Aristotle and Double Effect

There are some interesting similarities between Aristotle’s ‘mixed actions’ in Book III of the Nicomachean Ethics and the actions often thought to be justifiable with the Doctrine of Double Effect . Here I analyse these similarities by comparing Aristotle’s examples of mixed actions with standard cases from the literature on double effect such as, amongst others, strategic bombing, the trolley problem, and craniotomy. I find that, despite some common features such as the dilemmatic structure and the inevitability of a bad effect, Aristotle’s mixed actions do not count as cases justifiable through application of the Doctrine of Double Effect because they fail to meet the crucial necessary condition of the Doctrine according to which the bad effect can only be a merely foreseen side-effect and not an intended means.

...if a tyrant were to order one to do something base, having one's parents and children in his power, and if one did the action they were to be saved, but otherwise would be put to death... Something of the sort happens also with regard to the throwing of goods overboard in a storm; for in the abstract no one throws goods away voluntarily; but on condition of its securing the safety of himself and his crew any sensible man does so.(1110a5-20) Let us look at the similarities between those two examples: both cases are such that the agent is under external pressure to do something that she would normally not do -'do something base' and 'throw goods overboard'.Both are cases where capitulating to the external pressure appears to be prima facie morally preferable to not capitulating: 'do something base or your family will be killed' and 'throw goods overboard or you and your crew will die'.In this respect the latter example is even clearer than the former, contrasting goods with persons; the former is more problematic because we may imagine base things that are not prima facie outweighed by rescuing one's family.
We may indeed for both examples talk of two effects or consequences: doing something base against rescuing my family in the first case; and throwing goods overboard against rescuing myself and my crew in the second case. 3These cases are 2 See also tinyurl.com/7p722gy.This blog post and Pakaluk's piece are the only two cases known to me in which this suggestion is made. 3A peculiarity of the second example is that, supposedly, if I did not throw goods overboard to rescue myself and my crew, the goods would still get lost in the shipwreck.This is like imagining that, in the first example, the tyrant would tell me that if I did not do something base, then he would kill my family and also bring about that very same base thing.This does indeed make a possibly significant difference, both action-theoretically and morally, so that it is just best to imagine a case in which, if I did not throw the goods overboard, myself and the crew would die but the goods would be somehow saved (maybe to avoid shipwreck I must throw the goods similar to Double Effect cases in that in both cases there is a prima facie good effectrescuing my family and rescuing my crew -and a prima facie bad (or evil) effect -doing something base and throwing goods overboard.
Here I first explain what Aristotle means by 'mixed actions' by looking at his discussion of hekousion in NE III and then compare mixed actions with the standard scenarios in the double effect literature.

Aristotle
Aristotle thinks that these sorts of actions are 'mixed' because they are both hekousion and akousion (voluntary and involuntary).Here is the passage on mixed actions in full: But with regard to the things that are done from fear of greater evils or for some noble object (e.g. if a tyrant were to order one to do something base, having one's parents and children in his power, and if one did the action they were to be saved, but otherwise would be put to death), it may be debated whether such actions are involuntary or voluntary.Something of the sort happens also with regard to the throwing of goods overboard in a storm; for in the abstract no one throws goods away voluntarily, but on condition of its securing the safety of himself and his crew any sensible man does so.Such actions, then, are mixed, but are more like voluntary actions; for they are worthy of choice at the time when they are done, and the end of an action is relative to the occasion.Both the terms, then, 'voluntary' and 'involuntary', must be used with reference to the moment of action.Now the man acts voluntarily; for the principle that moves the instrumental parts of the body in such actions is in him, and the things of which the moving principle is in a man himself are in his power to do or not to do.Such actions, therefore, are voluntary, but in the abstract perhaps involuntary; for no one would choose any such act in itself (1110a5-20).
Here we must briefly say something about Aristotle's general discussion of hekousion in NE III.1 and III.2.Aristotle distinguishes between at least five kinds of actions: a) Chosen actions b) Voluntary actions c) Not voluntary actions d) Involuntary actions e) Mixed actions overboard here and now, and if I don't the ship will sink there and then, and here the goods cannot be recovered by divers but there the goods can be recovered.The peculiarity discussed here is pointed out by Pakaluk (2011: 215) as well, who does not though suggest reading the second example as I do here.
Aristotle starts with (d), involuntary actions: "Those things, then, are thought involuntary, which take place by force or owing to ignorance" (1110a1).This is pretty intuitive: the only thing to note is perhaps that, in the contemporary action theory literature, what happens as the result of physical coercion is normally distinguished by behaving in ignorance.Davidson (1963Davidson ( , 1971Davidson ( , 1978)), the trailblazer of contemporary causalism, holds that some movement is an action only if it is intentional under at least one description.On this account, movements that happen under physical coercion (Aristotle's own example is "if he were to be carried somewhere by a wind, or by men who had him in their power" (1110a2)) are intentional under no description; therefore these movements are not actions, following Davidson.
On the other hand, on Davidson's account, cases of ignorance will count as actions, precisely unintentional actions.Oedipus knowingly kills the man at the junction and unknowingly kills his father.Here killing his father is unintentional because Oedipus does not know (nor does he believe) that the man at the junction is his father; but it is an action because it is intentional under another legitimate description, namely killing the man at the junction.So this is an example, on Davidson's account, of unintentional action.And this is different from the examples of physical coercion because those are not even actions.
Aristotle also distinguishes between involuntary actions and actions that are simply not voluntary: "Everything that is done by reason of ignorance is not voluntary; it is only what produces pain and repentance that is involuntary" (1110b18-25).After having offered various examples of involuntary actions, Aristotle puts forward a definition of voluntary action: "Since that which is done under compulsion or by reason of ignorance is involuntary, the voluntary would seem to be that of which the moving principle is in the agent himself, he being aware of the particular circumstances of the action" (1111a21). 44 I did not want to unnecessarily complicate this discussion, but there may be still another sixth kind of action in Aristotle's text, distinct from the five discussed here: "Acting by reason of ignorance seems also to be different from acting in ignorance; for the man who is drunk or in a rage is thought to act as a result not of ignorance but of one of the causes mentioned, yet not knowingly but in ignorance" (emphasis in the original, 1110b-25).Here the distinction seems to be between ignorance is the sense of not knowing, like Oedipus; and ignorance in the sense of a sort of unawareness or unconsciousness, like the drunk.These two are genuinely different, and it is an interesting open question what the contemporary causal account that we have been David Charles (1984) in his book on Aristotle's philosophy of action formalises Aristotle's voluntary action as follows: z is a voluntary action of S's at t1 iff (1) z is a bodily movement of S's at t1; (2) S knows the relevant particulars involved in doing z (what he is doing at t1; to whom; with what: 1110b33); and (3) z is caused efficiently by one of S's desires at t1 (1980: 58).

5
This formulation is interestingly similar to the kind of Humean causal account of action Davidson puts forward in Actions, Reasons, and Causes (1963 -and indeed there Davidson takes himself to be defending an "ancient" (1963: 4) view, referring presumably to Aristotle).In this respect it is then no surprise that Charles thinks that 'voluntary' action in Aristotle is intentional action.But there is an important technical difference: Davidson's account is left as a set of necessary conditions because of deviant counterexamples.Presumably Charles's formulation of Aristotle's account, given as a set of necessary and sufficient conditions, would also be subject to deviant counterexamples (1963: 12 and also Davidson's essay Freedom to Act (1973)) 6 .
Voluntary actions, according to Aristotle, can be attributed to children and other animals too, and Aristotle introduces a further kind that seems to be reserved only to adult human agents, (a) choice: "Both the voluntary and the involuntary having been delimited, we must next discuss choice; for it is thought to be most closely bound up with virtue and to discriminate characters better than actions do.Choice, then, seems to be voluntary, but not the same thing as the voluntary; the latter extends more widely.For both children and the lower animals share in voluntary action, but not in choice, and acts comparing Aristotle to has to say about the case of the drunk, who does not seem to act intentionally under any description, but who will have to be hold responsible for the damage that she might cause (for more on this, see my work in action theory and free will : Di Nucci 2008, Di Nucci 2009, Di Nucci 2010a, Di Nucci 2010b, Di Nucci 2011a, Di Nucci 2011b, Di Nucci 2011c, Di Nucci 2012b, Di Nucci 2013d, Di Nucci 2013e, Di Nucci 2013f, Di Nucci 2014b, and Di Nucci (forthcoming b)). 5 Charles revises the third condition twice: "(3)' z is caused by S's desire to do a z-type action either for its own sake or as the means to achieving a further goal which he desires for its own sake… (3)'' z is an action which is caused either by S's desire to do a z-type action (for itself or derivatively) or by his desire to do a y-type action (for itself or derivatively), when S knows that in doing y he is also doing z" (1984: 61).Here I shall ignore the possible circularity in 3''. 6For a recent introduction to deviant counterexamples see Stout (2010).
done on the spur of the moment we describe as voluntary, but not as chosen" (1111b5-10). 7aving now looked at Aristotle's understanding of action, we have a better idea of why Aristotle claims that 'mixed actions' are voluntary: the agent is neither physically coerced to act in the case of 'mixed actions' nor does the agent act in ignorance.Less clear, it would seem, is why Aristotle says that 'mixed actions' are also in some sense involuntary.But take notice of the following two aspects: firstly, mixed actions are also, in an important way, forced upon the agent.Even though the agent is not physically coerced as when she may be carried by the wind, the agent is forced by the circumstances to act the way she does.8Take the tyrant's example: the point is that it is only because of the tyrant's threat that the agent may consider doing something base.And that it is only because of the threat of the storm that the agent will throw goods overboard.Secondly, Aristotle says in distinguishing involuntary actions from actions that are merely not voluntary, that involuntary action requires "pain and repentance"; mixed actions are certainly also a cause of pain; and the sort of difficult choices involved may provoke sometimes repentance and almost always lingering doubts.
Aristotle's suggestion, then, is that mixed actions are voluntary actions that have important features in common with involuntary actions: external circumstances force the agent in a dilemmatic situation out of which the agent cannot come without some bad consequences.Indeed, even when the balance of reasons may be clear as in the case of throwing goods overboard, the agent's only way out of the difficult situation involves some necessary bad consequences.Those are the very features of the sort of cases that are often tackled by the Doctrine of Double Effect: dilemma situations where there are no good effects without bad effects.
There is another similarity between Aristotle's mixed actions and standard cases from the literature on double effect: we have seen that Aristotle says that mixed actions are hekousion but also, in some sense, akousion.And we have seen that Charles's seminal treatment of Aristotle's action theory takes hekousion to be intentional action.At the same time we find, in discussions on double effect (see for example in the next section Bratman 1987), that the actions that the Doctrine is supposed to justify (Strategic Bomber's killing of the children, say) are intentional but not intended. 9 Let us now look at some classic double effect cases and compare them to Aristotle's mixed actions.

Double Effect
Aristotle's two cases are strikingly similar to the classic examples in the Double Effect debate.I will mention here five kinds of examples which I take to be the most common ones 10 : 1) Terror Bombing A good illustration of this case is provided by Bratman (1987): Both Terror Bomber and Strategic Bomber have the goal of promoting the war effort against Enemy.Each intends to pursue this goal by weakening Enemy, and each intends to do that by dropping bombs.Terror Bomber's plan is to bomb the school in Enemy's territory, thereby killing children of Enemy and terrorizing Enemy's population.Strategic Bomber's plan is different.He plans to bomb Enemy's munitions plant, thereby undermining Enemy's war effort.Strategic Bomber also knows, however, that next to the munitions plant is a school, and that when he bombs the plant he will also destroy the school, killing the children inside.Strategic Bomber has not ignored this fact.Indeed, he has worried a lot about it.Still, he has concluded that this cost, though significant, is outweighed by the contribution that would be made to the war effort by the destruction of 9 Here issues become quite quickly complicated and technical, as some authors on double effect take the actions that the Doctrine can justify to be intentional but unintended (like Bratman) while others take those actions to be full-blown unintentional.Importantly, here, there is no agreement in current philosophy of action on the notions of 'intentional' and 'intended'.Some endorse the so-called Simple View, according to which every intentional action is intended on exactly the same description under which it is intentional (see, for example, Adams 1986 andMcCann 1991).Most deny the Simple View, either because intentional actions must not necessarily be intended under their intentional description (see Bratman's Single Phenomenon View, 1984& 1987 or because intentional actions must not necessarily be preceded by any intention (see for example Hursthouse 1991, Pollard 2003& 2006, (also my own already cited work).Here I can't go into these sorts of detailed discussions, I just wanted to emphasize the common 'ambiguous' character of Aristotle's mixed actions and the actions justified by the Doctrine. 10These five cases are for example cited by McIntyre in her introduction to Double Effect for the Stanford Encyclopedia of Philosophy (she adds self-sacrifice to the list).Nothing really hangs on whether all of these cases are proper illustrations of double effect (self-defence may not be, for example): I mention them as illustrative of the Doctrine and of the debate and as cases that are similar to Aristotle's two examples.the munitions plant.Now, Terror Bomber intends all of the features of his action just noted: he intends to drop the bombs, kill the children, terrorize the population, and thereby weaken Enemy.In contrast, it seems that Strategic Bomber only intends to drop the bombs, destroy the munitions plant, and weaken Enemy.Although he knows that by bombing the plant he will be killing the children, he does not, it seems, intend to kill them.Whereas killing the children is, for Terror Bomber, an intended means to his end of victory, it is, for Strategic Bomber, only something he knows he will do by bombing the munitions plant.Though Strategic Bomber has taken the deaths of the children quite seriously into account in his deliberation, these deaths are for him only an expected side effect; they are not -in contrast with Terror Bomber's position -intended as a means...In saying this I do not deny that Strategic Bomber kills the children intentionally.(1987:  139-140) 11   2) Trolley Problem The Trolley Problem was first introduced by Foot (196712 ) even though its most influential formulation is Thomson's (1976Thomson's ( & 1985) ) variation on Foot's original case.
In that case you have been strolling by the trolley track, and you can see the situation at a glance: The driver saw the five on the track ahead, he stamped on the brakes, the brakes failed, so he fainted.What to do?Well, here is the switch, which you can throw, thereby turning the trolley yourself.Of course you will kill one if you do.But I should think you may turn it all the same (1985: 1397).
Consider a case -which I shall call Fat Man -in which you are stand-ing on a footbridge over the trolley track.You can see a trolley hurtling down the track, out of control.You turn around to see where the trolley is headed, and there are five workmen on the track where it exits from under the footbridge.What to do? Being an expert on trolleys, you know of one certain way to stop an out-of-control trolley: Drop a really heavy weight in its path.But where to find one?It just so happens that standing next to you on the footbridge is a fat man, a really fat man.He is leaning over the railing, watching the trolley; all you have to do is to give him a little shove, and over the railing he will go, onto the track in the path of the trolley (1985: 1409).

3) Craniotomy
Here we can go back to Foot's influential discussion (1967): It is said for instance that the operation of hysterectomy involves the death of the foetus as the foreseen but not strictly or directly intended consequence of the surgeon's act, while other operations kill the child and count as the direct intention of taking an innocent life, a distinction that has evoked particularly bitter reactions on the part of non-Catholics.If you are permitted to bring about the death of the child, what does it matter how it is done?The doctrine of the double effect is also used to show why in another case, where a woman in labor will die unless a craniotomy operation is performed; the intervention is not to be condoned.There, it is said, we may not operate but must let the mother die.We foresee her death but do not directly intend it, whereas to crush the skull of the child would count as direct intention of its death.

4) Death-inducing pain-alleviation
This is another classic of the literature on double effect, even though there is overwhelming evidence that the sort of dosages of opioid drugs which are effective against pain do not actually hasten death (Sykes and Thorns 2003).
A doctor who intends to hasten the death of a terminally ill patient by injecting a large dose of morphine would act impermissibly because he intends to bring about the patient's death.However, a doctor who intended to relieve the patient's pain with that same dose and merely foresaw the hastening of the patient's death would act permissibly (McIntyre, Stanford Encyclopedia of Philosophy).

5) Self-defence
In this case we may just go back directly to Aquinas: "Nothing hinders one act from having two effects, only one of which is intended, while the other is beside the intention.(…) Accordingly, the act of self-defence may have two effects: one, the saving of one's life; the other, the slaying of the aggressor."(Summa II-II, Qu. 64, Art.7) Let us now compare these classic examples from the Double Effect literature to Aristotle's two examples.Notice the similarity between, say, Strategic Bomber and Aristotle's captain.Both do things that might be thought to be absolutely forbidden (at least relatively to their role).Both do them unwillingly.Both seem to think that their wider goal is worth the sacrifice: in Strategic Bomber's case the wider goal of weakening enemy is supposedly worth the sacrifice of killing the children; and in Aristotle's Captain's case the wider goal of guaranteeing the safety of the crew is supposedly worth the sacrifice of the cargo.And in both cases we can imagine alternative characters who might do the thing in question willingly: Terror Bomber on the one hand, and a Captain who might want to compromise the profitability of that particular route by throwing goods overboard.Also, importantly, Aristotle endorses the captain's throwing goods overboard by saying that "any sensible man does so" just as the Doctrine of Double While it is not in all cases clear that the good effect is prima facie to be preferred to the bad effect, it is in all cases clear that there is a prima facie good effect and a prima facie bad effect. 13Here it is again worth emphasising the disanalogy between Aristotle's two examples: while Aristotle endorses the throwing of goods overboard, it is not clear that Aristotle would also endorse doing something base to save your family.Indeed, Aristotle states that "...some acts, perhaps, we cannot be forced to do, but ought rather to face death after the most fearful sufferings" (NE III.1 49).But that is not a problem, as the example involving doing something base to placate a tyrant is left open without specifying what the base deed should be (so that only the good effect is morally clear and assessable -saving one's family -while the bad effect is left as a variable.Pakaluk for example suggests that Aristotle may rather tend not to endorse doing something base for the sake of saving one's family, providing some Socratic applications: "Often it is supposed that Aristotle is presuming that the agent will act to save his family, and yet the passage gives no grounds for holding this.Actually, there is no need to suppose that Aristotle even meant to suggest a 'right' answer to the dilemma, which serves its purpose if both alternatives appear unpalatable.But suppose the dilemma was formulated in the Academy as an intensification of dilemmas that Socrates faced-its being wondered how Socrates would have acted in these more trying circumstances-then it is natural to think that Aristotle, if, as seems right, he took Socrates as a model of virtuous action, would at least have taken seriously that the agent should refuse to do the shameful thing.The two cases actually faced by Socrates were: the Thirty Tyrants commanded him to arrest an innocent man, and he refused; his friends tried to persuade him to escape from prison for the sake of caring for his family, and he refused.These dilemmas are intensified if one imagines that it is not Socrates but those he cares for who will suffer through his refusal to heed the Tyrants' command, or that the cost of his refusing to do something shameful, such as deliberately break the law in escaping from prison, was not merely that his family suffer from his absence, but also that they be tortured and killed" (Pakaluk 2011: 215-16).
Pakaluk suggests in relation to the case in which one's family is hostage to a tyrant "if in the second case one refused to do the disgraceful thing, then one might wish to defend this by an appeal to DE [Double Effect], claiming that the death of one's family, although foreseen, was not intended, an that the agent cannot be held responsible for the bad actions of others, when there are merely incidental to what he was aiming to do, that is, act always in an upright way: in avoiding the disgraceful action, one both gains an honourable good (the kalon) and incurs a loss (of philoi); it makes sense to prefer the one over the other; there is no third way out (ex hypothesi); and certainly one takes no pleasure in the death of one's spouse and children" (2011: 215-216).
This is different from my classification, but it is also a different course of eventsso that the two are not necessarily incompatible with each other.I have been comparing the case in which the agent saves his family with cases of double effect, while Pakaluk compares the case in which the agent does not save his family with cases of double effect.
Here then the bad effect is the death of the family while the good effect is that the agent does an honourable thing in refusing to do something base, supposedly.
There are two things to note about Pakaluk's alternative reading: firstly, notice that here the proposed application of the Doctrine of Double Effect is for an intentional omission rather than for an action: the agent refuses or omits to do the base thing; none of the standard scenarios in the debate on double effect are such that the Doctrine is used to justify an omission; the Doctrine is rather normally used to justify an intervention: think, for example, of the bystander who may omit to divert the trolley from the five to the one but who, according to the Doctrine, may justifiably divert the trolley from the five to the one.Think of the surgeon who may omit to practice a hysterectomy but who, according to the Doctrine, may justifiably intervene to save the mother even though that kills the foetus.
Secondly, notice that Pakaluk's reading depends on supposing that "one both gains an honourable good (the kalon) and incurs a loss (of philoi); it makes sense to prefer the one over the other" (2011: 216).Here, as we have said, since the course of events is different, there is no problem with Pakaluk's claim that refusing to do something base is an honourable good and that losing one's family is a loss; indeed, the two are both very plausible.But Pakaluk's next claim is less plausible, where he says that it makes sense to prefer the honourable good to avoiding the loss.The problem is not just moral intuition, which suggests that most people would probably rather save their family than obtain the honourable good of not doing something base (but, again, that does depend on what this base action consists in).The problem is rather that this would not be a clear cut case, even if we inserted instead of the variable base action something pretty bad, say killing someone or betraying an important secret or something like that.
Normally, in cases of double effect, the prima facie moral balance is, on the other hand, very clear: five workers as opposed to one in the trolley problem; or thousands of lives saved by shortening the war as opposed to some children in strategic bomber. 14nother thing that all the Double Effect scenarios have in common is that there are parallel cases also involving a comparable good effect and a comparable bad effect which do not meet the conditions to be justified by the Doctrine of Double Effect.Here the structure of the Double Effect cases starts to emerge: it isn't simply that there are two effects, one good the other bad, and that the good effect overwhelms, at least prima facie, the bad effect.That is because those two things are also true of all the above cases that are not supposed to be justifiable with the Doctrine.The idea is that, in each case, both the justifiable and unjustifiable case are consequentialistically identical but that there is an intensional difference between each pair of cases: the justifiable case, in each pair, is one in which only the good effect was intended while the bad effect was merely foreseen as a side-effect.Namely, the good effect may be an end in itself or a means to another good effect, while the bad effect may not be either an end in itself nor a means to the good effect -it can only be an unintended side-effect of the good effect.It is this intensional difference that is supposed to explain why the Doctrine justifies one but not the other in each pair.So, to cite just one case, Strategic Bomber intends to destroy the munitions factory and foresees that destroying the munitions factory (as a means to shortening the war) will also cause the death of the children as a side-effect.Strategic Bomber does not intend to kill the children either as an end in itself or as a means to shortening the war (nor is the death of the children supposed to be, according to double effect, a means to destroying the munitions factory).Terror Bomber, on the other hand, intends to kill the children as a means to shortening the war.Killing the children is, for

Scenario
Terror Bomber, an intended means to his end of shortening the war, and not merely a foreseen side-effect of his plan.
In the next section I will analyse Aristotle's mixed actions to see whether the distinction between intended means and merely foreseen side-effects applies to mixed actions.

Means and Side-Effects
Can we apply the distinction between intended means and merely foreseen sideeffects to Aristotle's examples?We may try to say the following: the captain, in throwing goods overboard, only intends to save the crew and does not actually intend to throw goods overboard: throwing goods overboard is merely a side-effect of saving the crew.
Similarly, we may say, of the agent who is asked by a tyrant to do something base otherwise his family will be killed, that he does not intend to do the base thing, he rather only intends to do save his family, and doing the base thing is just a foreseen side-effect of saving his family.We may, indeed, try to say the above: but it does not sound as plausible as it does with the previously illustrated double effect examples such as strategic bombing or Bystander at the Switch.Can we really deny that throwing goods overboard is a means to saving the crew?Can we really deny that doing the base thing is a means to saving the family?It could be suggested that we look for parallel examples: after all, it may be said, to say that Strategic Bomber does not intend to kill the children is not in itself plausible; its plausibility depends on contrasting Strategic Bomber with Terror Bomber, who clearly intends to kill the children.What are, then, the equivalent parallel cases for Aristotle's examples of mixed actions?
Let us take throwing goods overboard.Take an alternative captain who may throw goods overboard because he wants to compromise the profitability of the route, or because he may be opposed to the goods in question (suppose they are transporting alcohol or neuroenhancements).The consequences of the two cases are supposed to be the same: the ship's cargo is lost.And we can certainly say that the two captains have different attitudes to throwing goods overboard.But is this difference in attitude a matter of intention?It seems to be rather a matter of having different reasons or goals: Aristotle's captain's goal is to save the crew; the new captain's goal is to prevent the dissemination of intoxifying substances.We may argue about who has the morally preferable goal, but, again, the difference does seem to be in their goal and not in their intending to throw goods overboard.It seems, indeed, that in both cases the captains intend to throw goods overboard as a means to their respective goals of saving the crew and preventing the dissemination of intoxifying substances.We may try to say that Aristotle's captain does not really intend to throw the goods overboard because he would no longer throw goods overboard if the tempest would quiet down.But this strategy does not go very far, because the other captain would also no longer throw the goods overboard if he realises that he is actually carrying medicines; or maybe he reads a convincing argument in favour of the use of neuroenhancements.The counterfactuals are similarly close in both cases.
What would Aristotle's case have to look like so that we may convincingly talk of side-effects as opposed to means?Before attempting to modify Aristotle's example, we should look at his other scenario, involving someone who has been blackmailed by a tyrant to do something base: if he does not comply, the tyrant will execute his family.
Suppose that the one in question is a security official and that the base thing in question is to reveal a sensitive secret to the tyrant.Can we say that, when the security official passes the relevant information to the enemy, he does not intend to do so and that passing information to the enemy is not a means to saving his family but only a side-effect of saving his family?Again, that does not seem convincing, and this impression does not change even when we contrast our security official with another one who simply betrays his country for money.Again, they have different attitudes in that they have different reasons or goals; but that does not seem enough to be able to say that the former official does not pass on information as an intended means to save his family.
Maybe, it could be objected, we are just comparing Aristotle's examples to the wrong double effect case.Maybe if we compare them to a different case, we will get the relevant similarity.In this respect, self-defence looks promising.Indeed, Aristotle's case of doing a base thing does not seem far from a case of self-defence, as the victim is defending his own family from a threat.The structure is similar: if I don't do something that is prima facie base, such as killing or passing sensitive information to the enemy, I will lose something that is paramount to me (either my own life or the life of my children).Indeed, why think that the emergency in the case of self-defence is greater: most people would probably prioritize the life of their children over their own life.How is double effect supposed to justify self-defence and can we apply the same justification to Aristotle's case?In applying the Doctrine of Double Effect to self-defence we would probably say something like the following: I did not really intend to kill the attacker, I only intended to save my own life.And we can indeed say of Aristotle's example that the agent did not really intend to pass on the sensitive information, he only intended to save the life of his children.But the problem was that we cannot plausibly say that the agent did not pass sensitive information to the enemy as a means to saving the life of his children.Similarly, though, we cannot plausibly say that the victim who kills in selfdefence does not kill the attacker as a means to saving his own life.

Recall Mangan's Condition 3 in his classic definition of the Doctrine of Double
Effect: "that the good effect be not produced by means of the evil effect" (1949:43).The evil effect is not allowed to be a means to the good effect: so the killing of the attacker (bad effect) cannot be a means to saving my own life (good effect), and the base betrayal of my country (bad effect) cannot be a means to saving the life of my children (good effect).What that shows, though, is not just that Aristotle's example does not meet "The class of actions which Aristotle would describe as 'necessitated' is evidently the very same class to which a DE [Double Effect] analysis would apply.Both involve acting so as to bring about (or not acting to prevent) something bad, in circumstances of constraint such that this could not be done except by being responsible for something taken to be worse.Suppose there are two goods such that each is reasonably wanted but one is correctly preferred to the other, and suppose that circumstances are such that the goods have become so bound together that the only action which suffices to preserve the one is something which destroys the other; then we may say either that in the circumstances the agent is necessitated, by fear of losing the greater good, to act so as to destroy what he usually would want to preserve, or that in the circumstances the only action by which the agent may, as usual, preserve the greater good happens also to destroy the lesser good.An upshot is that DE applies to a much broader range of actions than is typically thought" (2011: 224-25).
Suppose we ignore condition 3 and we say that, just as the attacked did not intend to kill the attacker but only intended to save his life, so in Aristotle's first example I did not intend to do the base thing but only intended to save my family and similarly in Aristotle's second example I did not intend to throw the goods overboard but only intended to save my crew.The problem is that this is just an endorsement of utilitarianism, as the above can be said of all the classic counterexamples to utilitarianism.Take the judge who knowingly puts to death an innocent man to avoid a riot (Rawls 1955, Foot 1967): the judge did not intend to put to death an innocent man but only intended to avoid a riot.Take the surgeon who kills a healthy hospital visitor to save five patients in desperate need of organs.The surgeon did not intend to kill the one but only intended to save the patients.Even if one thinks that the Doctrine of Double Effect is a concession of the absolutist to utilitarian counterexamples, then if there is to be anything to the Doctrine of Double Effect then there must be cases that it does not allow despite their utility. 15akaluk's suggestion of extending double effect has exactly the same problem, which we can illustrate by looking at his example of the two goods who happen to be bound together: "two goods such that each is reasonably wanted but one is correctly preferred to the other, and suppose that circumstances are such that the goods have scenario to meet condition 3, it must be the case that the bad effect -throwing goods overboard -is not an intended means to the good effect -saving the crew.The bad effect of throwing goods overboard must be just a side-effect of saving the crew or of the action which results in saving the crew.Indeed, we need not imagine Aristotle's captain as standing on the edge of his ship literally throwing boxes in the water so as to reduce the ship's weight in order to increase the chances of withstanding the storm.
We may imagine that, in order to withstand the storm, the captain operates a sudden turning manoeuvre so as to direct the ship towards less dangerous waters.The captain knows that this sudden turning manoeuvre is likely to result in loss of cargo.Still, the captain does not want to or intend to lose cargo; he rather wants to and intends to save ship and crew from the storm.This scenario appears to be structurally similar to prominent side-effects scenarios from the double effect debate, such as Bystander at the Switch or Strategic Bomber.The captain foresees loss of cargo but does not intend to get rid of his cargo; were there to be a different available manoeuvre with similar probabilities to save ship and crew from the storm but which did not involve likely loss of cargo, the captain would rather have chosen this other manoeuvre.But this possibility is just not there, so that the captain, if he wants to do his best to try and save ship and crew, must choose the sudden turning manoeuvre that he knows will likely also cause loss of cargo.In this scenario, just like in Bystander at the Switch and the Trolley Problem, loss of cargo (the bad effect) appears to be a side-effect of the rapid turning manoeuvre to save ship and crew (the good effect).
Notice the similarity with Bystander at the Switch and the Trolley Problem: Now compare this interpretation of Aristotle's example to the following alternative possibility: the captain has been paid by a rival company to compromise the profitability of the route, his ship, and shipping company.Luck has it that the current storm gives the captain the perfect chance to realize his plan without being caught.The captain opts for a sudden turning manoeuvre which will according to his calculations cause the cargo to fall in the water and be lost at sea.He suddenly turns the ship in order to drop as much cargo as possible.it happens, the captain knows that this manoeuvre will also direct the ship towards less dangerous waters and minimize the chances of losing members of the crew.But this captain does not really care about that: indeed, he is being paid by the rival company in order to compromise his own shipping company.For all he knows, the loss of crew members may have also been an effective strategy; but his deal involved only loss of cargo and he is personally indifferent to what happens to crew members.Now, this captain does not appear to have just lost cargo as a foreseen sideeffect to his attempt to save the crew.On the contrary, he is indifferent to what happens to crew members and cares only about getting rid of as much cargo as possible so as to increase economic damage to his current employer.

Scenario
This story emphasises the side-effect character of the loss of goods in my previous interpretation of Aristotle's example.In this respect, then, given the equivalence between my interpretation of Aristotle's example and Strategic Bomber and Bystander at the Switch, then this new story should in turn be equivalent to scenarios such as Terror Bomber and Fat Man.But it is not.Here we certainly have an agent who intends the bad effect, loss of cargo.But here the bad effect is not an intended means to the good effect, saving the crew; while in Fat Man the bad effect -pushing the fat guy down the bridgeis a means to the good effect, saving the five.Similarly in Terror Bomber, the bad effect, killing the children, is a means to the good effect of shortening the war, even though it is not a means to the destruction of the munitions factory. 16n both the trolley problem twin scenarios and in the terror bombing and strategic bombing twin scenarios, the overall good goal remains stable: shortening the war is both Terror Bomber's goal and Strategic Bomber's goal, and saving the five workers is the goal of the bystander in both Bystander at the Switch and Fat Man.But in my reading of Aristotle's captain, saving the crew is not the overall goal in both cases, once pursued through bad means and the other time pursued through good or indifferent means but a bad side-effect.
Ironically, though, it is not difficult to find the candidate reading of Aristotle's example which is supposed to be equivalent to Terror Bomber and Fat Man so as to be contrasted with the side-effect case in which the captain operates a sudden turning manoeuvre in order to bring the ship in less dangerous waters while knowing that the sudden manoeuvre is likely to cause loss of cargo.The equivalent to Terror Bomber and Fat Man is just Aristotle's original case, in which the captain throws goods overboard in order to save his crew.Recall Aristotle's own words and contrast them to my side-effect reading of the scenario: Something of the sort happens also with regard to the throwing of goods overboard in a storm; for in the abstract no one throws goods away voluntarily; but on condition of its securing the safety of himself and his crew any sensible man does so.(1110a5-20) Here we have a case in which the agent brings about the bad effect (throwing goods overboard) as a means to achieving the good effect -saving the crew.Contrast this with my reading in which the captain suddenly turns the ship in order to direct it towards safer waters while knowing that the sudden manoeuvre will likely result in loss of cargo.
We have now identified two parallel scenarios -Aristotle's original example and my sudden manoeuvre reading -which are equivalent to the classic twin cases in the double effect debate such as the Trolley Problem and Terror and Strategic Bomber in respect to a common good goal (saving the crew, saving the five, shortening the war).
But, it may be objected, there is a difference between the two parallel scenarios from Aristotle and the others in the double effect literature: namely, the agents in the two parallel scenarios from Aristotle no longer perform the same action.One throws goods overboard, the other one suddenly turns the ship.But that's not a problem because, even if indeed in Terror Bomber and Strategic Bomber the two perform the same action of dropping bombs, in the Trolley Problem the two do not perform the same action, as the bystander in Bystander at the Switch must only flip a switch while the bystander in Fat Man must push the fat guy down the bridge.And it won't do to insist that the Trolley Problem is still different because one can reformulate Fat Man so as to have the bystander just flip a switch which will cause the fat guy to fall off the bridge (see Kamm 2007, Otsuka 2008); because the same reformulation can be offered for Aristotle's parallel examples, by stipulating, say, that the captain in both cases must just operate a lever, double effect scenarios 18 while it is life against goods in Aristotle's scenario.So the bad effect, in Aristotle's scenario, is just too far, morally speaking, from the good effect -and that would explain why we find Aristotle's verdict of moral permissibility plausible.

Intended Means and Bad Effects
Three aspects of this issue must be discussed: (1) Can we say that Aristotle's example is a counterexample to the Doctrine of Double Effect because it posits a case in which it is just intuitively morally obvious that it is permissible to choose a bad means to a good end?
(2) Can we reformulate Aristotle's example so as to find the kind of moral balance between bad effects and good effects that we have in other double effect scenarios?
(3) Aristotle's other example, of having to do something base in order to save one's family, may be more promising with respect to establishing the normal moral balance between bad effects and good effects that we find in standard double effect cases.
Aristotle says that any sensible man would throw goods overboard to save his crew.This must imply that Aristotle had in mind goods whose moral value cannot be compared with the lives of crew members.As we already said, we must suppose that, were the captain not to throw goods overboard, the lives of crew members would be lost but the goods themselves would not be necessarily lost (if one were to suppose that the goods would be lost no matter what, then the situation is obviously relevantly morally different).Now suppose the goods in question are the only boxes still in circulation of some life-saving medicine.People will die if these goods were to be lost.Now, on this supposition, Aristotle's example would have the sort of moral connotations of the standard double effect cases, where there is an alternative between life and life (see point (2) above).And then it is just a matter of getting the proportion right, according to Condition 4: "that there be a proportionately grave reason for permitting the evil effect" (1949: 43).In the Trolley Problem, the proportionately grave reason for permitting the 18 The same is true of the abortion case and of self-defence, even though the case of deathinducing pain-alleviation is slightly different: there it is life against pain.Still, that is also a case in which the most fundamental interests of persons are at stake, rather than, say, commercial interests.
evil effect of killing the one is the five lives that would be saved.In Terror Bomber and Strategic Bomber, the proportionately grave reason for permitting the killing of the children is the many lives that would be saved by shortening the war.Similarly, in Aristotle's example, the proportionately grave reason for throwing goods overboard that would save the lives of some sick patients at home is saving the lives of the sailors.
But what if the goods in questions were just, to use modern examples, mobile phones?Then, just as Aristotle says, any sensible man would throw them overboard to save the crew.But the moral imbalance between mobile phones and human lives is such that any sensible man would throw mobiles phones overboard to save the crew.The idea, here, is that if the proportionately grave reason is indeed grave enough (proportionately speaking, again), such as in the case of, say, the lives of 10 sailors against boxes containing hundreds of mobile phones, then it is irrelevant whether the throwing of the goods overboard is an intended means or a merely foreseen side-effect.Any sensible virtuous person would allow that throwing hundreds of mobile phones overboard to save the lives of 10 human beings is not just morally permissible but morally obligatory, and that such is the case even if throwing the goods overboard is indeed an intended means.
The problem for the Doctrine of Double Effect seems to be that there must be some sort of minimal threshold for the sort of bad effects that are not allowed to be intended means; if just any bad effect is not allowed to be an intended means, then the Doctrine runs into some serious problems of plausibility (see point (1) above).Let us illustrate this by looking at the four conditions of Mangan's classic definition: A person may licitly perform an action that he foresees will produce a good and a bad effect provided that four conditions are verified at one and the same time: 1) that the action in itself from its very object be good or at least indifferent; 2) that the good effect and not the evil effect be intended; 3) that the good effect be not produced by means of the evil effect; 4) that there be a proportionately grave reason for permitting the evil effect (1949: 43).
The point is that if the case of throwing mobile phones overboard as an intended means to save the lives of ten crew members cannot be morally justified by the Doctrine, then the Doctrine is implausible.Here a defender of the Doctrine will have to say that throwing mobile phones overboard is not an evil effect.And, indeed, one may want to argue that the loss of goods is not evil even if it has negative economic consequences that will affect the lives of human beings.But the problem with this reply is that, if throwing goods overboard is not evil, then Aristotle's case is no double effect scenario.And if throwing goods overboard is evil, then Aristotle's case is a counterexample to the Doctrine because the Doctrine would not allow one to throw goods overboard in order to save the crew.As we saw, there are two possible ways to resist this conclusion, both of which are problematic: one could try to argue that throwing goods overboard is not an intended means to saving the crew -we saw a possible reading along these lines, even though the most basic and natural interpretation of Aristotle's example is as an intended means.
The other strategy consists in denying that the goods are morally irrelevant and suppose, for example, that they are indispensable medicines which will save lives.But that does not sit very well with Aristotle's remark that any sensible man would throw the goods overboard.Here one would have to suppose that the medicines would save some lives but that the number of lives that the medicines would save is clearly inferior to the number of crew members that would die.On this interpretation, then, Aristotle's case would be somewhat like Fat Man: there is a relatively clear utility on the side of the good effect (saving the five, saving the crew) but the Doctrine still says that the agent is not morally permitted to intervene because the agent would bring about an evil intended means (killing the fat guy, throwing the medicines overboard).Still, in Aristotle's example, if the balance of utility is clear in favour of the lives of the crew, intuition still tells us that the agent is permitted to intervene (and, indeed, may be morally obligated to intervene) despite the fact that throwing goods overboard would be an intended means.
So that the only way to get moral intuition on the side of the Doctrine of Double Effect, in this case (namely to deny that it is permitted to throw goods overboard), appears to be to stipulate that the number of sick people that would be saved by the medicines is clearly greater than the number of crew members that would be saved: namely, overturning the balance of utility.But that, and here we are back at the beginning, would mean that the case is not one of double effect because then Condition 4 would not be met as then there is no proportionately grave reason to permit the evil effect.
What with Aristotle's other example, having to do something base to avoid your family being killed (as of point (3) above)?"...if a tyrant were to order one to do something base, having one's parents and children in his power, and if one did the action they were to be saved, but otherwise would be put to death... " (1110a5-20).This seems to be indeed an example in which there is the kind of balance between good and bad effect which is typical of double effect cases.But this seems to be, again, a case in which the agent who would submit and do the base thing would do it as an intended means to the end of saving one's family from the tyrant's menace. 19I shall just mention some representative major contributions to this particular stream of the debate: Foot 1967, Bennett 1980, Quinn 1989, Boyle 1991, Marquis 1991, Fischer/Ravizza/Copp 1993, McMahan 1994, McIntyre 2001, and Wedgewood 2011 (and I have already referred to my own work on the topic).
Summa II has sometimes been denied (notably by Anscombe), but Double Effect is not normally thought to go further back than Aquinas.Only recently scholars have started to point out a possible connection between the Doctrine of Double Effect and Aristotle's discussion of action in Book III of the Nicomachean Ethics (Pakaluk 2011 2 ).In discussing hekousion and akousion (normally translated with 'voluntary' and 'involuntary'), Aristotle offers some examples -of what he calls 'mixed actions' -which are strikingly similar to the sort of examples which are put forward in discussions of the Doctrine of Double Effect.Here are Aristotle's two examples in the classic translation of Ross (1925): Condition 3 of the Doctrine of Double Effect.It also shows that the standard account of self-defence fails to meet Condition 3 of the Doctrine of Double Effect.And, after all, if the Doctrine's most influential example is Aquinas's discussion of self-defence, it cannot be that self-defence does not fit the Doctrine.Should we then just drop condition 3? This seems to be for example Pakaluk's position in arguing in favour of a significant relationship between Aristotle's examples and the Doctrine of Double Effect: Let us take stock of our argument so far: we have looked in some detail at two examples of mixed actions from Aristotle and we have compared them with the classic examples that are brought to bear in the double effect debate.We have found that Aristotle's examples could be revised so as to be interpreted as the sorts of examples that the Doctrine of Double Effect could justify, but that the most basic and natural reading of Aristotle's examples is one in which the agent brings about the bad effect (throwing goods overboard, doing something base) as an intended means to the good effect (saving the crew, saving one's family).Now it may be objected to the claim that the most basic and natural reading of Aristotle's example is as intended means to a good end that the debate on the Doctrine of Double Effect shows that the distinction between means and side-effects is itself problematic and that it can therefore not be argued, on grounds of the distinction between means and side-effects, that Aristotle's cases are not structurally equivalent to standard double effect cases.Indeed, a lot of the debate around the Doctrine of Double Effect centres on whether it is at all possible to offer a coherent working distinction between intended means and merely foreseen side-effects.19But that whole debate is beside the point here because, if we do give up on the distinction between intended means and merely foreseen side-effects (or on the moral relevance of such a distinction for moral permissibility and moral justification), then we are just giving up on the Doctrine of Double Effect as a whole because, as the four conditions of the Doctrine make clear, the Doctrine depends on the distinction between intended means and merely foreseen sideeffects.So the option of denying the claim that Aristotle's examples are cases of intended means on the grounds that there is no coherent working distinction between intended means and merely foreseen side-effects cannot rescue the thesis that Aristotle's examples of mixed actions are cases of double effect.
Effect provides a justification for what Strategic Bomber does. 13I will not go through the comparison of Aristotle's examples with all of the aforementioned scenarios, but just summarize the structural similarities: