Wired for War – Singer

Wired for War by P.W. Singer (2009, Penguin books) is one of the most important books in security circles over the past 10 years. It is a milestone account of the state of technology up to 2009 and considered so many things that were at the cutting edge of innovation, that many of them still haven’t happened yet.


Without further ado, here are our thoughts:

Wired for War is an excellent book, and a comprehensive introduction to the impact of technology, and specifically robotics, in modern warfare. Some questions that arise from the text:

  1. iRobot’s mission statement (27) is somewhat disturbing, given their military links. Do we run the risk here of creating too much of a psychological distance between a product and its function?*
  1. It is interesting to note that books such as Starship Troopers and Ender’s Game appear in so many professional reading programs and military training courses (see 151 and 156). What are the potential ramifications of reading such books from a military perspective?
  1. What role does fiction and the arts have in the widespread acceptance of technology and military strategy? As with Q1., do we run the risk of ‘dumbing down’ the ethical, moral and social implications of these technologies? Too much emphasis on the rule of cool and ‘crash, bang, whallop’ and not enough intellectual engagement?
  1. The author suggests that robots can potentially reduce the instances of war crimes (393–408). But, with machine learning, will this remain the case? What about robots used on the other side? Will robots place equal value on the lives of friends and enemies alike?

Mike Ryder, Lancaster University

*Interesting related article: Dennis Hayes, ‘The Cloistered Work-Place: Military Electronics Workers Obey and Ignore’ in Cyborg Worlds: The Military Information Society, ed. by Les Levidow and Kevin Robins (London: Free Association Books, 1989).


Wired for War is probably one of, if not the, most important book for scholars of military technologies. Because this book was so forward thinking, and considered things really on the extremes of technological capability when it was written, many of the things that were prophesied as coming soon still haven’t appeared.

I wanted to draw attention to a passage in the book (272-276) which considers how artificial intelligence could be used to predict the incidents of terrorist attacks and other crimes. If the preparatory activities for attacks/crimes can be subject to data pattern analysis, there is a question here as to what to do with this information?

Arresting somebody for conspiracy to commit a crime of attack seems preferable, but the evidence is often difficult to bring together when the intended crime has not yet taken place. Deterring would-be criminals from carrying out the crime is another option. In a similar way to placing police cars outside banks when tipped-off of a possible robbery, an increased police presence at a site of planned crime/attack can have a deterring effect. Yet, we now commonly see determined and highly motivated terrorists carry out their actions despite knowing that they will suffer either arrest or a bullet from an armed response officer. It is certainly a difficult issue.

Recently, UKIP (a far-right UK political party) suggested the internment of terrorist suspects without trial. Yet, we know from experience with the IRA and al-Qaeda that such treatment can be used by those groups as a massive recruiting tool. It seems that the other option down this route of taking action before an actual attack happens is killing the potential attacker as happens with US/Israeli/UK/Russian targeted killings. If we think about the potential future implication of AI systems performing statistical analysis to essentially state that an individual is about to commit a terrorist atrocity, this could then result in a strike from an autonomous drone. We are well into the territory of a worrisome future with this. But, it is possible. This certainly raises questions about how much ‘meaningful human control’ society wants in its counterterrorism.

Joshua Hughes, Lancaster University


As always, see the contact tab if you want to join the network.

5 thoughts on “Wired for War – Singer

  1. I’m interested about this idea of predicting potential terrorists. How far will this prediction extend? I mean, is it ok to preemptively kill a ‘terrorist’ who may commit an attack (for example) in the next *month*? What about the next *year*, or then, maybe the next *decade*? Heck, why not make things easier and take them out while they are still young? Where does one draw the line?


  2. Hi Mike,

    This issue hits upon a really contagious issuing International Law. In terms of taking action in self-defence before something happens, say an attack on the state by either another state or terrorists (whether a state can claim self-defence against non-state actors is a whole other discussion), the issue of imminence has change in recent years from what was know as the Caroline standard where the requirement to act in self-defence in anticipation of an attack was ‘instant, overwhelming, leaving no choice of means, and no moment of deliberation.’ This envisions scenarios whereby an attack is about to happen and you act to prevent that attack and prevent loss of life.

    Contrast this with the Iraq War 2003, where it ended up being (wrongly) justified as self-defence because Saddam could launch WMDs in 45 minutes, meaning there was a constant possible threat, but no concrete threat. This pre-emptive action has been seen as completely unlawful, as it stretches the idea of an imminent attack too far.

    In terms of present day drone strikes, the original interpretation of imminence has been stretched far, but not Iraq-far. The claim by countries carrying out drone strikes is that they know an attack is going to happen imminently, but they don’t know exactly when. So, the need is ‘instant, overwhelming, leaving no choice of means, and no moment of deliberation.’ Add to this that many terrorists go in and out of hiding, and you have a situation whereby attacking the terrorists in self-defence when an attack is imminent, but not immediately incoming, is the only real chance of preventing the attack.

    You can see the UK’s legal position on this here: https://www.gov.uk/government/speeches/attorney-generals-speech-at-the-international-institute-for-strategic-studies


  3. You hit on a key problem relating to the terrorist here, in that you say they go ‘in and out of hiding’. Unlike a traditional soldier who goes home and becomes a ‘citizen’ and is not fighting in a ‘warzone’ as such, a terrorist never directly stops being a terrorist, which I think makes the problem of distinguishing threat and its immanence all the harder. Even if we can define what a ‘terrorist’ is, can we ever really define at what point they become a threat?


  4. Hi Mike, you’ve also managed to hit upon a key point as well. In the law of armed conflict, a combatant is a member of the armed forces which belong to a State, which terrorists or militants are not. So they are actually civilians who ‘directly participate in hostilities’. This means that they are targetable for as long as they are directly participating. The Red Cross would suggest that their direct participation only lasts during preparation, execution, and return from attacks. But, state parties to conflicts take the view that a member of an enemy group should be targetable at any time, on the same basis as their owns soldiers are. This only relates to attacks during an armed conflict, not to attacks away from an armed conflict.

    Outside of armed conflicts, It seems the threshold of threat which a state will accept before they act appears to still be the ‘Armed Attack’ standard from the Nicaragua case, meaning on the same scale as an attack by conventional armed forces. So, we can assume that terrorists killed by drone strikes are planing something big. Here’s an extract from the UK Government Legal Department regarding the drone strike against Reyaad Khan (https://www.parliament.uk/documents/joint-committees/human-rights/Letter_from_Govt_Legal_Dept_Leigh_Day_231015.pdf p.30):
    ‘The intended scale and effects of the threatened attacks against the United Kingdom by Reyaad Khan and those of ISlL with whom he was conspiring were judged to reach the level of a threatened armed attack capable of justifying a response under Article 51 of the UN Charter. There was clear evidence of Reyaad Khan’s planning and directing a series of attacks against the UK and the UK’s allies, including a number which were foiled. The evidence showed that the threat was genuine, demonstrating Mr Khan’s intent, and capability of delivering the attack. The threat of attack was current and could have become a reality at any moment and without warning. The airstrike that killed him was the only feasible means of effectively disrupting the attacks planned and directed by him. There was no realistic prospect that he would travel outside Syria so that other means of disruption could be attempted. The significantly disruptive impact of the strike outweighed any potential effects increasing the threat to the UK. The Government accordingly decided to act in self-defence of the UK against an imminent’armed attack, as it was entitled to do.’

    It seems, we can at least deduce that for a terrorist to be singled out and killed by a state, the level of threat which they pose much be pretty extreme.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s