The opinions expressed in this essay are the author's and do not necessarily reflect those of the United States Military Academy, the Department of the Army, or the Department of Defense.

 

 

Unintentional Terrorism?: Some Objections to Rodin’s “Terrorism without Intention”[1]

 

Stephen N. Woodside

United States Military Academy

 

 

David Rodin defines ‘terrorism’ as the “deliberate, negligent, or reckless use of force against noncombatants,”[2] though he concedes that his account “is at odds with the current common usage of the term.”[3]  While most believe that terrorism necessarily involves the intentional killing of noncombatants (NCs), Rodin disagrees.  His definition allows that some acts involving the unintentional killing of NCs, those that are reckless or negligent, are acts of terror.  And while he offers some support for his unorthodox definition, Rodin develops a more substantive argument that “many of the noncombatant casualties caused in the course of military operations (including those of Western nations)” are unjustified.[4]  Furthermore, those responsible for these casualties, whether or not we ought to call them ‘terrorists’, are morally blameworthy to the same degree as those who intentionally target the innocent.[5]  Rodin’s argument is based on his core premise that many of the actions involving the foreseeable killing of NCs (call them “standard cases”[6]) are reckless or negligent despite their satisfying the DDE.  He begins his support for this claim by arguing that the DDE is false, as it does not provide sufficient moral justification for an action.  I will argue Rodin’s objections to the DDE are unconvincing and that his argument is unsound primarily because his core premise is false.  Furthermore, I will argue that those responsible for killing NCs in the standard cases are not as morally blameworthy as those who commit traditional acts of terror (those that intentionally target the innocent), even if these actions are unjustified.     

 

I. The Doctrine of Double Effect (DDE).

Rodin’s argument that many of the standard cases are unjustified begins with his objecting to the doctrine of double effect (DDE).  He claims that an action’s satisfying the DDE is not sufficient moral justification for it since the DDE allows that some reckless or negligent actions are morally permitted.  He summarizes the DDE below:

One may never intentionally bring about an evil, either as an end in itself, or as a means to some greater good.  Nonetheless, one may use neutral or good means to achieve a good end which one foresees will have evil consequences provided that (i) the evil consequences are not disproportionate to the intended good, (ii) the action is necessary in the sense that there is no less costly way of achieving the good.[7] 

Rodin then offers two counterexamples to the DDE:  

Take the case of a motorist who drives across a crowded school yard to deliver a sick person to a hospital.  The motorist certainly has no direct intention to harm the children—he aims at their death neither as a means nor as an end.  His conduct may be necessary in the context, and if we imagine that the risk of death to his passenger and to the children are roughly balanced, then it will also be proportionate.  Yet if he strikes and kills a child he will be held liable, in law and morality, for manslaughter because of the recklessness of his actions.  Similarly, a medical researcher who prematurely tests an unsafe new vaccine on humans out of a desire to speed the development of the product and thereby save lives is guilty of negligent action and may be liable for manslaughter.  This is true even though his action may fulfill all the requirements of the double-effect principle.[8]

So according to Rodin, while the actions of the motorist and medical researcher both satisfy the DDE, they are reckless or negligent, and therefore, unjustified.  The DDE is too permissive.  Rodin believes, therefore, that we need a better account of what makes these actions, and others like them in combat, reckless or negligent, and hence, unjustified.[9]     

But I find these counterexamples unconvincing.  Let us first consider the motorist’s actions.  If we construct this case in such a way as to make it clear that it satisfies the DDE, then it is no longer clear to me that it is unjustified.  And conversely, if we construct it in such a way as to make it clear that it is unjustified, it then becomes unclear whether or not it satisfies the DDE.  To clarify, let us assume that the good of the motorist’s saving the sick person outweighs the risk of harm to the children.  His action is proportionate.  But is it really necessary in the relevant sense?  Is there no less costly way of achieving his good?  Are there not any other equally effective routes for him to travel?  Can he not swerve or slow his speed to avoid the children, or sound his horn to alert them while still achieving his good?  If the answer is “yes” to any of these, then I agree that his action is unjustified.  But in this case, it would fail the necessity condition of the DDE.  But if there really is no less costly way for the motorist to deliver the sick person, then I am not convinced that the action is unjustified.  The intuition that his action is unjustified seems forceful only when we imagine the motorist as a marauding daredevil who disregards any alternative less costly courses of action.[10]  What about the medical researcher?  While it seems more plausible that his action is proportionate, necessary, and unjustified, it does not follow from this that it satisfies the DDE.  The medical researcher’s action fails to satisfy the DDE because the harm that he brings about is a means to speeding the development of the vaccine and saving future lives.  In my opinion then, neither of these cases is a convincing counterexample to the DDE.  So where does all this leave us with the DDE?  I suggest that the DDE can account for the reckless and unjust nature of the motorist’s and medical researcher’s actions in spite of Rodin’s criticisms.  Nevertheless, Rodin believes that there is an additional moral requirement “which goes beyond the necessity and proportionality requirements, namely: is it justifiable to inflict such a risk upon this particular person?”[11]

 

II. Rodin’s Argument.

  Continuing with his explanation of the shortcomings of the DDE, Rodin writes, “The fact that the risks are necessary (from the perspective of the beneficiaries of the risky activity) and proportionate (from the impersonal perspective of the world at large) is not sufficient to defeat the personal right not to be endangered or used in this way.”[12]  So an action that satisfies the DDE, according to Rodin, might still be reckless or negligent, as it might fail to adhere to the moral requirement “not to undertake unreasonable risks,”[13] foreseeable risks inflicted upon those who maintain their right not to be harmed (hereafter “relevant right”).  So if an action results in the foreseeable harming of those who maintain their relevant right, then it is unreasonable.  And if it is unreasonable, then it is reckless.  Finally, if it is reckless, then it is unjustified. 

            Now all of this would imply that any act of war causing foreseeable harm to NCs is unjustified if there were no way for an NC to surrender his relevant right.  But Rodin thinks that there are circumstances under which NCs do this.  To explain, he offers three domestic examples in which those facing foreseeable harm have alienated their relevant right:

…it may be justified for a surgeon to perform an operation in which there is an extremely high risk of death to the patient, if the alternative surgery is an even higher risk of death or of great suffering.  Moreover, police cars and ambulances are permitted to exceed the speed limit and pass red lights in emergencies (though of course strict requirements of care remain, e.g., the requirement to display warning lights, the requirement to slow down when entering intersections, and the requirement to generally minimize the risk to others).  Finally, road engineers are permitted to make a reasonable trade-off between cost and safety when designing a road, though they know that this will lead to a certain number of otherwise avoidable fatalities.[14]

Rodin argues that “in all the above examples a crucial part of what makes the risk justifiable is that the party assuming the risk is also the beneficiary of the risk-producing activity,” and “one may rightly say that the community which enjoys the benefits of improved apprehension of criminals and an affordable road system also bears the risks incurred.”[15]  And although this is an important part of requirement for the surrender of one’s relevant right, Rodin suggests that more is required:

 [The alienation of the relevant right] links to the idea of autonomous agency and free acceptance of risk.  A patient decides whether to assume the risk profile of a given operation and every community makes its own collective decision about how to make the trade-offs involved in building roads and apprehending criminals.  The risks are justifiable in these cases, at least in part, because they have been autonomously assumed either individually or collectively by those who bear them.[16]   

But while Rodin thinks that some NCs do surrender their relevant right in similar ways, he believes that no such surrender takes place in the standard cases.  Before we proceed any further, let us consider the following standard case (call it “A1”):

A1: An American pilot drops a smart bomb on a known terrorist training camp that is set in the middle of a highly populated residential area.  The pilot drops this bomb to eliminate the terrorist threat to both the local populace and his own ground troops, as the terrorists frequently target both of these groups.  Based on his experience with these types of munitions, the pilot foresees that his action will cause 50-100 NC casualties, but he does not intend for this to happen, nor is his harming these civilians a means to eliminating the terrorist camp.  Furthermore, the good of his destroying the terrorist camp far outweighs the harming of these civilians.  Finally, his dropping the bomb is necessary in the sense that there is no less costly way of achieving his intended good.  His action satisfies the DDE.

Do the NCs harmed in A1 surrender their relevant right?  Rodin argues that they do not:

In standard cases of military conflict [like A1] there is no sense in which the party who bears the risk of harm benefits from the risky activity.  Neither have they autonomously chosen, either individually or collectively, to bear the risks of the bombardment.  It is extremely doubtful, therefore, whether the risks imposed on enemy NCs [in A1] can be justified in an analogous way to the risks imposed in domestic emergency situations.[17]

So according to Rodin, an NC alienates his relevant right only if (i) he stands to benefit (either individually or collectively) from the risk, and (ii) he freely accepts the risk.  But the NCs in A1 fail to satisfy both of these conditions.  So A1 imposes an unreasonable risk on the NCs.  Therefore, it is reckless.  It is unjustified even though it satisfies the DDE.  We can now summarize Rodin’s argument in the following way:[18]

1.      If an action causes the foreseeable harming of NCs who have not surrendered their relevant right, then it imposes an unreasonable risk on them.

2.      If an action imposes an unreasonable risk on NCs, then it is reckless. 

3.      If an action is reckless, then it is unjustified.

4.      An NC surrenders his relevant right only if both (i) he stands to benefit from the risk, and (ii) he freely accepts it. 

5.      It is not the case that the NCs in A1 both (i) stand to benefit from the risk and (ii) freely accept it.   

6.      Therefore, A1 is unjustified. 

Although there is one objection to premise five worth considering, I believe that it ultimately fails and this premise remains plausible.  I do believe, however, that there is a more convincing objection to premise four, but I do not see it as a serious setback to Rodin’s argument.  As I see it, the more serious problem with his argument lies with premise one.  I will discuss these objections in the following section.   

 

III. Some Objections.

            One might object to premise five by maintaining that the NCs in A1 do satisfy Rodin’s two necessary conditions for the alienation of their relevant right: they do stand to benefit from the risk and they freely accept it.  These NCs stand to benefit from the risky activity since, after all, they are subjected to these terrorist attacks as well.  And if they do not benefit individually (e.g. they are killed in the attack), then their community will certainly benefit.  Finally, we might also say that these NCs freely accept the risk of harm by allowing the terrorists to maintain their residence in the community.  Or perhaps they freely accept the risk in virtue of their deciding to maintain residence in a known combat zone.  If all of this is correct, then A1 satisfies both of Rodin’s necessary conditions.  Premise five would be false.

But while it may be plausible to argue that these NCs benefit from the risky activity, I am suspicious of the claim that they freely accept the risk.  A lack of armed resistance to the terrorists does not entail free acceptance of the risk of military attacks.  This would suggest that the NCs are morally obligated to resist the terrorists.  And while it might be permissible for them to resists the terrorists, it seems implausible to suggest that are morally obligated to do so.  It is more likely that their resistance would be supererogatory.  Furthermore, to suggest that NCs have freely accepted the risk of military attack simply in virtue of deciding to maintain residence in a combat zone seems inconsistent with our intuitions on analogous cases.  For example, I know that there are at least three gangs in the vicinity of my home.  Suppose that I am shot during a drive-by shooting.  Now, it might be appropriate to say that it would have been prudent of me to move out of the area, or to limit my outdoor exposure.  It might even be correct to say that by not moving and going about my daily routine I accepted the risk of being shot.  But it seems wrong to say that I freely accepted this risk.[19]  I think it is more likely that I freely accepted the risk of getting shot only if I somehow supported or at least condoned the gangs’ activities.  And while I may have done nothing to eliminate the gangs’ activities, I certainly did not support or condone them.  Moreover, surely when we consider certain groups like infants and civic leaders, the objection fails.  There is no sense in which infants, who have no choice but to remain, and civic leaders, who have professional (and perhaps legal and moral) obligations to remain in the area, have freely accepted the risk of military attack.  If I am correct, then the objection to premise five fails. 

            Although I am inclined to agree with Rodin that an NC’s freely choosing to accept risk is necessary for his losing his relevant right, I am not sure that it is necessary that he stands to benefit from the risky activity.  I suppose the rationale might go something like the following.  If we accept that an NC must freely accept the risk to lose his relevant right, then we need only accept the further claim that an NC freely accepts a risk only if he stands to benefit from it.  It would then follow that an NC alienates his relevant right only if he stands to benefit from the risk.  But while it is unlikely that an NC would freely accept a risk unless he stood to benefit from it, it is certainly possible.  In other words, his freely accepting a risk is probably a good indication that he stands to benefit from it, but it does not logically follow from this.  Therefore, premise four is false.[20] 

            The most serious problem with Rodin’s argument is with premise one, his core premise.  I agree that many standard cases like A1 involve the foreseeable harming of NCs who maintain their relevant right, and that these actions are reckless and unjustified if they pose an unreasonable risk to NCs.  But I believe that Rodin is mistaken in suggesting that the foreseeable harming of NCs who maintain their relevant right is ipso facto unreasonable.  I suspect that this claim is based on the idea that foreseeable harm that inhibits one’s relevant right entails the violation of that right.  But while it might be the case that violating one’s relevant right is ipso facto unreasonable, I do not believe that the foreseeable inhibition of one’s right entails the violation of it.  Let us return to A1.  The pilot foresees that he will inhibit the relevant right of the NCs, but it does not follow from this that he has violated their right.  Perhaps he has violated their right only if he inhibits it unreasonably.  But more is required for the unreasonable inhibition of a right than simply inhibiting it.  Perhaps this right would be inhibited unreasonably if he intended to do so, or if his action would result in far more NCs being killed than terrorists, or if there was some way to eliminate the terrorist threat with less NC casualties.  In these cases, we might say that an NC’s relevant right has been violated.  It has been violated because of the unreasonable risk of harm.  But the unreasonableness of this harm is adequately explained by the DDE. 

            One possible reply from Rodin is that I have misinterpreted his premise one.  Perhaps he is not claiming that the foreseeable harming of an NC who maintains his relevant right is, by definition, unreasonable.  Rather, it becomes unreasonable, a violation of his relevant right, if the proper standard of care is not undertaken by those who inhibit the relevant right.  He does suggest this in the following text:

If one is persuaded by these interlinked lines of argument then one will feel that the standard of care required of those engaged in military operations is high—higher than is currently reflected in the laws of war.  One will be inclined to view many of the noncombatant casualties caused in the course of military operations (including those of Western nations) to be culpably reckless or negligent.  One will feel this despite the considerations adduced by the principle of double effect.[21]

So what makes some of the standard cases unreasonable and reckless is that the proper standard of care has not been undertaken.  This suggests that in standard cases there is a level of care possible (even if it may be very high) such that the foreseeable harm to NCs who maintain their relevant right would be reasonable.[22]  But while I am more likely to agree with this reasoning, I do not believe that it is Rodin’s.  He seems to think that there is no appropriate level of care in these cases.  In other words, the only appropriate level of care would be to not harm the NCs.  He suggests this below:

This analysis, of course, raises a difficult question about how it can ever be justifiable to impose risk upon those who have not through their actions made themselves specifically morally vulnerable to it…my sense is that the answer has to do with two kinds of consideration.  The first is the kind of free collective consent to beneficial but risky activities assumed by communities [as discussed previously].  The second has to do with the much rarer set of cases in which there are such overwhelming consequentialist considerations that we are inclined to believe that it is justifiable to impose risks upon those who have done nothing to deserve their imposition…[23]   

So according to Rodin, while it might be justifiable to impose risks upon NCs who have alienated their relevant right in virtue of their freely accepting the risk, or upon those in “ticking time bomb” non-standard cases, it is never justifiable to impose foreseeable risks on NCs in the standard cases.  I believe that this supports my interpretation of premise one.  And if premise one is true, then one’s knowingly inhibiting another’s right entails the violation of that right.  This seems implausible. 

But let us suppose now that Rodin is right about all of this.  Suppose that A1 is unjustified for the reasons that Rodin suggests.  What would we then say about his further claim that the pilot in A1 (and the agents in other standard cases) is “morally culpable to the same degree and for the same reasons that typical acts of terrorism [those that intentionally harm NCs] are culpable?”[24]  Let us consider a typical act of terrorism (call it “A2”):

A2: An Al Qaida terrorist has set up a remotely controlled IED on the side of the road.  He allows a couple of US and Iraqi military vehicles to pass before detonating the bomb as a school bus full of children passes.  He intentionally kills the children.  He does this because he knows that many medical personnel will arrive at the scene shortly thereafter.  He intentionally kills all of these NCs because he knows that this will lead to further destabilization in the area, which will likely hamper Allied efforts.     

Rodin contends that a typical act of terror like A2 “has the significance it does in our moral thinking because the targets it attacks are morally inappropriate.”[25]  We find terrorism morally reprehensible primarily because it attacks NCs who have not alienated their relevant right.  He explains that it is “this feature of the core instances of terrorism which merit and explain the moral reaction which most of us have toward them.”[26]  So while A2 involves the intentional harming of NCs, this does not help account for or influence our moral reaction to it.  The only morally relevant feature is that it involves the foreseeable harming of NCs.  It unreasonably harms them and is therefore reckless.  If all this is true, then an act of unintentional “terrorism” (A1), which shares this same feature of recklessness, merits the same moral reaction.  A1 is equally blameworthy to A2.

But while I agree that the reckless nature of A2 accounts partly for our moral reaction to it, it does not explain all of it.  The terrorist merits additional blame for intentionally killing the NCs.  Even if his intentionally killing them is superfluous to his action’s being unjustified, surely it is relevant to his degree of moral culpability.  Certainly, we think less of the terrorist than we do of the pilot, even if both of their actions are reckless.  This is not to say that the pilot is deserving of no blame, that his lack of intention fully excuses him.  If his action is reckless, then he may very well be morally blameworthy to a high degree.  But not to a degree equal to that of the terrorist.  In short, while it may be plausible to suggest that intention is irrelevant to the moral status of an action, it is implausible to claim that it is irrelevant to assessments of moral blame.  Rodin seems committed to both of these claims.    

 

IV. Conclusion.

I have argued that Rodin’s argument that standard cases like A1 are unjustified is unsound.  For one, it seems unlikely that an NC alienates his relevant right only if he stands to benefit from the risk, although it is plausible that his free acceptance of the risk is necessary.  More importantly, although many NCs harmed in the standard cases do maintain their relevant right, this does not entail that an unreasonable risk has been imposed upon them.  It does not entail that these actions are reckless and unjustified.  And although I have not explicitly argued that a standard case’s satisfying the DDE is sufficient justification for it, I have suggested as much.  Finally, even if we accept that actions like A1 are unjustified, this does not commit us to the further claim that they are morally blameworthy to the same degree as typical acts of terrorism like A2.  This latter claim implies that intention is morally irrelevant.  This seems very unlikely. 

 

NOTES


[1] In Ethics 114 (July 2004): 752-771.

[2] Ibid., p. 755.

[3] Ibid., p. 767.

[4] Ibid.

[5] Ibid., p. 769.  Rodin himself concedes, “But in reality, very little turns on the purely terminological question of the usage of the word ‘terrorism’.  The important issue is whether one accepts the substantive argument put forward in this article: that the unintentional killing of some NCs in the course of military operations is morally culpable to the same degree and for the same reasons that typical acts of terrorism are culpable.”  However, one might wonder if Rodin is more concerned with his definition of ‘terrorism’ than he lets on.  After all, he seems eager to point out the irony in the United States’ fighting a war against terrorism using “terrorist” means.  Nevertheless, I will focus my objections on Rodin’s substantive argument.      

[6] The “standard cases” that Rodin has in mind are the attacks “against those targets within or adjacent to civilian populations which [are] almost certain to generate noncombatant casualties” (Ibid., p. 762).  In particular, he refers to the tactical air strikes conducted by Western powers and their allies in Kosovo, Afghanistan, and Iraq.  He argues that while many of these actions satisfy the DDE, they are nevertheless reckless, unjustified, and no better than traditional acts of terrorism.

[7] Ibid., p. 762.  So a pilot’s targeting a military facility while foreseeing that it will likely harm some innocent bystanders is justified just in case (i) the harming of the innocent is neither the end of his action nor a means to his end, (ii) the good of his action outweighs the risk of harm to the innocent, and (iii) there is no way to reduce the risk of harm to the innocent while achieving his good.

[8] Ibid., p. 764.

[9] I will return to this point shortly.

[10] I have similar concerns about the supposed proportionality of this action.  Although Rodin wants to construct the case in such a way as to satisfy proportionality, I am unconvinced that it does.  Part of the motivation for the intuition that the motorist’s actions are unjustified might also come from our considering that the motorist is risking the lives of several children to save one sick person.  This hardly sounds proportional.  

[11] Rodin, p. 764.

[12] Ibid., p. 765.

[13] Ibid.

[14] Ibid., p. 766.

[15] Ibid.

[16] Ibid., pp. 766-7.

[17] Ibid., p. 767.

[18] This is my interpretation of Rodin’s argument.  See Ibid., pp. 762-770.

[19] Perhaps I would have freely accepted the risk if the responsible gang leader had given prior announcement for the shooting.  If this is true, then perhaps the same could be said in the standard military cases.  I think it’s plausible to suggest that had the military warned all civilians and given them ample time to evacuate, then their staying in the area would constitute free acceptance of the risk.  But I am still not sure about this. 

[20] It might be the case that I misinterpreted Rodin’s premise four.  Perhaps all he means to say is that an NC’s freely accepting a risk is a good indication that he stands to benefit from it.  If this is true, then his premise four is more likely the following: An NC surrenders his relevant right only if he freely accepts the risk.  Premise five could then be modified to maintain the validity of the argument: It is not the case that the NCs in A1 freely accept the risk.  But while these modifications would make Rodin’s argument for plausible, I believe that there is a more substantial problem with premise one.  I will discuss this problem next.

[21] Rodin, p. 767.

[22] If this is true, then Rodin’s premise one would be something like the following: If an action causes the foreseeable harming of NCs who have not alienated their relevant right without the proper standard of care, then it poses an unreasonable risk.  For a similar view, see Michael Walzer, Just and Unjust Wars (New York: Basic, 1977), pp. 151-9.  In addressing the shortcomings of the traditional DDE, Walzer writes, “But we have to worry…about all those unintended but foreseeable deaths, for their number can be large; and subject only to the proportionality rule—a weak constraint—double effect provides a blanket justification” (p. 153).  He argues that what is also required is that “the foreseeable evil be reduced as far as possible” (p. 155).  But it is important to note that Walzer’s and Rodin’s views are different in at least two respects.  First, Walzer believes that intention is morally relevant to the justification of the action while Rodin does not.  Second, Walzer’s addition allows that soldiers can be justified in the foreseen killing of NCs.  Rodin’s view does not.   

[23] Rodin, p. 769.

[24] Ibid.

[25] Ibid., p. 757.   

[26] Ibid., p. 753.