint Battlespace
Environment with automatic robot weapons, “- all lightness, speed, information gathering,
information technology, and shared materiel” (Blackmore 2005, 7). This vision suggests the
elusive idea of automatic, bloodless warfare without humans. The computer-game perspective
of UCAV warfare with its distancing effects produced by tele-operated drones in Afghanistan
from Nevada, needs to be challenged. It seems most likely that these developments will end in
wars highly risky for civilians, in a dangerous and potentially endless spiral of high-tech arms
races as well as a destabilization of military balances.
Further critique of robotic warfare does not only need to focus on the contradictions to the law
of warfare and the technical problems such as hacking drones or the possible reconstruction
and misuse of robots by criminals, but also on the pervasion of civil society with robot
applications. The surveillance of borders, but also soccer games or demonstrations with
drones or the future deployment of police robots with tasers are examples of the militarization
of society – treating your own people a priori as a risk factor, building on surveillance and
monitoring instead of socio-political interaction and democratic processes.
At the same time, the limitations of AI resp. robot weapon systems must be repeatedly called
into memory to counteract the common glorification of the capabilities of future robot weapon
systems. Proposals of ethical software for killing systems must be questioned radically and
the imaginary of a clean and more humane wars via robot and other high technology must be
recognized as a vain dream if not cynical propaganda.
In the end, the dream of ethical killing machines is meant to silence the few remaining
Western voices which are critiquing Western high-tech warfare. This dream is meant to
silence those who remind us of the permanently exposed and terrified by (high-tech) war.
While the dream of ‘ethical’ high-tech warfare supports the common western media narratives
of regrettable collateral damages which will soon be avoided by more precisely functioning
machines, we need to acknowledge the on-going and very effective killing of civilian Afghan,
Pakistani, Palestine or Iraqi people by Western high technology.
I am very grateful to Jürgen Altmann, Cheris Kramarae, Angela Meyer, and Michael Nagenborg for
their helpful comments on earlier drafts of this paper.
Air Force Link. Official Website of the UNITED STATES AIR FORCE (2007): MQ-9
Reaper Uninhabited Aerial Vehicle. In: (last
access 31.08.07)
Altmann, Jürgen (2003): Roboter für den Krieg? In: Wissenschaft und Frieden, Nr.3, Vol. 21,
Altmann, Jürgen (2006): Trends in Cognitive Science and Information Technology. In: Annex
to Study: EU research and innovation policy and the future of the Common Foreign Security
Policy. A Report Commissioned by the Science and Technology Foresight Unit of DG
Research, European Commission. October 2006; ed. by Stephen Pullinger. In: www.isis- (last access 28.8.07)
Altmann (2009) in this volume
Arkin, Ron (2007): Governing Lethal Behaviour: Embedding Ethics in a
Hybrid/Deliberative/Reactive Robot Architecture. Mobile Robot Laboratory, College of
Computing, Georgia Institute of Technology, 2007. In:
lab/online-publications/formalizationv35.pdf (last access 1.02.08)
Arkin, Ron (2008): On the Ethical Quandaries of a Practicing Roboticist: A first-hand Look.
In: (last access
Asaro, Peter M. (2008): How Just Could a Robot War Be?, in Current Issues in Computing
And Philosophy, P. Brey, A. Briggle, and K. Waelbers, Eds. Amsterdam, Netherlands: IOS
Publishers, 2008.
Barry, Charles L. / Zimet, Elilu (2001): UCAVs – Technological, Policy, and Operational
Challenges. In: Defense Horizons. October 2001, Nr. 3, 1-8
Blackmore, Tim 2005, War X, University of Toronto Press, Toronto.
Boes, Hans (2005): An der Schwelle zum automatischen Krieg. In:
4/artikel/21/21121/1.html (last access 28.8.07)
Botturi, Luca / Capurro, Rafael / Donnarumma, Francesco / Gasson, Mark / Gill, Satinder /
Geordani, Alessandro / Laschi, Cecilia / Lucivero, Federica / Pingel, Christoph / Salvini,
Pericle / Santoro, Matteo / Tamburrini, Guglielmo / Warwick, Kevin / Weber, Jutta (2008):
Techno-Ethical Case-Studies in Robotics, Bionics, and Related AI Agent Technologies.
Deliverable 5 of the EU-Project ETHICBOTS. Emerging Technoethics of Human Interaction
with Communication, Bionic and Robotic Systems (SAS 6 - 017759), April 2008
(unpublished report)
Brook, Tom Vanden (2007): Faster, deadlier pilotless plane bound for Afghanistan. In: USA
Today (
Brzoska, M. (2006). Trends in Global Military and Civilian Research and Development and
their Changing Interface. Proceedings of the International Seminar on Defence Finance and
Economics, New Delhi, 13-15 November 2006, pp. 289-302.
Burgess, Lisa (2007): Reactivated wing is first combat unit with UAVs. In: Stars and Stripes,
Mideast edition, Thursday, May 3, 2007 (
Canning, John S. (2006): Concept of Operations for Armed Autonomous Systems. The
Difference between ‘Winning the War’ and ‘Winning the Peace’. In: (last access 3.09.07)
Butler, Judith: Commencement Address 2008. At: (last access 12.1.2009
Capurro, Rafael / Christaller, Thomas / Datteri, Edoardo / Giordani, Alessandro / Laschi,
Cecilia / Mock, Michael / Nagenborg, Michael / Operto, Fiorella / Salvini, Pericle /
Tamburrini, Guglielmo / Veruggio, Gianmarco / Weber, Jutta (2008): Ethically Motivated
Recommendations for EU Policies and Regulations concerning Robotics and its Convergence
with Bionic and Software Agent Technologies. Deliverable 6 of the EU-Project ETHICBOTS.
Emerging Technoethics of Human Interaction with Communication, Bionic and Robotic
Systems (SAS 6 - 017759), April 2008 (unpublished report)
Case, David (2008), The U.S. Military's Assassination Problem.
problem.html (last access 1.1.09)
Cerqui, Daniela / Weber, Jutta / Weber, Karsten (eds.) (2006): Ethics in Robotics.
International Review of Information Ethics 2/2006, http://www.i-r-i- (last access February 2008)
Christaller, Thomas et al. (2001). Robotik. Perspektiven für menschliches Handeln in der
zukünftigen Gesellschaft. Berlin et al.: Springer.
Common Military List of the European Union (adopted by the Council on 19 March 2007)
(equipment covered by the European Union Code of Conduct on Arms Exports) (updating and
replacing the Common Military List of the European Union adopted by the Council on 27
February 2006) (last access 29.03.07)
Cordeschi, Roberto / Tamburrini, Giuglielmo (2006): Intelligent Machines and Warfare.
Historical Debates and Epistemologically Motivated Concerns. In Magnani Lorenzo, Dossena
Riccardo (eds), Computing, Philosophy and Cognition. Proceedings of the European
Computing and Philosophy Conference (ECAP 2004), LONDON: College Publications, 1-20
Cordesman, Anthony C. (2008): Air Combat Trends in the Afghan and Iraq Wars. Center for
Strategic & International Studies (CSIS). March 2008. In: (last access 29.12.08)
Department of Defence (2007): Unmanned Systems Roadmap 2007-2032,
ETHICBOTS: Emerging Technoethics of Human Interaction with Communication, Bionic
and Robotic Systems.
Fischer, Hannah (2008): Iraqui Civilian Death Estimates. CRS report for Congress, August
2008. (last access 2.1.2009)
General Atomics (2007): MQ-9 Reaper. Predator-B Hunter-Killer UAV. In: www.defense- (last access 31.08.07)
Graham, Stephen (2006): Interrupting the Algorithmic Gaze? Urban Warfare and US Military
Technology. In:
df (last access 3.3.2008)
Graham, Stephen (2007): RobowarTM Dreams. Global South Urbanisation and the US
Military’s ‘Revolution in Military Affairs’. Crisis States Research Centre, Working Paper 20,
622F-AB7B-F4DA40512848&lng=en (last access 1.1.09)
Hanley, Charles J. (2007): Robot Air Attack Squadron Bound for Iraq. The Guardian, July 15,
2007. in:,,-6781081,00.html
Hayles, N. Katherine (1999): How We Became Posthuman: Virtual Bodies in Cybernetics,
Literature, and Informatics. Chicago
Herold, Marc W. (2008): The Matrix of Death: (Im)Precision of U.S. Bombing and the
(Under)Valuation of an Afghan Life. In:
imprecision-ofus-bombing-and-the-under-valuation-of-an-afghan-life.html (last access
ISIS Europe (2006): EU research and innovation policy and the future of the Common
Foreign Security Policy. A Report Commissioned by the Science and Technology Foresight
Unit of DG Research, European Commission. October 2006, ed. by Stephen Pullinger. In: (last access
Jane’s (2007). Jane’s Unmanned Vehicles and Aerial Targets. Coulsdon: Jane’s.
Khalifa, Omar: Israel’s all-seeing eyes., 19.July, 2008
Patrick Lin, George Bekey, and Keith Abney (2009)in this volume
Loyd, Anthony (2007): Afghans blame Nato, the invited peacekeeper, for civilian deaths. The
Times, 24th May 2007;
Marsiske, Hans-Arthur (2007): An der langen Leine. Roboter im Sicherheitsdienst. In: c’t –
magazin für computertechnik, 9/2007
Marte, Ana / Szabo, Elise (2007): Center for Defense Information: Fact Sheet on the Army’s
Future Combat Systems, August 7, 2007. In: www.cdi/org
Meilinger, Philip S. (2001): Precision Aerospace Power, Discrimination, and the Future of
War. Aerospace Power Journal 15, 3 (2001): 12-20
Mellenthin, Knut: Terror gegen Verbündete. Wie Washington die Souveränität eines
Partnerlandes respektiert: Die USA bombardieren regelmäßig den Nordwesten Pakistans.
Junge Welt, 18.10.2008
Mellenthin, Knut: Yes he can. Junge Welt, 26.01.09
Meyer, Josh (2006): CIA Expands Use of Drones in Terror War. Los Angeles Times, January
29, 2006
Miasnikov, Eugene (2007): Terrorists Develop Uninhabited Aerial Vehicles – On „Mirsad 1“
Flight Over Israel, Center for Arms Control, Energy and Environmental Studies at MIPT, (last access 31.08.07)
Miasnikov, Eugene (2004): Threat of Terrorist Uninhabited Aerial Vehicles: Technical
Aspects, Center for Arms Control, Energy and Environmental Studies at MIPT, June 2004,
http://www. (last access 31.08.07)
Moshkina, Lilia / Arkin, Ronald: Lethality and Autonomous Systems: Survey Design and
Results. In:
publications/MoshkinaArkinTechReport2008.pdf (last access 30.5.08)
Nagenborg, Michael / Capurro, Rafael / Weber, Jutta / Pingel, Christoph: Ethical Regulations
on Robotics in Europe. In: AI & Society. August 2007
Nikolei, Hans-Hermann (2005): Milliardenmarkt Drohnen. In: (last
access 5.1.08)
Rall, Ted (2006): U.S. Drone Planes Have a Nearly Perfect Record of Failure. In: Common
Dreams Newscenter. In: (last access
Rötzer, Florian (2007a): Einsatzregeln für Kampfroboter. In: (last access 1/2008)
Rötzer, Florian (2007b): Schwärme von Kampfdrohnen sollen Aufständische bekämpfen. In: (last access 1/2008)
Rötzer, Florian (2008): Der Luftkrieg im Irak und in Afghanistan. In: (last access 31.1.2008)
Sharkey, Noel: Robot wars are a reality. Armies want to give the power of life and death to
machines without reason or conscience. The Guardian, August 18, 2007,,,2151357,00.html
Noel Sharkey: Automated Killers and the Computing Profession. In: Computer. November
2007, Vol. 40, No. 11, pp. 124, 122-123
Sparrow, Robert (2007): Killer Robots. In: Journal of Applied Philosophy, Vol. 24, No. 1,
2007, 62-77
Sparrow, Rob (2009a): Building a Better WarBot. Ethical Issues in the Design of Unmanned
Systems for Military Applications. In: Science and Engineering Ethics (forthcoming)
Sparrow, Rob (2009b): Predators or Plowshares? Time to consider arms control of robotic
weapons. In: IEEE Technology and Society (forthcoming)
Tamburrini, Giugelmo (2009) in this volume
Tickner, J. Ann (2004): Feminist Responses to International Security Studies. In: Peace
Review 16:1, March 2004, 43-48
US Congress (2000): National Defence Authorization, Fiscal Year 2001. Public Law 106-398
von Schomberg, René (2007): From the Ethics of Technology towards an Ethics of
Knowledge Policy & Knowledge Assessment. European Commission, Community Research,
Working Document, EU 22429
Warren, Pete (2007): Robot Wars. In:
wars.htm (last access 28.8.07)
Weber, Jutta (2006): From Science and Technology to Feminist Technoscience. In: Kathy
Davis / Mary Evans / Judith Lorber (eds.): Handbook of Gender and Women´s Studies.
London: Sage 2006, 397-414; http://www.uni-
Weber, Jutta (2008): Human-Robot Interaction. In: Sigrid Kelsey / Kirk St. Amant (eds.)
Handbook of Research on Computer-Mediated Communication. Hershey, PA: Idea Group
Some Targeted Killing Missions by the US-Military in Afghanistan, Irak and Pakistan
According to the Center for Strategic & International Studies (Washington DC, USA), there
was a steady rise of close air support (CAS) resp. precision strikes by jet bombers as well as
U(C)AVs: In the Operation Iraqi Freedom the number of air strikes rose from 285 in 2004, to
404 in 2005. There were 229 in 2006 and 1,119 strikes by the Combined Forces in 2007. In
the Operation Enduring Freedom (Afghanistan[? Including Pakistan?]) the numbers rose from
86 in 2004 via 176 in 2005 to 1,770 in 2006 and 2,926 in 2007. As the number of sorties for
surveillance was relatively stable one can assume that more and more air strikes were carried
out by UCAVs – which probably contributed to the rising number of killed civilians in the
Iraq and Afghanistan wars.
USA Today, 10.11.2008 writes: “The use of drones, which supply 95% of the full-motion
video images commanders use to watch insurgent activity, has skyrocketed in recent years. As
recently as 2005, drones flew 100,000 hours, most of it in support of troops in combat in Iraq
and Afghanistan. In 2008, the number of hours in flight increased to nearly 400,000.”
As information on these target killing missions is still rare in (western) media, I document
some of the U.S. drone attacks I found in the literature during my investigations. This was not
undertaken as a systematically survey.
Sources:,, Guardian, Junge Welt, Los Angeles Times, Spiegel,
Süddeutsche, Telepolis, Times of India, Washington Post, Die Welt
November 2001, senior Al Qaeda military commander Mohammed Atef was killed by a
Predator strike
February.7, 2002: An armed Predator attacked a convoy of sport utility vehicles, killing a
suspected al Qaeda leader
March 4, 2002 a CIA-operated Predator fired a Hellfire missile into a reinforced al Qaeda
machine gun bunker
February 4, 2002 a Predator fired a Hellfire missile at three men, including one nicknamed
‘Tall Man’ who was mistaken by CIA operators for the 6’5” feet (=1.96 m] Ossama bin
Laden, near Zhawar Kili in Afghanistan’s Paktia province. The victims were poor civilians
gathering scrap metal from exploded missiles to sell for food.
May 6, 2002, a Predator UCAV fired a Lockheed missile at a convoy of cars in Kunar
province in an attempt to assassinate Afghan warlord Gulbuddin Hekmatyar. He was not there
but at least 10 civilians were killed.
August 22, 2008: “A ground force led by U.S. Special Forces (7th Special Forces Group
(Airborne) based in Shindand), allegedly came under fire as it approached the village,
carrying out a midnight raid to allegedly apprehend a Taliban commander, Mullah Siddiq.
The U.S. Special Forces called-in close air support and a fierce bombardment of the village
ensued involving both ground and air fire (including from a US Air Force Special Operations
AC-130 aerial gunship, Apache attack helicopters and Predator drones). The U.S. bombs
struck a large gathering of people who had congregated in Azizabad to honor a local leader
who had died months earlier. … The Afghan Independent Human Rights Commission
(AIHRC) said its investigators on-site concluded 91 people had perished: 59 children, 19
women and 13 men. Ahmad Nader Nadery, head of the AIHRC, said 76 of the victims
belonged to one large extended family – that of Timor Shah’s brother who is named Reza.
Reza was killed in the assault. Nadery confirmed reports from villagers that a memorial
ceremony was being held for a deputy militia commander allied with the Afghan police
named Timor Shah, who had died in a personal dispute several months ago.”
May 13, 2005, an al Qaeda explosives expert from Yemen was killed by a CIA-operated MQ-
1 Predator aircraft firing a Hellfire missile
December 3, 2005, an Al Qaeda chief and four others were killed in their sleep through a US
Predator UAV
January 13, 2006 several US Predators conducted an airstrike on Damadola village in
Pakistan where al Qaeda’s second-in-command Ayman Zawahiri was reportedly located.
Firing 10 missiles, 18 to 22 civilians were killed, including five women and five children.
According to Pakistan authorities the second leader was not present, but three other leading
figures were killed.
October 30, 2006, Bajaur airstrike, again the attempt to hunt down al Qaeda’s second-in-
command Ayman Zawahiri with predators and hellfire missiles. The strike hit a religious
school. 80-86 civilians were killed. The leader was not present.
August 20, 2008: Missiles, presumably fired from an UCAV, destroyed the house of the tribal
elder in Sari Nur in South Waziristan, near the border of Afghanistan. Six people were killed
and three injured.
August 30, 2008: Missile attack on a house in Korsai/South Waziristan by a drone or bomber.
Following local reports, four humans were killed and two injured. Two of the killed people
were said to be Canadians with Arabic heritage.
September 4, 2008: Missile attack via a Predator UCAV on a house in the village of Achar
Khel in Northern Waziristan. Six people were killed, said to be members of Al-Qaida
September 5, 2008: Missile attack by a drone on two houses in the village Garwek in Northern
Waziristan. According to local reports, seven men, three children and two women were killed.
September 8, 2008: Two predators fired seven missiles on a small village near Miranschah in
Northern Waziristan. 25 humans, mostly women and children, were killed. US officials said
that they were family members of Jalaluddin Haqqani, an important commander of the
Taliban in Afghanistan.
September 12, 2008: Another drone attack near Miranschah. According to local reports, 14
people were killed and 12 injured in the small village of Tolkhel. The target was said to be an
ex-school building inhabited by Pakistani Taliban combatants and their family members.
September 17, 2008: Air attack by drones with four missiles on the village Baghar Schina in
Southern Waziristan. According to local reports, seven people were killed and at least three
more were injured. The US government claimed that the target was a Taliban ordnance
factory. Only a few hours ago, Michael Mullen, US Chairman of the Joint Chiefs of Staff,
assured in Islamabad that the USA respects the sovereignty of Pakistan.
September 30, 2008: Drone attack with a missile on a house near Mir Ali in Northern
Waziristan. According to local reports, six people were killed. The reason for the attack might
have been that some tribal warriors shot live bullets on the drones which were circling in the
area for days.
October 1, 2008 A suspected US drone killed at least six people in a missile strike in North
Waziristan, officials said. Two missiles were fired at a house in the Khushali Torikhel area
near Mir Ali town at about midnight, according to local media reports. Pakistani intelligence
officials said the missiles struck the home of a local Taliban commander.
Oktober 3, 2008: Attack by a drone on the house of Mohammed Khel in Northern Waziristan.
24 people died, including the severely injured who died later in the hospital
Oktober 9, 2008: Attack by two drones on a house in the village of Ghundai in Northern
Waziristan. According to local reports, nine people died, five of them were civilians. The
attacked house was owned by a man who had some relation to the pro-governmental tribal
Oktober 11, 2008: Drone attack on a house near Miranschah in Northern Waziristan. Five
people died, two were injured. According to official reports, the house was owned by a
member of the Taliban.
Oktober 16, 2008: A drone fired two missiles on a house in the village of Sam in South
Waziristan, an area known to be a stronghold of Baitullah Mehsud, the head of Pakistan's pro-
Taliban movement. At least four people died.
Oktober 27, 2008: Peschawar/Islamabad A drone attack on two houses in Schakai / Southern
Waziristan in Pakistan killed about 20 people.
December 29, 2008: Suspected U.S. drones fired at least two missiles into Pakistan’s
Waziristan region on the Afghan border on Monday, killing seven people, intelligence
officials and residents said.
September – December 2008: “U.S. forces have carried out nearly 30 air strikes in Pakistan
this year, according to a Reuters count, more than half since the beginning of September. The
attacks have killed more than 220 people, including foreign militants, according to a tally of
reports from Pakistani intelligence agents, district government officials and residents.”
January 23, 2009: Under the new U.S. President Barack Obama the bombing is continued:
There was an air attack against a building in Mir Ali in Northern Waziristan, another hit a
house in Wana, in South Waziristan. 21 dead people were counted. According to Pakistani
media, the house in Wana was owned by a government-friendly tribal elder, who was killed
together with many family members such as his three sons and his five-year old grandchild.
December 23, 2002: An Iraqui MiG-25 shot down a Predator performing reconnaissance over
the no fly zone in Iraq, after the predator fired a missile at it. This was the first time in history
an aircraft and an uninhabited drone had engaged in combat. Predators had been armed with
AIM-92-Stinger air-to-air-missiles, and were being used to ‘bait’ Iraqui fighter planes, then
run. In this incident, the Predator didn’t run, but instead fired one of the Stingers. The
Stinger’s heat-seeker became ‘distracted’ by the MiG’s missile and so missed the MiG. The
MiG’s missile did not miss.
July 2005 - June 2006: the 15th Reconnaissance Squadron fired 59 Hellfire missiles, surveyed
18.490 targets, escorted four convoys, and flew 2,073 sorties for more than 22.833 flying
hours. The number of dead civilians caused by these attacks is not known.
U.S. troops are going to train more soldiers for the operation of drones in Iraq and
Afghanistan. A high-ranking officer of the U.S. Air Force stated that until 2011, 50 instead of
30 uninhabited aerial vehicles are supposed to monitor Iraq and Afghanistan. At the moment,
U.S. forces are testing a new program to train 1100 pilots and operators who will tele-
operate so-called predators – uninhabited combat aerial vehicles equipped with missiles and
bombs – from their air base in Nevada. (Associated Press / the German newspaper ‚Junge
Welt’, 24.10.08;
... Adversaries can hack into computers, implant viruses and worms, shut down systems, or order fake commands and actions [7]. High-volume data transfers are vulnerable to interference of noises, wiretap, and interception of credentials [8]. It is therefore probable that hostile forces will launch attacks to disable autonomous systems by hacking their communication infrastructure [8]. ...
... High-volume data transfers are vulnerable to interference of noises, wiretap, and interception of credentials [8]. It is therefore probable that hostile forces will launch attacks to disable autonomous systems by hacking their communication infrastructure [8]. Hacked autonomous systems could be dangerous not only to the military, but also to civilians. ...
... Furthermore, autonomous systems, such as unmanned air vehicles, can be reconstructed quite easily. The availability, stability, and low cost of the components as well as the modularity of the systems make it feasible to rebuild surveillance drones [8]. These drones can also be equipped with destructive weapons that may harm civilians. ...
Conference Paper
With the increasing excitement about the emergence of autonomous systems, there is also a level of caution associated with the possible vulnerabilities of such systems. As systems become more independent, it may become more difficult for humans to trust them, particularly in settings where they may be vulnerable to malicious behavior by others. This paper presents a game theoretic model for autonomous systems subject to hacking. We propose a hacking game that models the interaction between an autonomous system and a hacker. Simulations were conducted to examine the outcomes for the victims (rational autonomous agents) when attacked by humans with different motives. The implications of the results for the decision-making process of the players are discussed. This paper also explores how game theory can be used by policy makers to choose a policy when autonomous systems have been attacked.
... The first revolution was due to gunpowder and the second one is due to the development of nuclear weapons [13]. There are supporting and opposing arguments on the use of lethal autonomous systems [6][7][8][9][10][11][12][13][14][15][16][17][18][19][20][21][22][23]. All these arguments point toward an increasing and urgent need for a regulation on the use of lethal autonomous weapon systems. ...
... The use of military drones in war zones is increasing remarkably [5,14,17]. There are many supportive arguments for the use of military drones. ...
... Eventually, however, he felt that he had lost control of and destroyed his machines. Tustin (1972) describes the behaviour of David, an autistic child, in the course of his psychoanalytic treatment. To protect his psyche, David built a cardboard model of an armour's headpiece and hand. ...
... persecutory objects or autistic objects. The latter are typically experienced as a total ''me'' fending off a threatening ''not-me'' (Tustin, 1972), so as to protect one from subjective fragmentation anxieties. ...
Full-text available
Psychological attitudes towards service and personal robots are selectively examined from the vantage point of psychoanalysis. Significant case studies include the uncanny valley effect, brain-actuated robots evoking magic mental powers, parental attitudes towards robotic children, idealizations of robotic soldiers, persecutory fantasies involving robotic components and systems. Freudian theories of narcissism, animism, infantile complexes, ego ideal, and ideal ego are brought to bear on the interpretation of these various items. The horizons of Human-robot Interaction are found to afford new and fertile grounds for psychoanalytic theorizing beyond strictly therapeutic contexts.
... Daten werden mehr oder weniger mit Fakten identifiziert, und so behauptet man, dass die Software aufgrund ihrer ›objektiven‹ Algorithmen komplexe Fakten besser repräsentieren und daher die automatisierten Entscheidungen vornehmen könne (Kitchin und Dodge 2011;Kropp 2020). Vor allem in jenen Bereichen, in denen die Implikationen dieser Annahmen in ihrer Radikalität besonders sichtbar werden -z.B. wenn Künstliche Intelligenz (KI) in autonomen Waffensystemen die Auswahl und Tötung von Menschen übernehmen soll -wurden schon früh kritische Fragen laut (Weber 2009;Schuppli 2014). ...
Full-text available
Wie verändern sich gesellschaftliche Praktiken und die Chancen demokratischer Technikgestaltung, wenn neben Bürger*innen und Öffentlichkeit auch Roboter, Algorithmen, Simulationen oder selbstlernende Systeme einbezogen und als Beteiligte ernstgenommen werden? Die Beiträger*innen des Bandes untersuchen die Neukonfiguration von Verantwortung und Kontrolle, Wissen, Beteiligungsansprüchen und Kooperationsmöglichkeiten im Umgang mit intelligenten Systemen wie smart grids, Servicerobotern, Routenplanern, Finanzmarktalgorithmen und anderen soziodigitalen Arrangements. Aufgezeigt wird, wie die digitalen »Neulinge« dazu beitragen, die Gestaltungsmöglichkeiten für Demokratie, Inklusion und Nachhaltigkeit zu verändern und Macht- und Kraftverhältnisse zu verschieben.
... Daten werden mehr oder weniger mit Fakten identifiziert, und so behauptet man, dass die Software aufgrund ihrer ›objektiven‹ Algorithmen komplexe Fakten besser repräsentieren und daher die automatisierten Entscheidungen vornehmen könne (Kitchin und Dodge 2011;Kropp 2020). Vor allem in jenen Bereichen, in denen die Implikationen dieser Annahmen in ihrer Radikalität besonders sichtbar werden -z.B. wenn Künstliche Intelligenz (KI) in autonomen Waffensystemen die Auswahl und Tötung von Menschen übernehmen soll -wurden schon früh kritische Fragen laut (Weber 2009;Schuppli 2014). ...
Full-text available
Wie verändern sich gesellschaftliche Praktiken und die Chancen demokratischer Technikgestaltung, wenn neben Bürger*innen und Öffentlichkeit auch Roboter, Algorithmen, Simulationen oder selbstlernende Systeme einbezogen und als Beteiligte ernstgenommen werden? Die Beiträger*innen des Bandes untersuchen die Neukonfiguration von Verantwortung und Kontrolle, Wissen, Beteiligungsansprüchen und Kooperationsmöglichkeiten im Umgang mit intelligenten Systemen wie smart grids, Servicerobotern, Routenplanern, Finanzmarktalgorithmen und anderen soziodigitalen Arrangements. Aufgezeigt wird, wie die digitalen »Neulinge« dazu beitragen, die Gestaltungsmöglichkeiten für Demokratie, Inklusion und Nachhaltigkeit zu verändern und Macht- und Kraftverhältnisse zu verschieben.
... But in the eye of the critics, it is precisely these strategic and tactical advantages that constitute the problem: the drone makes tracking down ever more targets possible literally anywhere and anytime, without exposure to risk (Sauer and Schoernig 2012;Zenko/ Kreps 2014). The promise of precision deploys its own logic and story of success (Krasmann 2016;Weber 2009). The presumed precision of drones also is made into an argument to justify aerial bombings in "cities under siege" such as Damascus, Baghdad or Gaza City even as 21st century warfare has turned out to be predominantly asymmetric and therefore urban (Boyle 2013;Graham 2011;Weizman 2006;. ...
... Assim como a Internet (Ruthfield, 1995), os drones foram inicialmente desenvolvidos para fins militares durante o período da Primeira Guerra Mundial (1914 -1918), por serem capazes de atacar alvos sem colocar tripulações em risco (Taylor, 1977). Recentemente os drones ganharam atenção devido ao grande número de mortes de civis provocadas por ataques dos Estados Unidos utilizando UAV's no Afeganistão e no Iraque (Weber, 2009). É crescente o número de projetos científicos que estão utilizando drones em pesquisas sobre agricultura (Costa et al., 2012), ambientes marinhos e oceânicos (Reed et al., 2007) e florestas tropicais (Koh & Wich, 2012). ...
Full-text available
This paper identifies and evaluates the Points of Culture in the city of Santos, Sao Paulo State, aiming to detect possibilities for future developments of experimental laboratories as nuclei for co-management. The survey, which used as source the Network Catalog of Culture Points of Sao Paulo State to identify points (BRAZIL 2012), was carried out between June and July 2012. It was identified two Points of Culture: the “Project Parcel”, located in the continental area, and the “Station of Citizenship and Culture”, in the insular area of Santos. Both have potential to develop experimental activities related to the theme and could subsidize activities of co-management.
Full-text available
This chapter outlines the main developments of roboethics 9 years after a worldwide debate on the subject – that is, the applied ethics about ethical, legal, and societal aspects of robotics – opened up. Today, roboethics not only counts several thousands of voices on the Web, but is the issue of important literature relating to almost all robotics applications, and of hundreds of rich projects, workshops, and conferences. This increasing interest and sometimes even fierce debate expresses the perception and need of scientists, manufacturers, and users of professional guidelines and ethical indications about robotics in society. Some of the issues presented in the chapter are well known to engineers, and less known or unknown to scholars of humanities, and vice versa. However, because the subject is transversal to many disciplines, complex, articulated, and often misrepresented, some of the fundamental concepts relating to ethics in science and technology are recalled and clarified. A detailed taxonomy of sensitive areas is presented. It is based on a study of several years and referred to by scientists and scholars, the result of which is the Euron Roboethics Roadmap. This taxonomy identifies the most evident/urgent/sensitive ethical problems in the main applicative fields of robotics, leaving more in-depth research to further studies.
Full-text available
Already prior to the decline in absolute numbers, military r&d had declined in relative importance due to major increases in civilian r&d. Even in the US, civilian spending is now substantially larger than military spending. In global terms, civilian r&d is about 10 times as large as military r&d. Most of the civilian r&d is privately funded. The largest private r&d spenders, large companies, have higher r&d budgets than gov- ernments can muster for military purposes, with the exception of the United States. While a good part of civilian r&d is in sectors of lesser relevance to the military, much occurs in technology lines important for the production of military goods. In general, the differences between civilian and military technology lines have shrunk.
Full-text available
This working document of the services of the European Commission of which Dr. Rene von Schomberg is the author, reflects upon the difficulties of developing ethical frameworks for (new) technologies in the context of an ethics of responsibility. The principle shortcomings of contemporary ethical theory with regard to the challenges of scientific and technological development are discussed. A case is made for the need of an ethics of collective co-responsibility. Such an ethics should focus on the ethics of knowledge assessment and knowledge policy in the framework of deliberative procedures, rather than on the ethics of technologies as such. Furthermore, the document tries to identify the deliberative procedures and processes in the science-society interface in which ethical issues concerning new technologies are discussed. The documents ends with a preliminary and descriptive overview of how currently the issue of nanotechnology is addressed in deliberative frameworks at the European Level.
Engineers at the Massachusetts Institute of Technology (MIT) are studying how robots and humans can work closer together on assembly lines. The Interactive Robotics Group in the Computer Science and Artificial Intelligence Laboratory (CSAIL) is focusing on dynamic sequencing scheduling, spatial awareness and other areas. Julie Shah, an assistant professor of aeronautics and astronautics at MIT who heads up the Interactive Robotics Group, and her team are studying robot spatial awareness by using overhead cameras and a motion-capture system attached to robots. They have also harnessed artificial intelligence technology to enable the machines to learn from experience, so the robots will be more responsive to human behavior. The MIT engineers are also working with a major automaker and attempting to install mobile robotic assistants on assembly lines.
The Bush administration and Congress are in concert on the goal of developing a fleet of unmanned aircraft that can reduce both defense costs and aircrew loses in combat by taking on at least the most dangerous combat missions. Unmanned combat aerial vehicles (UCAVs) will be neither inexpensive enough to be readily expendable nor--at least in early development--capable of performing every combat mission alongside or in lieu of manned sorties. Yet the tremendous potential of such systems in widely recognized, and allies as well as potential of such systems is widely recognized, and allies as well as potential adversaries are moving quickly to mount their own research and development programs. The United States is committed to fielding UCAV capabilities by 2010, principally for the missions of suppression of enemy air defense and deep strike, which are among the highest risk tasks for the Air Force and naval aviation.
With the development of the General Atomics MQ-1 Predator, robotic weapons came of age. The operations of this unmanned aerial vehicle (UAV) in Iraq, Afghanistan, Pakistan, and northern Africa in the last few years have given us a glimpse of the future of high-tech war. It is a future in which thousands of miles separate those firing weapons from those whom they kill, in which joystick jockeys have replaced pilots and soldiers, and in which the psychological barriers to killing are greatly reduced by the distance between weapon operators and their targets. Perhaps more importantly, it is a future in which wars are more likely, in which decisions about when weapons are fired and who they are fired at are increasingly in the hands of machines, and in which the public has little knowledge of or control over what is being done in its name. Finally, it is a future that is likely to come about not because it represents a better, less destructive, way of fighting war but because the dynamics driving the development of unmanned weapon systems (UMS) are likely to dictate that they be used more and more often. Now that we have had a glimpse of this future, it is time to begin thinking about whether - and how - we might avoid it by adopting an arms control regime designed to limit the development and deployment of robotic weapons.