This would not beg the question. Those who deny the normative significance of the distinction don’t usually deny that the distinction itself can be drawn, and that killing thus has essential properties that letting die doesn’t.

It seems to me that this ‘would show just the reverse’ only if we are all agreed that the property in question is indeed essential to either killing or letting die, and that its being essential is morally significant. But someone who disagreed that there is a significant moral distinction between killing and letting die is unlikely to concede this.

Given constraints on time and space, it would be impractical for me to address all of your recent comments (many of which I think I agree with, though some of which perhaps not). And, I’m not an expert on Kant himself. But, I wanted to comment in particular on one theme that seems present in your thoughts . . . .

I also agree (if this is part of your point) that few, if any, moral principles ‘exist’ as part of the universe itself in some sort of context-less, disembodied, ‘be-rational-only, please’ sense. Principles must, I believe, be understood in context, and can be ‘grounded’ in and explained by a larger context if indeed they are valid. For example, viewed this way, the Golden Rule (in its various closely-related versions) is one central outcome of human reciprocity, which itself is an outcome of our evolutionary development as social beings of a particular type, which itself is an outcome of … etc. And, the Kantian notion (I think) that some action is only OK if there would be a good/acceptable outcome if everyone did it, should be understood in the context within which it is valid and not thought of as if it exists in some sort of context-less, disembodied, absolute sense. Put simply and very imprecisely (and at risk of misinterpretation), the valid aspects of this principle are applicable and beneficial, to the degree that they are wise in a wide range of situations, because of the ways humans sense, think, and ‘work’; because of the social nature of our species; and (importantly) because of the ‘effective’ role that those aspects play as part of our overall social-moral dynamics to help facilitate human survival. In other words, valid principles or ‘dynamics’ can be grounded, and if a principle or ‘dynamic’ can’t be grounded, it probably doesn’t exist merely by virtue of itself. If it can’t be grounded, it might only exist as such in the human imagination. And, it may not even be a good idea.

A recent post of mine (number 7 in the Kamm chapter 2 thread) expresses more of my views, and earlier posts do as well.

The first mistake is to assume humans act only “rationally,” which all forms of Kantianism mistake. The second mistake is that any human can act “unconditionally impartial,” behind a “veil of ignorance,” or some similar impossible nostrum. The third mistake is to assume only categorical imperatives have “moral grounding.” The injunction, “Do No Harm,” is not derived from a categorical imperative, but from self-referential considerations into analogous situations, and “Do No Harm” is the only moral imperative I know that is moral, proscriptive, universal, and indubitable — without a bit of categorical logic. The fourth mistake, similar to Aquinas’s Natural Law theory, mistakes “instrumental reasoning” (e.g., practical reason) for the “means and ends” of natural objects, e.g., persons. Valuation of ends by means is perfectly valid, but evaluation of a person (fact) as a “means” (valuation) crosses the Fact/Value Divide. On these four mistakes alone, the Kantian moral system fails.

Granted, it’s “intuitional” similarity with the Golden Rule has kept it alive, usually among religionists, and particularly Lutherans. (Wouldn’t Kant have loved that?) But even Kant’s transcendental deity is merely a hypothesis to ground his morality, because the absence of “consequential” considerations from the hyper-rational categorical imperative leaves one with what? No judgment follows from any categorical imperative as to its suitability, ergo, a need for a mythic deity (just in case)!

As modern moral and ethical thought evolves, clearly some of morality’s demands, such as “duty,” and “self-control” have features that must be incorporated into our overall axiological systems. But all the deontological systems are entirely inadquate to the task “in themselves,” whether utilitarianism or categorical imperatives. As James Q. Wilson observed in 1993, the “moral sense” comes from many processes, including empathy, fairness, duty, and self-control. Even a virtue theorist such as myself holds the failsafe position of “Do No Harm,” as a duty one cannot trespass. Where two or more harms are involved, the choices will have to be weighted and valued on a host of considerations, not a single trump card.

Clearly, empathy, as the Scottish philosophers suggested, plays a significant role in our axiology. But that “analogous emotion” also precludes the “veil of ignorance,” and the standard of “hyper-impartiality.” Or consequentialism alone. The standard “moral puzzle” is to save either (i) one’s beloved, or family member, or (ii) five strangers. Anyone who chooses (ii) by some artificial calculus is not human. And to presume otherwise is to DEMAND more than ethics or morality SHOULD.

Philosophy teaches us to seek wisdom, not merely reason. The difference between all the moral deontological systems, based on hyper-rations, versus the smartness of virtue theory is prudence. As long as axiologists hold humans to improbable, if not impossible, standards, such as utilitarians and Kantians, no one is paying attention to either ethics or morality. One encounter with Peter Singer has people running for their humanity, Barbara Herman, while Aristotle, Solomon, Foote, Wilson, et alia offer sane suggestions of making “right choices.”

So, to answer your final question, should our moral sense reflect our understanding or human behavior and intuition, the answer is an unqualified YES. And utilitarianism, divine commands, and Kantianism reflect neither our biological natures or our bases for valuing the best choices. But even T. H. Huxley avoided the Naturalistic Fallacy. Ethics must take our biological natures into account, but not then make them morally or ethically normative, lest we get Social Darwinism of Herbert Spencer. But I think Wilson’s on the right track by his pluralist approach, which incorporates all valuation factors, motivations, and systems, rather than “fit” a single rational construction.

Thanks very much for your helpful response. Regarding the second half of your response, i.e., re knock-on consequences and intuition, … I agree that intuitions should be handled with (i.e., interpreted with) great care, for a variety of reasons of course. That said, you mention as a possibility that human intuitions may be influenced by our assumptions about knock-on effects. I agree, and I think it’s much more than a possibility. If I understand the terms correctly (e.g., “knock-on”), I do think that our human psychology, intuitions, and behaviors are influenced by the fact that we know (or at least sense) — consciously, “subconsciously,” and “intuitively” — that, if we choose an action that creates a consequence for someone, it is at least a possibility that they or someone else could someday do something similar to us. Would we like it if they did? Mirror neurons and/or their related mechanisms are at least part of this equation. (Aside from the question of direct reciprocity from person-to-person, I’m talking here about lessons, example, precedence, way of thinking, and related matters.)

Consider one of the simpler Trolley Cases, for example: At least one of the reasons that someone might choose to save five and endanger (effectively, kill) one by switching a run-away trolley from one track to the other is that people can understand (at least a little) that they are at risk if they choose to walk on a trolley track. A person wouldn’t want to be sacrificed just because he was walking on an inactive (he thought) stretch of track, but he can at least understand the situation. On the other hand, a person walking on a bridge (over a track), or standing safely to the side of a track, does not expect to be abruptly pushed from behind by another person onto a track. The very thought of such a thing greatly threatens our human sense of stability and control and “right” to life, even more so than does the notion of walking on a side track and then getting hit by a trolley that has been switched to that track. If we permit (in our minds) the idea of pushing people onto tracks, or into roads, or whatever, we intuitively realize that “all hell will break loose” and that our very way of understanding the world, and our “security” in the world, will have to change. This “intuitive” knowledge (i.e., leading to a choice to switch the track to save five, but not leading to a choice to push a bystander in order to save five) reflects, I would argue, the “intuitive” understanding on the part of an agent that “if I were the third party, would I want this done to me, or would I at least be ‘comfortable’ with it and ‘understand’?” We intuitively know that we do have to live with each other, and we intuitively want at least some senses of control and security over our lives, and we know that people do learn from our example (and we from theirs), and we know that, if we adopt a way of thinking, other people can and may adopt it too. I.e., this type of knock-on consequence is real, and it is (also) really a part of our intuitions, I believe.

Thus, if my understanding (including of terms) is correct, and if we are searching for a moral theory that is as sound and accurate as possible, then, given the choice, wouldn’t it make more sense to choose one that understands and reflects this aspect of human behavior and “intuition” rather than one that disallows it from being part of the explanation? Of course, I’m not quite sure yet what Kamm is saying in these regards.

Brian,
Yes, I agree that it seems implausible to account for permissibility in Loop by reference to the agent’s actual intentions, since the act would still be permissible if the harm were intended. And I like your suggested interpretation of Kamm according to which it is only the possibility of not intending the harm – which perhaps is determined by the structure of the case – that is relevant to the explanation. (It will be interesting to see if Kamm eventually endorses something like this.) The question then is whether the possible intentions really add anything to the structural explanation. If not, perhaps intentions could be left out of the explanation altogether.

Jeff,
It seems to me that referring to hypothetical consequences (such as what would happen if everyone did something) might, as you suggest, turn out to be a good way of explaining (im)permissibility. And it may be that such consequences should always be part of the explanation. I doubt, though, that the latter would be the case on Kamm’s methodology (as I understand it from pp14-15). On her approach, whether hypothetical consequences should be invoked will depend, I think, on whether they track facts about (im)permissibility, which are to be identified in the first instance through reflection on the cases.

Regarding knock-on consequences such the effects of setting a precedent, I suspect Kamm would want to rule these out as potential explanatory factors since I presume her cases are meant to be causally-insulated from the rest of the world. It seems possible, though, that our intuitions about the cases are influenced by our assumptions about knock-on effects even if there are supposed to be none, meaning that the intuitions should be handled with care…

Regarding the roles of the intention and mental state of the agent, answers to those questions can be informed by the following, I think: The answers should involve (address?) two questions, among others: First, what would happen if everyone did it? (Kant); and Second, how is the lesson/example component of the intention-action combination accounted for by the various theories? Indeed, the latter question, I think, involves the question of differences (and relationships) between consequentialism and non-consequentialism themselves. When someone acts under certain conditions and with a certain mental view or intention, are the only “consequences” that are counted in the equation the living or dead bodies resulting immediately from the incident and action? Or, are the following also considered to be “consequences” of the action in any way: the “lesson” conveyed to others; the precedent set; the “example”? By asking “what would happen if everyone did this?” (i.e., did the combined intention-action) and by considering likely future ripple-consequences of the “lesson” that was “given” to others (of the intention-action and how it was responded to), you may be able to tip the scales in favor of one conclusion or the other. People, i.e., third parties, learn (in some ways) from examples they see. And consequences, of course, don’t cease one week after an action. I hope this is helpful in some way. Sorry if not. Good luck.

You bring up a number of interesting issues related to Kamm’s remarks about what explains (im)permissibility. As Matthew points out, Kamm claims that bombing in the Munitions Grief Case is permissible due to the fact that one is acting because the effects will occur, though not intending that they occur. So she does, at least here, seem to think that one’s mental states can affect the permissibility of his acts. But as you point out, one could bomb with the intention that the effects occur; and I’m strongly inclined to think that whether or not one has this intention is irrelevant to the permissibility of the act. This is especially clear, I think, with respect to the Loop Case. If diverting the trolley in that case is only permissible if one does not intend the death of the one on the looping track, then if the person on the looping track is someone that one despises and would like to kill, such that one could not possibly, psychologically speaking, divert the trolley without intending his death, it will be impermissible to save the five. But this seems clearly wrong; diverting the trolley still seems permissible, and both the structural facts criterion and the kind of victim-centered approach that Kamm says she intends to pursue would seem to be able to account for this. If one can permissibly be killed in order to save five in the ordinary Loop Case, then the fact that the person who diverts the trolley has a certain mental state shouldn’t affect the permissibility of that killing.

Perhaps Kamm’s view is that it is the structural features of a case that ultimately determine the permissibility of acts, but also that whenever an act is permissible given the structural features, it will be at least possible to perform the act without intending any of its harmful side-effects. On this view one’s actual mental states make no difference to permissibility, so that diverting the trolley is still permissible in my altered version of the Loop Case. The permissibility can be partially explained by the fact that one could (in some non-psychological sense of could) have diverted it without intending the death of the one. Sound plausible?

But it seems to me that there is only a contingent relationship between (a) and (b). Though we are inclined not to attribute an intention to inflict evil to the agent in Loop (or in any other case with the same structure), the agent could, it seems to me, intend the evil in such cases. (For example, the agent in Loop could see it as an advantage of her diverting the trolley that it will run over the person on the track.) Whether she does seems to depend in part on what mental states she has at and before the time of her pressing the button. But these mental states are not (clearly) specified in the description of the case.

Since the causal/constitutive structure of the case and the agent’s psychology can come apart in this way, it seems that Kamm must decide which explains the permissibility of diverting the trolley. Regarding Loop, I wasn’t clear which way she would go.

More generally, I wasn’t sure whether Kamm thinks that (im)permissibility (i) is always to be explained by reference to structural facts, (ii) is always to be explained by psychological facts about the agent, or (iii) may be explained either by structural or psychological facts (or both) depending on the case. I suspect (iii), but this leaves me unsure about what is meant to be doing the explaining in specific cases such as Loop.

Kamm’s later comment that she intends to adopt a victim-centred rights-based approach suggests that neither the psychological facts about the agent nor the causal/constitutive structure of the case ultimately explains (im)permissibility. Rather, this is explained by the rights and moral status of the victim(s). But clearly, either the structural or psychological facts, or perhaps both, are still meant to play some role in the explanation. Presumably, whether the victim’s right is breached in a particular case will depend on the structure of the case, or the psychology of the agent, or both. But I’m not sure which. Perhaps this is something that will become clear later.

I agree that intuitions vary (to a degree) among people. If my background (partly involving research with people) is typical, I suspect that, regarding any given moral case, the results (on a large number of humans) would show a mainstream of answers along with many different variations, e.g., roughly in a bell curve or in multiple bell curves superimposed on each other. The variability is probably greatest with some of the more difficult, nuanced, and rare/contrived cases. So, how to explain the mainstream answers AND the variations? Based on what I’ve read, which is only chapter 1 as well as this thread, I’m still curious as to whether Kamm is seeking to find (and claim) highly logical moral principles (which sometimes involve exceptions to exceptions, etc.) that, as she sees them, exist within all humans? If so, what form might an “error theory” take to explain variations? Will the “error theory” include such things as: Variation A exists because Person A drank wine before responding; and Var B exists because Person B imagined himself looking into the scared faces of the five people he would save while imagining that he could only see the back of the person he would push under the trolley; and Var C exists because Person C wasn’t raised well?

I agree wholeheartedly that intuitions can tell us a lot that can inform any theory of morality. And, I agree that we should understand them. But, I doubt that humans have specific exacting logical rules etched somehow inside of us. (Again, I don’t know if Kamm claims this. My apologies if she doesn’t.) We do have social-moral “faculties” and enabling biological mechanisms, and certain types of predispositions, but (in my view) they don’t exist in the form of specific logical rules as we normally think of moral rules. Explaining the variations between these Kammian types of specific rules and the varying intuitive human answers using an “error theory” such as that mentioned above does not seem very compelling. There are other theories that can explain mainstream-and-variations as part of the theories themselves, and they are grounded in a scientific understanding of how humans most likely came to be. Perhaps that’s where Kamm sees intuitions as being grounded, i.e., in evolved human nature? If anyone can help further clarify Kamm’s view for me on these issues, that would be most helpful. Looking forward to reading chapter 2.

Brian and John: Kamm would, I think, ask you to specify the cases to which you have different intuitions than she does. Then, she would think about whether it is a genuine conflict of intuition or just a verbal conflict. If genuine, she would then try to come up with an error theory as to why your intuition may be in error. Failing that, she may give up her intuition. If you follow her intuitive approach, she would think that you should do the same.

Brian: in fact, Kamm develops DTE in order to show that one can intend the ends without intending the means. So she would agree with you about that point. I think she’ll discuss this in greater detail in later chapters (using a party case example).

John: Consider the Transplant Case. Is one permitted to take the kidney of a healthy person to save five other people? If the intuition/judgment is no, then this case may begin to demarcate the limits of consequentialist thinking. Of course, consequentialists may be able retell the story in terms of consequences. But then they would be providing an error theory regarding this intuition and its putative implication. Seen in this light, Kamm is similarly trying to accommodate the (consequentialist) intuition that aggregation matters, but in a nonconsequentialist way.

Kamm is not a meta-ethicist, and to a great extent the value of her approach is demonstrated by her first-order achievements in this book and previous work. She is hardly the first to appeal to intuitions — there is a sense in which most utilitarians also rely on such appealsso. Sidgwick certainly did!

One of the things that are unique about Kamm’s methodology is the extent that she appeals to intuitions. Sidgwick was happy with a handful, Ross with quite a few more, but Kamm is remarkably persistent in pursuing intuitive nuances and identifying (and labelling) what she takes to be subtle normative differences between cases. She is certainly different from Rawls (and most other ethicists) in not employing the method of reflective equilibrium in a way that involves fairly rapid movement from intuition to principle and back again, with many intuitions being dropped fairly quickly when they don’t seem to fit some suggested principle. Kamm gives a far greater priority to intuitions, and covers a far greater range of them before coming up with a ‘covering’ principle.

It seems to me that if one accepts that moral intuitions have a role to play in substantive moral thought — and I doubt that many truly deny this — then it is hard to see why it should be better to work with a narrow rather than broad range of intuitions.

Kamm briefly suggests two ways of understanding what she is doing. One is that she is tracking a ‘psychologically real’ mental structure — she approvingly cites Mark Hauser’s claim that we might all share an innate ‘moral grammar’. The other suggestion is that she is tracking what is implictly presupposed by our responses to various cases. (There are views of mental content on which there wouldn’t be a real distinction between the two.)

This means that Kamm consciously sees herself as mapping our existing moral scheme. This raises the question some of you have already raised — the question of the move from what we think about moral matters (assuming for the moment that Kamm is really tracking that, rather than an utterly personal set of responses) to what is actually true. But Kamm (unlike, say, Ross) is quite happy to admit that there is this gap. She thinks that there is a further stage that is required once we’ve identified the deep principles that guide moral thought — we need to ground them in some plausible set of values or theory. She warns us that she will do rather little of this in this book. It’s a pity she doesn’t say more about this further stage. In assessing whether the principle we identify are justified, can we help appealing once again to our intuitions? If not, then what could we appeal to? (Robert Audi’s recent defence of ‘Kantian intuitionism’ is one possible answer.)

Also, I am confused on the extent to which we can use the consequences to determine how we should act. Kamm argues that a deontologist does not soley rely on the consequences of an action to determine what we should do, so it seems that the consequences do play some role. What is the demarcation between when we can be consequentialists and when we should be utilzing her deontogical principles? If it is our intuitions that provide this demarcation then some argument needs to be forthcoming that justifies the privlidged position these intuitions play.

Matthew – I’m inclined to agree with your point about the Munitions Grief Case. I’m inclined to side with Sidgwick on many of these issues, so I tend to have difficulty accepting many of the distinctions that Kamm, as well as proponents of the classical DDE attempt to make to make (e.g. “in order to” vs. “because”, “intending” vs. merely “foreseeing”). Still, I think Kamm’s treatment of this case is particularly difficult to accept. The response that you suggest on Kamm’s behalf would seem to put the bar for intending a consequence implausibly high, such that in order to intend that it occur one must be intensely focused on bringing it in particular about; it must figure prominently in one’s thought process regarding what he will do (your willingness to try multiple means of bringing it about criterion is helpful in highlighting this, but since intention is a mental state I think it misses the point). But this seems implausible, since at the very least I often (I’m inclined to think always, despite Kamm’s arguments that we can intend the ends without intending the means) intend the means to my ends despite the fact that it is the ends that I am focused on and that guide my decision-making process.

Jiang Yi: Kamm’s nonconsequentialism is different from Rawls’s in a number of ways. For example, she is not a contractualist. Also, her endorsement of Sobjectivity 4 means that she does not think that one always has to give priority to the worse off.

Arto: If there is only one pilot available, and the pilot can’t change his motive within the timeframe, I would say that he fails to do what he ought to do if he acts. Ought he in such a case then refuse to act? I think an affirmative answer would follow from a version of DDE, according to which one should not intend harm even if the harm will be part of a greater good. Recall that the original version of DDE says that one should not intend harm, even when the harm will be a means to a greater good. In this case, the harm is not even a (necessary) means to a greater good. So it seems that DDE or a version of it would forbid it.

Kamm might respond, but we have already agreed that the act of bombing the munitions factory is justified because of its great good consequences, even though there will be some foreseen harms. But I would respond that the act was initially justified because the harms were described as “foreseen” harms, which on DDE would be justifiable. But once the bomber intends to act only in order to produce the harm, these harms are no longer foreseen but are instead intended. And on DDE, this is forbidden.

Kamm might say instead that surely because of its great good consequences, this act is justified no matter what. But such a response would just be begging the question against DDE. Whether DDE is a valid principle or not is of course a separate question. But I don’t see that this example challenges it.

Kamm does rely on the distinction between rightness of acts and goodness of motives, and you are right that the distinction may threaten her DTE. In fact, I think we will see later that Kamm will argue that it is a mistake to try to find the source of the permissibility status of an act in the intention of an agent. This is consistent with her “victim focused” nonconsequentialism, I think.

Jeff: I’m not sure about your distinction, but Kamm believes that human intuitions can have probative force regarding moral truths, and the human rational faculty can develop valid moral principles by reflecting on these intuitions and deciding which should be kept and which are errors. See, e.g., Chapter 1 part 2 for her methodology.

not an easy one to summarize, plenty of interesting stuff (a great idea to have the reading group!). Your comments A & B made me think of this:

One reason to think that tactical bombing with bad intentions may be permissible relies on the distinction between rightness of acts and goodness of motives. The act is permissible (whatever the motive), but the motive is morally bad. (Compare: saving someone from drowning may be the right thing to do, but if one’s motives are selfish they have no moral worth).

[Is your suggestion that if there is only one pilot available, who has bad motives, which cannot be changed in the timeframe, he ought to refuse to do the tactical bombing, because it would be impermissible with bad motives?]

I’m not sure if Kamm relies on that distinction though. If she does, maybe it should be applicable to the Munitions Grief Case as well. So that if it is permissible to bomb in such a situation, it should be permissible with bad intentions as well (if the badness of the intentions does not affect the rightness of the act). But that threatens to undermine the whole distinction between “because” and “in order to”. I suppose there’s more on this in the next chapters. Any thoughts?

Arto L