War photographer John Cantlie in the Pech valley, Afghanistan June 2012. Wikimedia Commons/Unknown. Some rights reserved.Hostage-taking has recently been a major issue in the war against terrorist insurgent groups in Africa and the Middle East. The position held by the UK Government is that one should never bargain with hostage-takers. The government’s view is that any other position is foolish in that it only encourages more hostage-taking. However, whether you believe this policy to be desirable or not, there is something strange and misplaced about the intransigence with which it is held.

First, let’s assume that the hostage-takers are bargaining in good faith: that they will hand over the hostage to the government in return for money or perhaps in exchange for prisoners that the home government – or a third party – holds. It is the duty of the government to safeguard its own citizens and it is prepared to invest in order to reduce their likelihood of death, in automobile accidents for example. Therefore it would seem reasonable at least to consider whether some form of exchange should take place with hostage-takers. It would seem sensible if the government were prepared to pay, say, £50k to reduce expected deaths on a dangerous stretch of road by one, or spend the same money on a surgical procedure that would save one life, that they should be prepared to pay the same fee to release a hostage.

However, the government’s viewpoint is that such a policy is weak and counterproductive because it simply fuels more hostage-taking. The argument is a 'rule utilitarian' one and is rigorously developed in game theory. The argument is as follows. Hostage-taking is costly and dangerous for terrorists. Hostages must be guarded, fed, and perhaps moved frequently in order to evade the SAS. If the government adopts a rule that it never pays a ransom, then this will deter any hostage-taking because there is a positive cost but no monetary return to the hostage-taker. If the government can credibly claim that it will always act tough and never pay the ransom, this policy then takes on the magic property of allowing the government to never actually have to act tough, because no one will take hostages.

Free Alan Johnston. Flickr/Martin Deutsch. Some rights reserved.The main game theory problem is: how do you make the rule credible when in any particular case it might well be in the interest of the country’s inhabitants, especially the hostage herself and her friends and family, to pay the ransom? Although it might be thought that the UK government has a duty to defend its citizens from harm, it has even made it illegal for insurance companies to pay such a ransom. This is for people who are working in dangerous circumstances overseas, often in humanitarian roles. The argument is this: one has to think about the potential hostage-takings that will be avoided in the future as a result of not paying this ransom demand. If the government has a rule to never pay a ransom and concedes to break this rule once, it will have to pay the ransom whenever it is asked again. It is, therefore, a case of weighing up the refusal to pay the ransom of one case today, against possible hostage-takings in the future that could be generated by an alternate decision.

What are the weaknesses with this argument? Let us begin with the model (or game) in its pure form. If the ransom is always paid, there are no direct deaths. There will be future hostage-taking but no deaths. Therefore in refusing to pay the ransom in any one case the government is allowing a death in order to reduce monetary payments in the future. Clearly, this raises issues about the value of life and does not suggest that there is an obvious, unquestionable policy. It could (and almost certainly should) be argued that it is sensible to interpret the model (or game) less strictly. Now, it should be noted right from the start that as soon as one does this one is likely to move away from categorical outcomes coming from the model. The results are more likely to be contingent and dependent on conjectures about unknown probabilities and trade-offs.

It could be argued that hostage-taking is risky by nature and involves potential deaths, so reducing future hostage situations will reduce future deaths. If the money that it transferred to the hostage-takers goes to violent terrorists, the denial of this money will also reduce future deaths. However, this sort of argument seems to introduce further problems. Whether the lives of actual hostages being held today should essentially be bartered for the lives of potential and unknown future hostages is a morally tricky argument. Also if the usage of the ransom money becomes a central issue, then this further weakens the notion that there is an open and shut case in favour of not paying ransoms.

However, there is a rather obvious weakness to the position of the UK government: terrorist groups appear to keep taking hostages, so the theory, in its strong form at least, is not working. If the response is that this simply means that we should stiffen our resolve, how many times do we need to resist paying ransoms before we change the policy? A second problem is that it is possible that terrorist groups pick up hostages as part of their other operations. If there were no value to hostages for them, because of the policy of not paying ransoms, perhaps the terrorists would just kill potential hostages. Therefore the policy might not be costless, even if the rule is totally credible and the authorities are never asked to pay a ransom.

Jordanian Hostages Released. Flickr/UNAMID. Some rights reserved.The further weakness is that if countries differ in their attitude to ransom paying, the policy of not paying ransoms would be ineffective unless the hostage-takers are aware of this and are able to differentiate between nationalities. Essentially, it is an ineffective policy if terrorists are not aware that the UK does not pay ransoms when other countries do. On the other hand it is possible that terrorists are sophisticated and are very knowledgeable about different countries’ policies. But is there any evidence that terrorists take more hostages from those countries that pay a ransom? Also, with regards to the UK government's argument, if it thinks that terrorists are well informed, the UK government should encourage other countries to pay ransoms (if it wants to save UK citizens from becoming hostages by being tough itself). If terrorists only take French hostages, for example, the issue then purely rests on what the hostage-takers will do with the money and how that affects our citizens through other (non-hostage-taking) activity, not on the principle of the payment.

Again, perhaps the hostage-taker has more than one option with the hostage: a ransom or a gruesome message. Maybe the latter in itself makes it worth taking the hostage, regardless of whether a ransom is paid. In this case the treatment of these citizens is driven by our role in fighting terrorism (and therefore the force of the message) rather than whether we pay ransoms or not. Again, this does not make the case clear cut in any sense, but involves other considerations.

Finally, if the argument against paying ransoms is so watertight, why is it not adopted in other parts of our legal system? Logically, the country should adopt very harsh punishments for even trivial crimes as they do in Singapore, because these punishments will deter people from committing crimes in the future. David Cameron does not wish to employ the death sentence domestically, even for very serious crimes, but it might be said that a similar sentence is in effect being imposed on innocent UK nationals being held as hostages, all in the name of deterrence.

It is important to stress that our argument is against UK government’s adoption of the TINA (there is no alternative) principle to the non-payment of ransoms. Whenever this principle is invoked, alarm bells should begin to ring. It might be that this is the most effective policy but it is not self-evidently the case. Furthermore, we are not arguing against game theory which might be seen to be implying a logical necessity to such a form of action. Nothing could be further from the truth. Game theory identifies how even a slight change to some of the assumptions can radically affect the equilibrium outcome. Instead, we are arguing for a rational discussion of this issue informed by evidence and facts and also for a more reasonable treatment of the individuals and their friends and families caught up in hostage situations.