Print This Post Print This Post

I. INTRODUCTION

Chapter 9 builds on the topic, introduced in chapter 8, of what happens when there are conflicts between different rights. Previously, Kamm has considered what we (or, ‘an agent’) ought to do when faced with a conflict between different people’s negative rights, and has argued that we should minimise transgression of rights. The discussion of rights becomes more complicated in chapter 9, as she introduces consideration of positive rights, and also looks at the factors (including, but not limited to, interests) that should sway us in the direction of according one right rather than another.

In what follows, I will loosely structure my summary around Kamm’s section headings, for the purpose of helping people to find the topics discussed here in Kamm’s chapter. My own comments are enclosed within square brackets.

II. TYPES OF CONFLICT

Kamm begins by noting that the allocation of rights may not be the only factor relevant to deciding how conflicts of rights should be resolved. To illustrate this point, she observes that, whilst we may agree that the right not to be killed is ‘stronger’ than the right not to have one’s leg cut off, someone threatened with having his leg cut off may justifiably kill his attacker in self-defence, since in such a case the attacker’s right not to be killed is ‘weakened’ by his presenting himself as a threat. To avoid such complicating factors, Kamm tells us that she will consider only conflicts among ‘unweakened’ rights.

[Kamm does not explicitly define what she means by rights being ‘stronger’, ‘weakened’, etc., but from her discussion is seems that rights that are ‘stronger’ relative to others are those that we ought to accord (or have good reason to prioritise), for whatever reason. Given that she says she will discuss only conflicts among ‘unweakened’ rights, she apparently does not take rights to be ‘weakened’ by their coming into conflict with other rights, or by those who bear them being in the sort of situations she describes in the thought experiments throughout the chapter. I wonder whether others think (1) that a case can be made for some rights being ‘weakened’ by circumstances, and if so, (2) whether the cases she discusses in the chapter are indeed all cases of ‘unweakened’ rights.]

Kamm then introduces her typology of rights, and says that she is interested in cases where rights conflict because the duties correlated with them conflict. She categorises rights as follows:
– negative vs positive rights (correlated with, respectively, duties of non-interference and duties of provision). These may protect
o the same interests (e.g. the negative right not to be killed vs the positive right to be saved)
o different interests (e.g. the negative right not to have one’s leg broken vs the positive right to be saved)
Further, a positive right may be:
derived from a violation of a negative or positive right (e.g. someone whom we have harmed, or whose positive right we have earlier failed to accord, may now have a right to our help)
pure (e.g. a positive right to aid arising from a promise, which may be contractual)
mixed (e.g. when one person has a duty to prevent another from violating the negative or positive right of a third person. Kamm notes that some such cases may more accurately be described as involving pure positive rights, and because of this I think that it remains a little unclear exactly how mixed positive rights differ from other positive rights).

She summarises all of the different possible conflicts between different types of rights in a list on page 287. This list can be doubled, since a conflict can be agent-relative (one agent has conflicting duties, or an internal conflict about which right to accord) or agent-neutral (the rights in question give different agents conflicting duties, in the sense that one agent’s according one right prevents the other agent from according the other right). She refers to agent-relative conflicts as type A, and agent-neutral ones as type B.

The list of rights conflicts structures the discussion that follows. I will not reproduce the list here, but will indicate as we go along which conflicts are being addressed, and where in the table they can be found.

III. SPECIFIC CONFLICTS

A. Negative versus Negative

Kamm begins by considering conflicts between negative rights covering the same interest, giving a single agent conflicting duties. This is (A)(1)(a), where ‘(A)’ denotes agent-relativity, and (1)(a) locates the conflict in the list on page 287. Having noted in chapter 8 that Jeremy Waldron believes that making all rights negative and agent-relative rules out conflicts of rights, she introduces a counterexample to Waldron’s view:

Trolley Scenario 1. We may send a trolley down either track A (where it will kill Joe) or track B (where it will kill Jim), otherwise it will go down track C and kill five people. If we want to save the five, then contra Waldron we face a conflict between respecting either Joe or Jim’s negative right not to be killed. Kamm believes that, in such a case, ‘Joe and Jim each has a right that the agent use a random decision procedure to select which one of them will be killed’ (p. 288). Aside from indicating this belief, she provides no argument for why a random decision is the correct way to resolve the conflict. Kamm takes this case to show that, even when a conflict of rights reflects a conflict between fulfilling different duties, the conflict may arise from reasons other than our doing our duty (since it may not be our duty to save the five), and that a conflict of negative rights arises from redirecting the trolley does not make redirecting it impermissible.

[Kamm does not elaborate on this right to a random decision procedure, which Joe and Jim both have, and which apparently comes into play only when Joe and Jim’s other rights are threatened. What are our duties with regard to this right, and how do they compare with our duties with regard to Joe and Jim’s other rights? Imagine that we have a very small reason that inclines us to save Joe over Jim (perhaps we find his company slightly more entertaining than Jim’s): we are then faced with a conflict between respecting Joe’s right to life and Jim’s right to a random decision procedure. How might we resolve this conflict? In this case, we are dealing with a conflict between Joe’s negative right not to be killed and Jim’s derived positive right that a decision be made in a certain way. (Kamm would no doubt wish to claim that our small reason for preferring Joe is morally irrelevant; however, since its introduction changes the type of rights conflict at issue, I think a case can be made for its relevance.) Kamm later says that ‘a negative right not to be killed can be stronger than (even an aggregate of) positive rights derived from the negative right protecting the same interest’ (p. 291), which appears to sway the decision in favour of saving Joe. This suggests that some such dilemmas can be characterised in terms of more than one type of rights conflict, and our intuitions about what we should do may vary depending on which characterisation we use.]

She then considers Trolley Scenario 2. We can either send the trolley along track A, where it will kill Joe, or along track B, where it will kill Jim and Susan. Kamm believes that we may send the trolley to Joe, because ‘In resolving this sort of case, each person has a right only to be balanced against his equal and opposite number, and the remaining person on one side helps to determine the outcome’ (p. 288). She notes that this conclusion requires further argument, possibly like that of chapter 2, which has received some discussion here. Next is:

Trolley Scenario 3. This is like Scenario 2, except that instead of being killed, Susan would be only slightly bruised. Should Susan’s negative right not to be bruised affect the decision about where to send the trolley? Kamm’s ‘sense’ is that Susan’s right is ‘morally irrelevant in this context’ (p. 289), and that there therefore seems to be a Principle of Irrelevant Rights (which she does not define further, but compares it to the Principle of Irrelevant Goods discussed in chapters 1 and 2). Her explanation for this conclusion is that what Susan ‘would suffer is much less than what Jim would suffer if his right is transgressed; their rights would be transgressed in the same way, and from Jim and Joe’s personal points of view, it is not irrelevant whether Jim or Joe is the one to survive’. Kamm takes this case to show that ‘an agent who is faced with a conflict of negative rights, where an equal number of rights protecting the same interest lies on either side, should not necessarily always minimise violations of negative rights, in particular when other rights at stake protect lesser interests’. She notes, however, that some ‘lesser interests’—such as Susan’s standing to lose a leg rather than being bruised—may be significant enough to affect the decision.

[Kamm’s explanation of the moral irrelevance of Susan’s getting bruised strikes me as an adequate explanation of why Kamm feels it to be irrelevant, but not as a good explanation of why it ought not to factor in the decision. Of course getting bruised is not nearly as bad as getting killed, and because of this we may be inclined to resist allowing it to factor in the sort of decision she is asking us to consider, but without further argument this does not show it to be irrelevant. As with many of the conclusions that Kamm derives from such thought experiments, there is a need to step back from her intuitive reactions to individual, highly contrived, imaginary scenarios, and to consider whether the intuitive conclusions she has generated form the sort of coherent pattern that is needed to support the moral claims she makes. Without doing this, it is difficult to tell whether the fact that she finds certain proposals (such as allowing Susan’s getting bruised to affect the decision about where to send the trolley) intuitively unappealing plausibly rebuts those proposals; or whether allowing that a few intuitively unappealing proposals may, all things considered, be plausible is the price we have to pay for some maximally plausible moral theory.]

Kamm next turns to agent-neutral conflicts between negative rights covering the same interests ((B)(1)(a)). Unless R kills Joe, P will kill Jim and Susan. If Jim and Susan have a mixed positive right against P, P will face a conflict between respecting Joe’s negative right and fulfilling his mixed positive duty to Jim and Susan. Kamm points out that she argued in chapter 8 that ‘it is not irrational for a rights theory to prohibit one agent from transgressing one person’s negative right for the purpose of minimising the transgression of other people’s negative rights by other agents in cases where there is no specific right these other people have to be aided’ (p. 289). She now claims that this argument implies that such a prohibition can remain even when the other agent’s victims have mixed positive rights to assistance against the agent.

She then considers whether conflicting negative rights protecting the same interest could be different in strength, such that ‘in a conflict, we should not necessarily be indifferent as to which negative right we transgress’ (p. 290). She illustrates this with the Multiple Option Case, in which a villain directs a trolley towards five people, and we can save their lives by:
(i) pushing John in front of the trolley, which stops it but kills John
(ii) redirecting the trolley to a track where it will kill Rachel
(iii) redirecting the trolley to a track where it will kill Peggy and Rita.
Kamm claims that we should choose (ii) over (i), but in the absence of (ii) we should choose (iii) over (i): ‘Hence, in this conflict, we should kill more people—transgress more negative rights—in one way rather than transgress someone else’s negative rights in another way’. She returns to consider the Multiple Option Case in later discussions about the strength of rights.

Kamm moves next to agent-relative conflicts between negative rights that protect different interests. An agent must choose between directing a trolley towards Joe, killing him, or towards Jim, breaking his leg. She tells us that ‘Other things being equal, the agent should avoid violating the right that protects the more important interest, at least when the differential strength of each right is a function solely of the interests it protects’ (p. 290). Some differences, however, are morally irrelevant: e.g. if we face a choice between breaking Paul’s leg or breaking Simon’s leg a bit more severely, we should not allow the slight difference in severity to sway us towards breaking Paul’s leg. This, she claims, is another application of the Principle of Irrelevant Rights.

[It seems to me that, especially in cases where we are considering conflicts between rights protecting different interests, the Principle of Irrelevant Rights is going to be applicable only in a very limited number of cases, since it will not always be obvious whether one of two possible outcomes will be worse than another, or whether such a comparison can even be made. E.g., if we face a choice between paralysing Paul and bringing it about that Simon’s wife and children reject him, it is not obvious which, if either, of the two outcomes is worse than the other (and whether the extent to which it is worse makes the difference relevant or not). Further, it seems that we cannot hope to calculate which is worse without knowing more about Paul and Simon: the extent to which paralysis is bad for Paul might vary depending on whether he is a professional athlete or a couch potato, and the extent to which being rejected by his family is bad for Simon might vary depending on whether he is a cantankerous loner or a committed family man.]

What if we face a trolley-redirection choice between breaking the legs of Jim and five others, and killing Joe? Kamm’s answer is that, ‘Because on Jim’s side each of the six as an individual would suffer a far smaller loss if his negative right is transgressed than Joe would, it would be wrong to avoid transgressing the large aggregate of all of their rights. The principle that would justify this is giving preference to the person who will be worst off if his right is transgressed’ (p. 290). This conclusion holds even if Jim and friends face the more undesirable outcome of paralysis—however, Kamm claims (without further elucidation) that this is consistent with ‘there being some greater number of people, each of whom would suffer such a significant loss, for whose sake it is permissible to cause Joe the greater loss’.

Kamm tells us next that the ‘different interests’ case is further complicated if the strength of the rights is not solely a function of the interests they protect. She asks us to consider again the Multiple Option Case, in which choosing option (i) would paralyse John rather than kill him. She maintains that we should choose (ii) over (i) and, in the absence of (ii), we should choose (iii) over (i): ‘Hence, it is not always the case that when there is a conflict among fulfilling negative rights, we should transgress the one that protects the weaker interest’ (p. 291).

[Kamm is stretching our intuitions about this case to the limit here, to the extent that her conclusion begins to look like a sound basis for a reductio of her own argument. She is no doubt right that most would find the idea of killing someone by pushing them in front of the trolley intuitively more objectionable than killing someone by redirecting a trolley; but intuitively most would also agree that the death of two people is worse than the paralysis of one. The claim that one intuition should trump the other needs to be defended with more than mere assertion.]

B. Negative versus Positive

Kamm considers the following agent-relative negative/derived positive conflict of rights protecting the same interests ((A)(2)(a)(i)): I must kill Joe to save Jim and Susan from a threat that I introduced to them in the past, and which is about to materialise (that is, I have a positive derived duty to save Jim and Susan). The threat facing Jim and Susan is the same as that with which I would kill Joe. In considering ways in which I might kill Joe, Kamm wishes to argue for the following claims: (1) the interests of each party remain constant throughout the different cases, yet killing is sometimes permitted and sometimes not; (2) that it is sometimes impermissible to kill Joe shows that a negative right not to be killed can be stronger than (even an aggregate of) derived positive rights protecting the same interest. Echoing the earlier discussion, these claims suggest that the strength of a right does not depend solely on the interests it protects.

Kamm lists five ways in which I might kill Joe ‘in connection with’ saving Jim and Susan:
(a) I need to kill Joe to harvest his organs for Jim and Susan
(b) To divert a trolley away from Jim and Susan, I must set a bomb which I foresee will kill Joe
(c) I must push Joe in front of a trolley headed for Jim and Susan
(d) I must redirect towards Joe a trolley headed for Jim and Susan
(e) My saving Jim and Susan from a killer trolley will result in their continued breathing directing deadly airborne germs towards Joe.
Kamm believes that (a), (b), and (c) are impermissible, but that (d) and (e) are permissible. Since the interests at stake are the same in each case, the variations in permissibility are not due to any difference in interests, but to ‘a difference in the strength (or perhaps even existence) of rights not to be killed that is not a function solely of the interest it protects, but of how we kill’ (p. 292). She introduces, in two parts, a possible nonconsequentialist principle (which she later, although not here, refers to as a principle of permissible harm) to ‘account for’ the differences: (1) ‘We may permissibly transgress someone’s negative right as an effect of a greater good (or component thereof), or of means having the greater good as its noncausal flip side, even when the person’s right being infringed plays a causally useful role in sustaining the greater good by dealing with new threats that arise from our efforts to produce the greater good’; and (2) it is impermissible to transgress someone’s negative right as a direct effect of, or as part of, what we do to produce the greater good—though this does not mean that the right against doing this is absolute’ (p. 292). She notes that this principle focuses not on the agent, but on the victim of the transgression, and that it implies that (d) is permissible but not (a).

[The moral relevance that Kamm attaches to various means of causing a state of affairs is confusing. The principle she describes here is certainly consistent with holding that (a), (b), and (c) are impermissible whilst (d) and (e) are permissible, but it does not help us to understand why this is. To my mind, it no more addresses a morally relevant aspect of the different cases than would a principle asserting that—assuming that (a), (b), and (c) take place during daylight hours and (d) and (e) at night—we may permissibly transgress someone’s negative rights at night but not during the day. Perhaps she means the principle to be action-guiding without having to be explanatory, but even in this case it needs to be motivated with an appeal to morally relevant considerations.]

Kamm goes on to point out that we can employ the Choice Test and the Goal Test (introduced in chapter 8) to show that the negative right not to be killed is stronger than the positive right to be saved. I will not recap her discussion here, but it can be found on pages 292-3.

Kamm claims that her conclusions about agent-relative negative/derived positive conflicts of rights protecting the same interests ((A)(2)(a)(i)) can also be applied in cases where the positive right in question is pure or mixed (that is, in cases (A)(2)(a)(ii) and (iii)), and where the conflict is agent-neutral (that is, in cases (B)(2)(a)(i)-(iii)): in all such cases, it is permissible to kill using methods (d) and (e), but not using methods (a), (b), or (c). In what follows, she takes options (a)-(e) to characterise different ‘manners’ of killing.

The discussion of the (2)(a) category ends with consideration of negative/positive conflicts involving the same interests, in which one or more side also has rights protecting lesser interests: ‘For example, Joe’s negative right not to be killed conflicts with Jim’s contractual right to have his life saved and Susan’s contractual right to be saved from a broken leg. Or, the right of Jim not to be killed and the right of Susan not to be bruised conflicts with Joe’s contractual right to have his life saved’ (p. 293). Kamm asks two questions: first, whether in such cases we should give equal chances to each side, ‘even if this means killing n people to save n people’ (she suggests that ‘we should not be willing to kill to save merely to provide an equal chance’). Second, she asks whether anything but ‘the major interest (i.e., life)’ should determine whether we kill, and reminds us that earlier she claimed that, in conflicts between negative rights, lesser interests could sometimes be relevant to deciding which option to choose. She gives an affirmative answer, supporting her decision with the following case: ‘I suggest that it would be permissible to redirect a trolley away from Jim and Susan to fulfil positive rights to save him from death and her from being completely paralysed, even though we foresee that Joe will be killed. Hence, the significant lesser loss (paralysis) covered by a right makes killing Joe permissible, even though Jim on his own would have had a right to an equal chance that he be saved or that Joe be killed’ (p. 294).

[I am not convinced that consideration of this case tells us anything about the strength of positive rights in such conflicts. Were the case to involve no positive rights but only negative rights, Kamm would agree that it would be permissible to direct the trolley towards Joe (since she would allow that the prospect of Susan’s paralysis is morally relevant to the decision about whether to send the trolley away from Jim). As a result, the introduction of positive rights seems superfluous here: our intuitions about negative rights may plausibly be doing all the work.]

Finally, Kamm briefly considers negative/positive rights conflicts in cases where the interests at stake are different. There are two possibilities: the negative right protects the stronger interest, or the positive one does. To illustrate the former type of conflict, Kamm asks us to ‘suppose that I must kill someone in order to fulfil a contractual obligation to save each of 5 people from being completely paralysed, assuming that total paralysis is not as bad as death’ (p. 294). Kamm claims that this would not be permissible, but notes that her conclusion may change as the number of people at risk of being paralysed increases.

[As above, it seems plausible that our intuitions about negative rights are doing all the work here.]

To illustrate the second type of conflict, Kamm asks whether we may break Joe’s leg to save Jim from death. She suggests that we may, providing we do it in manner (d) or (e), but that ‘compensation is owed for this harmful interference’.

C. Positive versus Positive

Mercifully, Kamm tells us that she will ‘only consider some of the many possible cases’ involving this type of conflict (recall that, with agent-relativity versus agent-neutrality; pure, derived, and mixed positive rights; and same versus different interests, there are twenty-four different types of positive rights conflicts). In order to decide whether there is anything analogous to her principle of permissible harm (that is, a principle governing cases involving not-aiding rather than killing), she considers analogues of cases (a)-(e):
(a) Jim and Susan have a positive right against a doctor to an organ transplant. Since there is a shortage of organs, the doctor omits to give Joe his medicine in order that he will die, leaving his organs free for the doctor to use to fulfil Jim and Susan’s rights.
Kamm claims that this would be impermissible: ‘Hence, we should not always decide whom to aid merely on the basis of the number of rights protecting the same interests that we can fulfil’ (p. 295).
(b) We are setting a bomb that will divert the trolley heading towards Jim and Susan, who have a positive right to our aid. Therefore, we cannot simultaneously implement means to save Joe from another trolley, and he also has a positive right to our aid.
Kamm concludes, ‘it is our using certain necessary means to help Jim and Susan that results in Joe’s death, but it is permissible not to save Joe. (This is by contrast with the impermissibility of using means whose direct effect is causing Joe’s death in [the original (b)].)’
(c) Joe is about to be blown off a bridge, into the path of a trolley headed for Jim and Susan. This would kill him but stop the trolley, saving Jim and Susan. We can easily save Joe. All parties have a positive right to be saved.
It is impermissible not to save Joe, according to Kamm.
(d) Joe has a positive right to a life-saving raft from us, which is on its way to him. Then we see that Jim and Susan are drowning. We redirect the raft to them, fulfilling their positive right to a raft from us.
Kamm: it is permissible to redirect the raft, thereby failing to fulfil Joe’s right.
(e) If I fulfil Jim and Susan’s positive right against me to be saved, they will consume resources that I would need were I to fulfil Joe’s positive right that I save him.
Kamm: it is permissible that I save Jim and Susan.

Since some of the cases described are permissible, and some not, despite the interests involved remaining constant, Kamm concludes that ‘This is a step in showing that there is a principle dealing with not-aiding that is in some ways similar to (though in some ways different from) the particular principle of permissible harm described earlier’ (p. 295). She does not, however, attempt to formulate such a principle.

Kamm goes on to consider agent-relative conflicts between a pure positive right and a derived positive right, where both rights protect the same interest ((A)(3)(a)(ii)). She asks whether Joe’s bodyguard should ‘forego saving Joe from paralysis in order to stop or (somehow) undo his (the bodyguard’s) impermissibly paralysing Jim’ (p. 295). She answers that he should: avoiding depriving Jim of an ability that he has independently of the bodyguard is more important than his (contractual) promise to Joe. She compares this with the following agent-neutral case: ‘if the bodyguard faced a conflict of positive duties (correlative to rights) between (a) saving Jim from someone else’s failure to abide by Jim’s negative right and (b) saving Joe from paralysis due to a natural disaster, (a) would not necessarily take precedence over (b)’ (pp. 295-6). So, that each case involves a rights violation does not itself determine whom to help when the interests at stake are the same. She concludes that ‘This contrast between the role of rights violations in A (agent-relative) and B (agent-neutral) cases is interesting because some [e.g. Scheffler and Scanlon—see footnote 25] seem to have concluded on the basis of the second (agent-neutral) case that a rights violation makes no difference to whom we should choose to aid when interests are the same. The case where the bodyguard would himself have violated the right of one of the people he must now aid defeats this conclusion’ (p. 296).

[I am not familiar with the history of this debate (i.e. the Scheffler and Scanlon material that Kamm mentions), but it seems to me that there is not enough here to draw any useful conclusions about the role that rights violations play in deciding how to resolve conflicts. Other differences between the two cases (and other possible variations) could account for the conclusions that Kamm draws, such as agent-relativity versus agent-neutrality, or contractual versus non-contractual rights. At the very least, we need to consider more cases, preferably explaining and analysing our intuitions about them as we go.]

Kamm then applies these conclusions about rights violations at the social level. Suppose that the government is the agent who acts to fulfil positive rights. Citizens have positive rights to both medical care and police protection (against the violation of negative rights by others). Her conclusion about rights violation means that the right to police protection would ‘not necessarily’ take precedence over the right to medical care. On the other hand, if we take one role of the government to be helping people to ‘share the burdens of each one’s duty to prevent his violation of negative rights and to fulfil positive rights that derive from prevention (or repair) of violations of negative rights’ (p. 296), and if these duties take precedence over fulfilling duties associated with other, pure or derived, positive rights, then ‘perhaps we could justify according a right to police protection before a right to medical care’. This is because the role of the government means that the mixed positive rights in question are reduced to derived positive rights, and changes the case from agent-neutral to agent-relative.

Next, Kamm considers conflicts between pure positive rights protecting the same interests, noting that the now-familiar issues of aggregation and balancing are relevant here; and she also questions the relevance of whether a right is contractual or ‘a human (person) right to aid (assuming that there are such rights)’ (p. 297). Despite this, she does not address directly the relevance of a right’s being contractual. Instead, she introduces the following case: ‘Suppose that Joe and Jim each contracted and paid for a portion of a life-saving drug to which no one has an equal right simply in virtue of being a person. Joe is on one island and Jim on another, and we can only get to one in time to save him. I believe that each has a right to an equal chance to the drug. However, what if Susan is near Jim, also needs the drug to save her life, and could successfully use the part of it that Jim does not need, but she did not contract for it’. In this case, Kamm decides that Susan’s need should not sway the decision in Jim’s favour, but that if the drug is sent to Jim, Susan should also be helped.

[This seems an inadequate treatment of contractual rights. The example treats contractual rights as a special case of positive right, but it does not tell us anything about how contractual rights compare to other sorts of rights, or how to resolve conflicts between contractual and ‘human’ rights—and whilst Kamm does not claim to address these issues, they seem highly relevant to her discussion, especially as several of her examples (e.g., those involving bodyguards) involve contractual rights.]

Kamm then returns again to the issue of aggregation: of what to do when we face a choice between according positive rights protecting weighty interests and according positive rights protecting lesser interests. She tells us that ‘because we will not have to transgress a negative right of the person who has the weightier interest … fulfilling the aggregation of rights that involve significant but lesser losses may often override fulfilling the positive right of the person with the weightier interest’ (p. 297). However, the importance of the because of/in order to distinction arises again here: we may not omit to save Joe’s life in order to make available his organs for the development of a serum that will save 100 people from paralysis; but we may choose to distribute paralysis-reversing medicine to the many rather than save Joe’s life, even though we know that the medicine will have a potentially fatal side-effect on them, and we do this because we know that once Joe dies we can use his organs to create a serum which will save the lives of the many.

Finally, Kamm considers an implication of her claim that it may sometimes be permissible to accord many positive rights protecting lesser interests instead of a single (or fewer) positive rights protecting weightier interests: ‘Where x, y, and z are decreasing losses, n and m represent the number of such losses, and m>n: If n(y)>x, and m(z)>n(y), transitivity implies that m(z)>x’ (p. 298). This implies that it is permissible to relieve the headaches of a billion people instead of saving Joe’s life (where all parties have a right to our aid). To avoid this conclusion, Kamm suggests that we compare the sizes of x and z to ‘make sure they are not too far apart’. If they are too far apart, the aggregate of m(z) can never outweigh x, although it may outweigh n(y), and n(y) outweighs x. She recognises that this entails that our choice between two options may depend upon the availability of a third (not-to-be-chosen) option, and claims that this is not irrational.

The dubious plausibility of according relevance to such third options has already received discussion in this blog (see, for example, Toby Ord’s discussion of a case from Alastair Norcross (comment number 14 on chapter 2) and Nir Eyal’s introduction of a Sidney Morgenbesser case (chapter 6 summary)), and Kamm has her own justification for it. She asks us to suppose that, if we are unable to choose more than one option, it is permissible to save 1,000 people from quadriplegia over saving Joe’s life, and to save 10,000 people from paraplegia over saving 1,000 people from quadriplegia, but that we should choose to save Joe’s life over saving 10,000 from paraplegia (because not to do so would be ‘disrespectful of Joe’). What if we choose to save 1,000 people from quadriplegia (for brevity, ‘save 1,000 quadriplegics’) instead of saving Joe, and as a result of this decision Joe dies before we have acted: should we switch our resources to saving 10,000 paraplegics? According to Kamm, ‘This would be wrong … because the only legitimate reason for Joe now being dead (and so not available for help) is that we had a superior reason to help the quadriplegics rather than him. We should act only on the reason (i.e., help the quadriplegics) that would have justified our letting Joe die, given that he is dead because we decided to act on that reason. We must not forget the history of how we came to be at the point of helping the quadriplegics when we decide between them and the paraplegics’ (p. 298).

[Whilst Kamm may be satisfied that the appeal to history justifies her unwillingness to change her decision in this case, it seems to me that it will not be a good guide to action in other, analogous, cases. For example, imagine that Joe is so ill that the resources we have would be insufficient to save him, meaning that if we used them on him they would be wasted. Since we do not want to waste our resources, helping Joe is not an option, and so we are faced with a choice between saving 1,000 quadriplegics and saving 10,000 paraplegics. We therefore choose to save 10,000 paraplegics. As a result of our decision, but before we have acted, two things happen: Joe recovers just enough for our resources to be sufficient to save him, should we choose to use them in this way; and the threat of quadriplegia facing the 1,000 disappears, meaning that they no longer have need of our help. What should we do: should we go ahead and help the 10,000 paraplegics as we originally intended, or should we change our minds and save Joe? Following Kamm’s reasoning, the only legitimate reason for Joe now being well enough to be able to make use of the resources is that we had a superior reason to help the paraplegics, and given this history, we should now act on that reason. As a result, we face a conflict between, on the one hand, 10,000 people’s rights to be saved from paraplegia plus Kamm’s imperative that we act on our original reason (to help the paraplegics), and on the other hand, Joe’s right to have his life saved plus the fact that, according to Kamm, it would be ‘disrespectful of Joe’ to help the paraplegics over him. At the very least, there is a plausible case to be made for helping Joe, and therefore changing our original decision—and my view is that this is what we ought to do.]


Comments

  1. 1. Posted by Fiona Woollard | September 4, 2007 9:05 am

    I really enjoyed your summary, Rebecca. I agreed with most of your comments, particularly your questioning of Kamm’s methodolgy in, for example, the discussion of the relevance of Susan’s leg bruise. So, I thought I’d make a small picky point.

    In your final paragraphs you discuss Kamm’s claim that it is not irrational for our choice between two options to depend upon a third (not to be chosen) option. I find that the idea that such unused options can make a difference distinctly uncomfortable, however, I do not think your counterexample to Kamm’s appeal to history works. In your example, it does not seem as if the only legitimate reason for Joe now being well enough to be able to make use of the resources is that we had a superior reason to help the paraplegics. It is not as if Joe would have been killed if the paraplegics had not been there. Presumably, if we had not had a superior reason to help the paraplegics we would have already used the resources to help the quadreplegics, so there is a sense in which the only legitimate reason for us to be able to help Joe is our superior reason to help the quadreplegics. However, I think Kamm could argue that there are significant disanalogies.

    (a) In Kamm’s case, Joe no longer needs our help because he has died, whereas in your case, the quadraplegics no longer need our help because they have recovered. It may be disrespectful to ignore the history when a group can no longer be helped because they have lost out, but not when they can no longer be helped because they already acheived the desired benefit by other means. (How do your intuitions about your case go if we make the quadraplegics die rather than recover?)

    (b) In Kamm’s case, Joe is no longer able to be helped because the reason to help the quadraplegics was superior to reasons for “helping Joe or helping the paraplegics”. Thus we ignore this history if we now decide that we have more reason to help the paraplegics. In your case, Joe is able to be helped because the reason to help the paraplegics was superior to the reason to help the quadraplegics. (Helping Joe was not an option). Thus we do not ignore this history in the same way if we now help Joe.

  2. 2. Posted by S. Matthew Liao | September 4, 2007 5:13 pm

    As with Fiona, I too enjoyed your summary/analysis, Rebecca! I have a question regarding what Kamm has said, and some questions for you.

    1. On pp. 291-292, Kamm discusses five ways in which Joe might be killed in connection with saving Susan’s and Jim’s lives in (A)(2)(a)(i) to fulfil their derived positive right (against, e.g., me, who started the threat to them):

    (a) I need to kill Joe to harvest his organs for Jim and Susan
    (b) To divert a trolley away from Jim and Susan, I must set a bomb which I foresee will kill Joe
    (c) I must push Joe in front of a trolley headed for Jim and Susan
    (d) I must redirect towards Joe a trolley headed for Jim and Susan
    (e) My saving Jim and Susan from a killer trolley will result in their continued breathing directing deadly airborne germs towards Joe.

    Kamm points out that killing Jim in (a), (b), and (c) is impermissible, and killing him in (d) and (e) is permissible, and then draws the conclusion that this shows that interests do not determine the variations in permissibility in these cases. She takes this to be a further point against Interest Theories of Rights.

    I don’t think that this conclusion is warranted, because arguably, the interests at stake in these five cases are not the same. In (a), (b), and (c), an advocate of an Interest Theory of Rights could argue that the interest at stake is an interest not to be killed as a (necessary) means to some greater end, while in (d) and (e), the interest at stake is an interest not to be killed as a side effect. On this basis, the advocate could argue that the right not to be killed is grounded differently in these cases, which in turn would explain the difference in their permissibility.

    2. You said,

    Kamm’s explanation of the moral irrelevance of Susan’s getting bruised strikes me as an adequate explanation of why Kamm feels it to be irrelevant, but not as a good explanation of why it ought not to factor in the decision. Of course getting bruised is not nearly as bad as getting killed, and because of this we may be inclined to resist allowing it to factor in the sort of decision she is asking us to consider, but without further argument this does not show it to be irrelevant.

    Is your worry here that you think that Susan’s getting bruised *is* morally relevant? If so, I think Kamm would just modify the example. For example, suppose Susan will only get a scratch to her leg. If you don’t think that Susan’s getting a scratch to her leg is morally relevant in this case, then Kamm’s argument would still go through, I think.

    3. You said,

    It seems to me that, especially in cases where we are considering conflicts between rights protecting different interests, the Principle of Irrelevant Rights is going to be applicable only in a very limited number of cases, since it will not always be obvious whether one of two possible outcomes will be worse than another, or whether such a comparison can even be made. E.g., if we face a choice between paralysing Paul and bringing it about that Simon’s wife and children reject him, it is not obvious which, if either, of the two outcomes is worse than the other (and whether the extent to which it is worse makes the difference relevant or not). Further, it seems that we cannot hope to calculate which is worse without knowing more about Paul and Simon: the extent to which paralysis is bad for Paul might vary depending on whether he is a professional athlete or a couch potato, and the extent to which being rejected by his family is bad for Simon might vary depending on whether he is a cantankerous loner or a committed family man.

    I think Kamm would respond that there is/should be a ceteris paribus clause in operation here. After all, suppose A and B are drowing, and you can save only one, and suppose A is a brilliant scientist and B is a couch potato. All things considered, factors such as how much contribution A or B can make to the society could affect whom we should save. But presumably, in the cases that Kamm has in mind, such factors are ruled out.

    4. You said,

    The moral relevance that Kamm attaches to various means of causing a state of affairs is confusing. The principle she describes here is certainly consistent with holding that (a), (b), and (c) are impermissible whilst (d) and (e) are permissible, but it does not help us to understand why this is.

    I think Kamm offers some explanations for these judgments in Chapter 4. As she summarizes them on p. 292, (a), (b), and (c) are impermissible because

    it is impermissible to transgress someone’s negative right as a direct effect of, or as part of, what we do to produce the greater good—though this does not mean that the right against doing this is absolute.

    And, (d) and (e) are permissible, because

    We may permissibly transgress someone’s negative right as an effect of a greater good (or component thereof), or of means having the greater good as its noncausal flipside, even when the person’s right being infringed plays a causally useful role in sustaining the greater good by dealing with new threats that arise from our efforts to produce the greater good (p. 292).

  3. 3. Posted by Dave Simon | September 5, 2007 3:32 pm

    S. Matthew, you said:

    I don’t think that this conclusion is warranted, because arguably, the interests at stake in these five cases are not the same. In (a), (b), and (c), an advocate of an Interest Theory of Rights could argue that the interest at stake is an interest not to be killed as a (necessary) means to some greater end, while in (d) and (e), the interest at stake is an interest not to be killed as a side effect. On this basis, the advocate could argue that the right not to be killed is grounded differently in these cases, which in turn would explain the difference in their permissibility.

    I’m not sure why it would make a difference morally if you characterize (a), (b), and (c) as “an interest not to be killed as a (necessary) means to some greater end,” while characterizing (d) and (e) as “the interest not to be killed as a side effect.”

    The “interest not to be killed” may not be absolute, but its meaning is universal. Thus, whether an agent is killed as a “necessary means” or “a side effect,” is morally irrelevant. In all of these cases the right not to be killed is at stake. In all of these cases that right is trampled on. The thread common to all of these hypotheticals is scienter: in every case we know that our action will cause Joe to die. I don’t see how the distinction you make is relevant since in every case we kill Joe.

    Moreover, I’m not sure that the word “necessary” accurately characterizes what you want to say. If by necessary you mean that it is necessary that Joe dies as a result of our action, then we can describe our actions in every case as necessarily causing Joe’s death because that is the consequence of our actions. In (e), for example, we necessarily cause Joe’s death by allowing Susan and Jim to breathe deadly germs at Joe, resulting in his death. (If you mean something else, please comment.)

    Of course, I’m assuming that Joe dies in scenarios (d) and (e). If he does not, and there is a mere risk of death, then our situation changes and so to does the right at stake. If you meant that the right changes in this way, you did not say so. You said, in both cases, the right not to be killed was at stake but characterized that right as different vis-a-vis the means used to achieve the end (death).

  4. 4. Posted by S. Matthew Liao | September 5, 2007 6:35 pm

    Thanks for your comments, Dave.

    You said,

    The “interest not to be killed” may not be absolute, but its meaning is universal. Thus, whether an agent is killed as a “necessary means” or “a side effect,” is morally irrelevant.

    Let us suppose that we all have an interest in not dying, other things being equal. Even so, in addition to that interest, we also have, so I am claiming, interests in not dying in certain manner. For example, it is in my interest to have a dignified death (e.g. dying peacefully in my sleep) rather than to be brutally murdered. If this is right, we can begin to see how whether an agent is killed as a “necessary means” or “a side effect” can be morally relevant.

    You said,

    Moreover, I’m not sure that the word “necessary” accurately characterizes what you want to say. If by necessary you mean that it is necessary that Joe dies as a result of our action, then we can describe our actions in every case as necessarily causing Joe’s death because that is the consequence of our actions. In (e), for example, we necessarily cause Joe’s death by allowing Susan and Jim to breathe deadly germs at Joe, resulting in his death. (If you mean something else, please comment.)

    To see what I mean by ‘necessary’ here, consider the following: Imagine that you could achieve some greater end by killing me or without killing me. Suppose you chose to kill me. It could be argued that your action was more blameworthy because you chose to kill me when it was not necessary to do so to achieve the greater end. My adding ‘necessary’ here is meant to reduce the potential blameworthiness of (a), (b), and (c).

    You said,

    Of course, I’m assuming that Joe dies in scenarios (d) and (e). If he does not, and there is a mere risk of death, then our situation changes and so to does the right at stake. If you meant that the right changes in this way, you did not say so.

    No, I don’t mean that the right changes this way. You are correct to assume that Joe dies in (d) and (e).

  5. 5. Posted by Dave Simon | September 5, 2007 7:27 pm

    Thanks for your response, Matthew.

    I still have some questions/comments.

    Let us suppose that we all have an interest in not dying, other things being equal. Even so, in addition to that interest, we also have, so I am claiming, interests in not dying in certain manner. For example, it is in my interest to have a dignified death (e.g. dying peacefully in my sleep) rather than to be brutally murdered. If this is right, we can begin to see how whether an agent is killed as a “necessary means” or “a side effect,” can be morally relevant.

    What interest are you talking about here? It doesn’t seem like you are still talking about “the interest in not being killed.” An individual’s “interest to have a dignified death” is different, but not mutually exclusive from, “an interest not to be killed.” So, I assume that you mean individuals retain an interest in “being killed in certain manner.” Intuitively that is true. But is a “brutal” murder really *worse* than one where the murderer slips the victim a cyanide pill? I understand where that distinction lies, and how you can build upon this. Yet, just because there is a distinction in this *kind* of case does not mean that it will be a relevant distinction in *other kinds* of cases. The example given by Kamm, for instance, does not seem to implicate this kind of “death interest.” What difference does it make to Joe whether he is killed by a train or germs. But even assuming this interest is relevant here, it would actually militate in favor of options (a), (b), or (c) since death by germs (d) would probably more painful and gruesome than death by sudden impact with a train!

    You said:

    To see what I mean by ‘necessary,’ consider the following: Imagine that you could achieve some greater end by killing me or without killing me. Suppose you chose to kill me. It could be argued that your action is more blameworthy because you chose to kill me when it was not necessary to do so to achieve the greater end. My adding ‘necessary’ is meant to reduce the potential blameworthiness of (a), (b), and (c).

    That makes sense. I see how “necessary” reduces the moral blameworthiness in that sense. But you state that (a), (b), and (c) that “the interest at stake is an interest not to be killed as a (necessary) means to some greater end, while in (d) and (e), the interest at stake is an interest not to be killed as a side effect.” It seems that you are comparing being killed as a “necessary means” with being killed “as a side effect.” That implies that one can never be killed as a “necessary side effect.” If that’s the case, I disagree, and I think the point I made about necessity still holds: being killed as a side effect is necessary in that hypothetical, at least inasmuch as it is necessary to kill Joe in any of the other examples. We are not merely killing Joe in (e) as a side effect for no reason at all. Thus, I don’t see how the other options are better, on a “necessary” ground, than option (e).

    Please stop me if you think these points are without merit.

  6. 6. Posted by Dave Simon | September 5, 2007 7:45 pm

    I apologize for all my typos! (E.g., “But you state that (a), (b), and (c) that “the interest at stake is an interest not to be killed as a (necessary) means to some greater end, while in (d) and (e), the interest at stake is an interest not to be killed as a side effect.”

  7. 7. Posted by Rebecca Roache | September 6, 2007 9:29 am

    Fiona, thank you for your comments – I’m especially pleased that you chose to comment on this particular example, because I’d been thinking a lot about analogies of Kamm’s thought experiment.

    You said:

    In your example, it does not seem as if the only legitimate reason for Joe now being well enough to be able to make use of the resources is that we had a superior reason to help the paraplegics. It is not as if Joe would have been killed if the paraplegics had not been there.

    In talking of legitimate reasons here, I am borrowing from Kamm’s description of her own example. She doesn’t elaborate on what ‘legitimate’ means, but it seems to mean something like ‘relevant’. By stipulation, then, Joe recovers sufficiently to use our resources as a result of our deciding to help the paraplegics, and (presumably) either he would not have recovered had we not decided to help the paraplegics, or if he had recovered for some other reason, this reason is not one that we are entitled to take into account in deciding whether to change our original decision. (If anyone else has an alternative interpretation I would be interested to hear of it.) I agree that it is difficult to think of a realistic scenario in which Joe recovers as a result of our decision to help the paraplegics, but in my view it’s no more difficult than thinking of a scenario in which Joe dies as a result of our deciding to help the paraplegics in Kamm’s example.

    To address your disanalogies, you said:

    (a) In Kamm’s case, Joe no longer needs our help because he has died, whereas in your case, the quadraplegics no longer need our help because they have recovered. It may be disrespectful to ignore the history when a group can no longer be helped because they have lost out, but not when they can no longer be helped because they already acheived the desired benefit by other means. (How do your intuitions about your case go if we make the quadraplegics die rather than
    recover?)

    I have to confess to having no clear intuitions about what is disrespectful in such cases, and what is not. I was trying to work with Kamm’s intuitions here. Part of what I find frustrating about many of her examples is that she introduces concepts like this, but neither explains exactly what she means nor offers us any additional examples so that we can get a feeling for how the concept might work in different scenarios. My view is that, if one party has a stronger right to help than another, we should help the party with the stronger right. Whether one or the other is disrespected in some way strikes me as something to be balanced against other considerations (in the way that we are asked to balance Susan’s being paralysed or bruised against someone else’s right not to be killed in some of the earlier examples) rather than something that should sway the decision in a given direction in all cases of this type.

    Next, you said:

    (b) In Kamm’s case, Joe is no longer able to be helped because the reason to help the quadraplegics was superior to reasons for “helping Joe or helping the paraplegics”. Thus we ignore this history if we now decide that we have more reason to help the paraplegics. In your case, Joe is able to be helped because the reason to help the paraplegics was superior to the reason to help the quadraplegics. (Helping Joe was not an option). Thus we do not ignore this history in the same way if we now help Joe.

    It’s a different history, but I’m not sure what you mean by saying ‘we do not ignore this history in the same way’. If it is permissible to ignore some sorts of history but not others, Kamm would need to explain why. As it stands, the relevance of the history to the decision is as obscure to me as the relevance of the third option whose relevance the history is meant to explain.

    Matthew, thanks also for your comments – I’ll respond to them later!

  8. 8. Posted by Fiona Woollard | September 7, 2007 8:14 am

    Rebecca, thanks for your response! I hope my reply is not too late.
    Your comments certainly gave food for thought. I hadn’t realised that it was supposed to be a stipulation, rather than a conseqeunce, of your example that the only legitimate reason Joe recovers is our decision to help the paraplegics. I had taken the truth of the legitimate reason clause in Kamm’s example to be based on something like this:

    (a) if we had not decided to help the quadraplegics we would have been required to help Joe.
    (b) Thus the only legitimate reason we can offer to explain why we have acted in such a way that Joe is now dead is that we decided to help the quadraplegics.
    (We might need to add something like ‘given we have decided to help one of the group’)
    This seemed to be true of Kamm’s example without further elaboration of the case. I do not think it could be true of your example, even if we stipulate that Joe’s recovery would only have happened if we had decided to help the paraplegics. For even in this case, it seems that our decision to help the paraplegics is not required as a reason to justify why we have acted in such a way that Joe has now recovered – for we do not need to justify acting in such a way that Joe recovered. (Although we may need to justify our actions under another description.)

    On point (b):
    Given the history of Kamm’s example, in this situation, the original ordering of the options was: Option B (helping the quadraplegics) is better than Option A (helping Joe) and Option C (helping the paraplegics). Thus according to the orginal ordering B is preferable to C (in this context). To then decide to do C rather than B goes against this ordering and so ignores the history.

    In contrast,in your example, the history included only an ordering of Option C over Option B. We do not act against this ordering when we then decide to do A rather than C. The history does not include any previous ordering that has implications about the relative appeal of A and C.

    As you say, Kamm needs to expand on what it is involved in ignoring the history of a situation. I think that she will be able to do so in such a way that *if* unused third options can change how we order options, the results about when we must keep to an original decision and when we should change will be intuitive. However, as I noted above, I find the idea that an unused third option can change which of two other options I must take distinctly uncomfortable.

  9. 9. Posted by Rebecca Roache | September 14, 2007 10:53 am

    Matthew, thanks for your comments, and sorry to take so long to respond! I will answer the questions you addressed to me.
    With regard to my comment that Kamm has not provided a good explanation of why getting bruised should not be relevant to the decision about where to direct the trolley, you asked:

    Is your worry here that you think that Susan’s getting bruised *is* morally relevant? If so, I think Kamm would just modify the example. For example, suppose Susan will only get a scratch to her leg. If you don’t think that Susan’s getting a scratch to her leg is morally relevant in this case, then Kamm’s argument would still go through, I think.

    My issue with Kamm here is not merely that I believe that Susan’s getting bruised (or scratched) is morally relevant. In this case, I share Kamm’s uneasiness about according relevance to such outcomes. My point is that she has not provided a satisfactory explanation of why it is morally irrelevant. In the case she describes, the conflict between Joe’s right not to be killed and Jim’s right not to be killed is not the only conflict at issue. Kamm also tells us that Joe and Jim each has the right that we use a random procedure to decide which of them will be killed, and it is this right that comes into conflict with Susan’s right not to be bruised: we have to decide between using a random decision procedure (thus fulfilling Joe and Jim’s rights) or avoiding bruising Susan (thus fulfilling her right). Kamm focuses on telling us how to resolve the former conflict (between Joe and Jim’s rights not to be killed), but not the latter. The moral relevance of what stands to happen to Susan needs to be weighed against Joe and Jim’s rights to a random decision. Kamm says nothing about this directly: we are left to infer (from her conclusions about these cases) that she believes the right to a random decision to be stronger than the right not to be bruised, and weaker than the right not to be paralysed; but in my view we are left without a satisfactory explanation of why this is so.
    With regard to my comment that, in conflicts of rights protecting different interests, it may be impossible to compare two different outcomes between which we must choose, you said,

    I think Kamm would respond that there is/should be a ceteris paribus clause in operation here. After all, suppose A and B are drowing, and you can save only one, and suppose A is a brilliant scientist and B is a couch potato. All things considered, factors such as how much contribution A or B can make to the society could affect whom we should save. But presumably, in the cases that Kamm has in mind, such factors are ruled out.

    You are right that I have confused the point by introducing factors other than the badness of the outcomes, which may help us decide which to choose. Let me modify my example: Paul and Simon are identical twins, with identical careers, interests, family situations, etc. We must choose between paralysing Paul and bringing it about that Simon is rejected by his family. There doesn’t seem to be any obvious way of comparing these outcomes and deciding which is worse, and therefore which corresponding right is the stronger.
    You then reminded me of some earlier comments from Kamm, in which she tries to explain why some means of causing a state of affairs are permissible whilst others are not. However, what she offers in those comments is simply more of the same: I still fail to see why some actions are (im)permissible because of the way in which they are caused. I accept that, intuitively, some means of bringing about a state of affairs make us uneasy and some do not, but this might not entail that there is a morally significant difference between the different means. A consequentialist opponent of Kamm’s could respond that all that matters morally is the outcome of the decision, and our fluctuating intuitions in such cases demonstrate that our intuitions are not always a reliable guide to what is right and wrong. Without further argument, it seems that Kamm is ill-equipped to defend her view against such an objection.
    Fiona, thanks for your further comments, and apologies to you too for taking so long to respond.
    You are quite right that, in Kamm’s case, our decision to help the quadriplegics can be seen as justification for letting Joe die in a way that is not required in my case. Because of this, perhaps Kamm would say that it is more important not to ignore the history in her case than it is in mine. But, this is still insufficient to explain why we should stick to our original choice. Assuming that it is important that we act in a way that justifies our letting Joe die, sticking to our original choice (even after circumstances relevant to this choice have changed) would be important only if changing our minds would somehow undermine the justification; but it is far from clear that it would. And, even if it would, the importance of not undermining the justification is something that would need to be weighed against the importance of helping the paraplegics over the quadriplegics (which is permissible in cases where Joe is not also involved) before deciding how to act.
    On the topic of Kamm’s ordering of the different options, I’m not sure that what she says is strong enough to imply that option B is better than option A and option C. She says that it is ‘permissible’ to choose option B over options A and C, and option C over option B in the absence of A. This could be because option B is ‘better than’ options A and C, and option C ‘better than’ option B, but it could also mean that each set of options is approximately equal in strength. What she says, for example, is consistent with it being permissible to choose option A over option B (although not with choosing option C over option A). She does, however, structure the examples as if she means ‘better than’, so it did occur to me to wonder why she didn’t phrase the ordering more strongly.

Post a comment

Name: (required)

Email Address: (required) (will not be published)

URL:

Comments:

(Spamcheck Enabled)

This work is licensed under a Attribution-NonCommercial-NoDerivs 3.0.