This long essay, “A Theory of Human Dignity,” presents and defends a general theory of human dignity, with special attention paid to spelling out its background metaphysics, formulating and justifying a basic set of dignitarian moral principles, and critically addressing hard cases for the theory.
“A Theory of Human Dignity” is being made available here in serial format, but you can also download, read, and/or share a .pdf of the complete text of this essay HERE.
This twenty-second installment contains section V.4.
TABLE OF CONTENTS
II. Refuting the Dignity-Skeptic and Debunking a Dignity-Debunking Argument
III. The Metaphysics of Human Dignity
III.2 Real Persons and Minded Animals
III.3 A Metaphysical Definition of Real Personhood
IV. Nonideal Dignitarian Moral Theory
IV.0 How Nonideal Can a World Be?
IV.2 How to Solve the Universalizability and Rigorism Problems
IV.3 How to Solve the Problem of Moral Dilemmas
IV.4 Policy of Truth: The Murderer-at-the-Door Revisited
IV.5 One Last Thing, By Way of Concluding This Section
V. Some Hard Cases For Broadly Kantian Nonideal Dignitarian Moral Theory
V.0 How Hard Can Hard Cases Be?
V.1 Abortion and Infanticide: Introduction
V.1.1 The Neo-Person Thesis, Neo-Persons, and Non-Persons
V.1.2 A Five-Step Argument for the Neo-Person Thesis
V.3 Non-Human Animals and Their Associate Membership in The Realm of Ends
V.3.1 Real Persons and Different Species
V.3.4 Kindness to Animals Revisited: Harming without Torture or Cruelty
V.3.5 Kindness to All Living Beings: Associate Membership in The Realm of Ends
V.4 Treating People Merely as a Means
VI. Enacting Human Dignity and The Mind-Body Politic
VII. Conclusion
V.4 Treating People Merely as a Means
Human real persons are what I call subjects of dignity, and also targets of sufficient respect. Because every human real person is both subject and also a target in these ways, it’s strictly morally impermissible to treat people merely as a means to someone’s end. This doesn’t entail that it’s morally impermissible to enter into instrumental relationships with other people—for example, the person who cuts your lawn, the people who work at the grocery store, the bus driver, or your tax-preparer—since they can give their explicit or implicit rational consent to that instrumental relationship, and above all you don’t treat them merely as a means. Hence it’s perfectly permissible to enter into instrumental relationships with other people provided that you also treat them with sufficient respect for their human dignity.
Nevertheless, there appear to be at least some cases in which it’s morally permissible to treat people merely as a means. But if so, then it would be at once strictly impermissible to treat people merely as a means, because it violates their human dignity, and also sometimes permissible. Contradiction! So, we must find an adequate way of dealing with this kind of case that smoothly avoids global inconsistency in the heirarchical structuralist system of broadly Kantian nonideal dignitarian moral principles. Well, precisely what kinds of cases are we talking about? As Kant correctly pointed out, since (i) all human real persons have dignity, and since (ii) all human real persons naturally desire happiness, and since (iii) all human real persons belong to The Realm of Ends and morally owe each other equal consideration, then it follows that (iv) I have a duty to promote the happiness of all other human real persons:
Concerning … duty to others, the natural end that all men have is their own happiness. Now humanity might indeed subsist if nobody contributed anything to the happiness of others, provided he did not intentionally impair their happiness. But this, after all, would harmonize only negatively and not positively with humanity as an end in itself, if everyone does not also strive, as much as he can, to further the ends of others. For the ends of any subject who is an end in himself must as far as possible be my ends also, if that conception of an end in itself is to have its full effect in me. (GMM 4: 430, emphasis in the original)
When it comes to my promoting happiness as an end that is also a duty, thus must therefore be the happiness of other men, whose (permitted) end I thus make my own end as well. (MM 6: 388, emphasis in the original)
But this seemingly unexceptionable principle, when it’s explicitly taken by broadly Kantian nonideal dignitarian moral theory to be a first-order substantive ceteris paribus objective moral principle, leads directly to the apparent permissibility of treating people merely as a means. Correspondingly, in her famous essay, “Abortion and the Doctrine of the Double Effect,”[i] Philippa Foot spelled out what is now commonly known as The Trolley Problem:
If it’s judged morally permissible for the driver of a runaway trolley which is heading straight towards five innocent people to turn his trolley onto a spur occupied by one innocent person, thereby killing one in order to save five (aka Trolley Driver), then why is it judged impermissible for a surgeon to kill one innocent patient and distribute his organs to five other dying patients in order to save those five people (aka Transplant)?
Otherwise put: What’s the morally relevant difference between Trolley Driver and Transplant? If, to almost everyone who rationally considers these thought-experiments, it seems sometimes morally permissible to kill a few innocent human real persons, and therefore create a few innocent casualties, in order to save significantly more innocent human real persons from dying—thereby choosing and acting not only in accordance with the first-order substantive ceteris paribus objective Kantian moral principle to promote the happiness of other human real persons, thereby apparently treating the innocent casualties merely as a means, which is also in at least superficial conformity with altruistic act-utilitarian principles—then why isn’t this always morally permissible to treat people merely as a means, which of course would lead to global inconsistency—Explosion—in the hierarchical structuralist system of broadly Kantian nonideal dignitarian system of moral principles? This is what I’ll call The Trolley Problem of Saving Lives.
Now Foot’s own proposed solution to The Trolley Problem of Saving Lives is to use the killing vs. letting die distinction,[ii] and claim that (i) killing one is worse than letting five die, and (ii) killing five is worse than killing one. Put in terms of the classical negative duties vs. positive duties distinction, Foot is saying that it is worse to violate a negative duty not to harm one, than it is to violate a positive duty to save five, and also that it is worse to violate a negative duty not to harm five, than it is to violate a negative duty not to harm one. The trolley driver in Trolley Driver will either kill five or kill one, so he must kill the one. By contrast, the surgeon in Transplant will either kill one or let five die, so he must let five die.
But as Judith Jarvis Thomson very effectively shows in “Killing, Letting Die, and the Trolley Problem” and its sequel “The Trolley Problem,”[iii] this cannot be correct. That is because of the following variant on Trolley Driver, which seems clearly morally permissible to almost everyone who rationally considers it:
You are standing beside a switch at the spur, which you know how to operate, and you see the runaway trolley without a driver, so you decide to turn the trolley onto the spur and thus onto the one, so that you kill him, thereby saving the five (aka Bystander at the Switch).
Your choice here is between killing one and letting five die, hence it’s false that it’s always worse to kill one than to let five die. So, given the clear permissibility of Bystander at the Switch, the killing vs. letting die distinction doesn’t solve the Trolley Problem.
Now at this point an orthodox Kantian might appeal to the Categorical Imperative’s Formula of Humanity as an End-in-Itself, namely,
so act that you use humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means (GMM 4: 429),
and say that the moral difference between Trolley Driver and Transplant is that whereas the surgeon in Transplant uses the one “merely as a means” to saving five, neither the trolley driver in Trolley Driver nor the bystander in Bystander at the Switch uses the one “merely as a means” to saving the five. If the one by some miracle disappears or is otherwise removed from the track (say, by a swooping air-sea rescue helicopter) before the trolley reaches him, then the intentions of the trolley driver’s act and the bystander’s act are satisfied just the same.
But unfortunately for broadly Kantian nonideal dignitarian moral theory, this suggestion is apparently refuted by Thomson’s ingenious “loop variant”on Bystander at the Switch.[iv] This variant extends the spur and loops it around back onto the five, and thus strictly requires the death of the one—presumably, the automatic brakes of the trolley are triggered as it runs over the one—as a means to saving the five, hence it treats the one as a mere means to saving the five. And Derek Parfit and Frances Kamm have also devised similar thought-experimental cases to the same effect, that they deem morally permissible, again apparently showing that it’s at least sometimes morally permissible to treat people merely as a means.[v]
Let’s provisionally suppose, for the purposes of advancing the argument, that Thomson’s loop variant case, and also Parfit’s and Kamm’s cases to the same effect, are as morally permissible as Bystander at the Switch. Thomson then considers the idea that what accounts for the morally relevant difference between Trolley Driver and Transplant is the fact that some “right in the cluster of rights one has in having a right-to-life”[vi] of the one is violated in Transplant,but not violated in Trolley Driver. She calls this sort of right a “stringent right,” because it is a non-interference right or liberty right not to be harmed. But this does not seem to be sufficient, since both in Trolley Driver and Bystander at the Switch it appears that some “right in the cluster of rights one has in having a right to life” of the one is in fact violated. Thomson then draws our attention to another salient difference between Trolley Driver and Bystander at the Switch on the one hand, and Transplant on the other:
In Trolley Driver and Bystander at the Switch an existing threat is deflected from five onto one, whereas in Transplant a new threat is introduced and imposed on the one.
Nevertheless, at this point in Thomson’s argument, it’s somewhat unclear just why the distinction between deflecting old threats and introducing new threats makes a genuine moral difference.
One distorting feature of both the original Trolley Driver and Transplant cases is that trolley drivers and surgeons may, by virtue of their social roles, have positive duties to provide certain goods and services for other people. Bystander at the Switch, by contrast, is morally analogous to Trolley Driver but does not include this distorting feature. Therefore, Thomson introduces a corresponding non-distorting analogue of Transplant, which she calls Fat Man:
You are standing on a footbridge over the trolley track. Beside you is a really fat man. You see the runaway trolley below you heading towards the five, and realize that if you push the fat man down onto the tracks he will stop the trolley. So you decide to push the fat man off the bridge and save the five.[vii]
Almost everyone who rationally considers these cases thinks that your choice or act in Fat Man is morally impermissible. So now the non-distorting version of The Trolley Problem is this:
What’s the morally relevant difference between Bystander at the Switch and Fat Man?
By way of a proposed solution Thomson offers the following principle, which I’ll call Thomson’s Trolley Principle, by combining her thought about rights violations with her thought about deflecting threats:
It’s morally permissible to kill one in order to save five if and only if we do so by deflecting an existing threat in such a way that the act of deflection itself violates no stringent right of the one.
Otherwise put, Thomson is saying that it’s permissible to kill a few innocent people in order to save significantly more innocent people, as long as we introduce no new threats and also do not violate any stringent rights in the means we use to get the existing threat onto the one.
I think that Thomson’s Trolley Principle clearly fails, if we continue to suppose that it’s at least sometimes permissible to treat people merely as a means. This can be seen in a case I will call, for lack of a more elegant label, the shoving-the-small-person-aside variant on Bystander at the Switch.[viii] In this variant, everything is the same as Bystander at the Switch,except that there is now one innocent small human person standing in front of the switch, and no one on the spur, and you cannot get to the switch except by shoving her aside. Sadly, the small person then falls right in front of the trolley and is killed. You do not try to push her in front of the trolley, and in fact you shove her with only the minimum amount of force necessary to clear her out of the way. Nevertheless, you do foresee that because she is so very small, it is almost inevitable that she will fall that way and be killed, yet you go ahead and shove her aside anyway, and she is killed. So you are using and treating, and in fact killing, the small person “merely as a means” to saving the five. It seems very clear that if Bystander at the Switch and the loop variant are both morally permissible, then so is the shoving-the-small-person-aside variant. The small real human person, sadly, is just an innocent casualty of that mortal threat situation. But the shoving-the-small-person-aside variant violates a stringent right of the one in the act of deflecting that threat, namely, her right-to-life, not to mention treating her merely as a means. So Thomson’s Trolley Principle doesn’t correctly isolate the morally relevant difference between Bystander at the Switch and Fat Man. But if Thomson’s Trolley Principle fails, and we’re also and even more problematically, saddled with an apparent violation of the strict impermissibility of treating people merely as a means, then what’s the correct solution to The Trolley Problem of Saving Lives? In my view, the correct solution is what I call The Morality De Re Solution.
According to The Morality De Re Solution to The Trolley Problem of Saving Lives, it’s crucial to recognize that correctly isolating the morally relevant difference between Bystander at the Switch and Fat Man, thereby showing why the former is morally permissible while the latter is morally impermissible, also rules out the apparent counterexamples to the strict impermissibility of treating people merely as a means, by showing how Thomson’s loop variant, Parfit’s and Kamm’s cases, and the shoving-the-small-person-aside variant are all in fact morally impermissible. This is for two reasons.
First, even though the human real person who is sacrificed in Bystander at the Switch is indeed being killed while saving five other people, nevertheless at the same time it’s strictly “nothing personal,” in the sense that it’s not in any way required that the actual unique life of this or that person be destroyed in order to save five other innocents. If on the contrary it were “something personal,” and the sacrificed person were to be specially selected for sacrifice, as in Fat Man, Thomson’s loop variant, Parfit’s and Kamm’s cases, and the shoving-the-small-person-aside variant, whereby the victim is as it were pulled out from behind the Rawlsian veil of ignorance[ix] and made to take a fatal hit, then that would be treating her merely as a means and like a mere thing, thereby harming her by violating her dignity as a human real person. So all those cases are morally impermissible. Again, and now put in terms of the well-known de revs. de dicto distinction in philosophical logic,[x] in Bystander at the Switch it’s not specifically required of they themselves, the sacrificed one, the unique human real person that they actually are,in this context, that they and they alone must become an innocent casualty in order to save five (de re). Rather it’s only required that someone or another, who just by sheer accident happens to be them in this context, become an innocent casualty in order to save five (de dicto). Let’s call this The Nothing-Personal Criterion, and it specifically captures the de dicto standpoint of the harming agent, or sacrificer, as they look into causally-accessible nearby possible act-worlds in order to find some way of permissibly saving the five.
Second, the one who is sacrificed in Bystander at the Switch can indeed give their explicit or implicit rational consent to being treated in this way, because it’s nothing personal and a sheer accident that they’re being sacrificed while saving five other people, hence they’re not being treated merely as a means or as a mere thing, hence it’s no violation of their human dignity. By sharp contrast, the sacrificed one in Fat Man, in Thomson’s loop variant, in Parfit’s and Kamm’s cases, and in the shoving-the-small-person-aside variant alike, cannot give their explicit or implicit rational consent to being treated as they are, precisely because it’s something personal and no sheer accident that they’re being sacrificed, in fact they themselves are being killed precisely in order to save five others, hence they’re being treated merely as means and as mere things, thereby violating their human dignity. Let’s call this The Something Personal Criterion, and it specifically captures the de re standpoint of the harmed agent, or sacrificial victim, as they look towards the oncoming threat and seeing their own deaths required in order to save the five, and justifiably refusing their actual or implicit rational consent to being treated in this way, in effect cry out with tragic moral force: “why does it have to be me?”
Taken together, The Nothing Personal Criterion and The Something Personal Criterion tell us when innocents may be morally permissibly harmed or even killed while saving many mortally endangered others, and when this is morally impermissible.
The Morality De Re Solution to The Trolley Problem of Saving Lives says that the absolute, non-denumerably infinite, intrinsic, objective value, aka dignity, of human real human persons entails that it is absolutely always morally impermissible to treat people either as mere mere means or mere things, hence without their actual or possible rational consent, even if, in some very special contexts in which “other things are not equal,” it is morally permissible to kill them while saving five others, as per Bystander at the Switch. This entailment carries over directly into the semantic content of The Nothing Personal Criterion and The Something Personal Criterion alike. The Nothing Personal Criterion is satisfied in Bystander at the Switch, thereby making it a morally permissible case, and The Something Personal Criterion is satisfied in Fat Man, in Thomson’s loop variant, in Parfit’s and Kamm’s cases, and in the shoving-the-small-person-aside variant alike, thereby making them all morally impermissible cases. By virtue of some ineluctably contingent contextual differences in proximity, distance, and causation, the Fat Man is originally placed in a position to be an innocent bystander alongside an ongoing mortal threat situation, and is morally immune to sacrifice with respect to that situation. But he has nevertheless been forced into that very situation and is being treated “as a mere trolley-stopping thing,” and not as an end-in-himself, and thereby he is also being treated without his explicit or implicit rational consent. So the trolley-stopping has become something degradingly personal for him: in that context, it has to be the Fat Man who does it. Otherwise put, in that context, he is being treated as nothing but a sufficient causal trigger for the trolley’s brakes, so that the five can be saved, and the greater good promoted. So the Fat Man and the other ones similarly placed in Thomson’s loop variant, in Parfit’s and Kamm’s cases, and in the shoving-the-small-person-aside variant, can all rationally fully expect never to receive a sufficiently justified answer to the morally tragic question: “why does it have to be me?,” and therefore they will all refuse to give their explicit or implicit rational consent to being sacrificed. By sharp contrast, the ones in Bystander at the Switch, and all relevantly similar cases, are all placed in positions to become innocent casualties of those ongoing threat-situations, for whom The Nothing Personal Criterion is satisfied, although this still happens by virtue of some ineluctably contingent contextual differences in proximity, distance, temporality, and causation. None of them can ask “why does it have to be me?” and rationally fully expect never to have a sufficiently justified answer. This is simply because it didn’t have to be them. It was nothing personal at all. They all just happened to be in the wrong place at the wrong time. So, sadly, it was just very bad moral luck.
What’s essential for Bystander at the Switch, and all relevantly similar cases, is that it’s not the intention of the harming agent or sacrificer to kill the one as “something personal” and no sheer accident. It’s not the intention of the harming agent to treat the one “as a mere trolley-stopping thing,” or as a mere causal trigger for the trolley’s brakes, in order to bring about the greater good of saving the five. If, counterfactually, there had been any other way of saving the five without sacrificing the one’s life, then the harming agent would have used that other causal means to the greater good. Hence the one can morally permissibly be sacrificed.
Obviously, everything turns here on the morally fundamental idea, captured in The Formula of Humanity as an End-in-Itself formulation of the Categorical Imperative, that it is absolutely always impermissible to treat anyone as a mere means or a mere thing, because this violates that human real person’s dignity. So more precisely then, what does it mean to treat someone “as a mere means or a mere thing”? At the very least, to treat someone as a mere means or a mere thing is to regard or treat the other person as if they had no dignity. It is therefore to regard or treat the other human real person as if whatever moral value they had was at best either (i) relatively intrinsic, such that their value as an end is solely the result of my desiring the existence of that person or some property of that person in a self-interested or aesthetically disinterested way, or else (ii) merely extrinsic, such that his or her value is merely instrumental to some further relatively intrinsic self-interested or aesthetically disinterested moral value. In turn, there are at least two different ways of having merely extrinsic or instrumental moral value. The first way involves the idea that X’s being treated as a mere means, that is, as a mere instrument or tool, also requires, or at least does not inherently rule out, the continued existence and functionality of X. I will call this re-usable merely extrinsic or instrumental moral value. And the second way involves the idea that X’s being treated as a mere means not only does not require the continued existence and functionality of X but in fact also strictly rules out this continued existence and functionality, by entailing the consumption or destruction of X. I will call this disposable merely extrinsic or instrumental moral value.
Correspondingly, there’s a sense in which being treated as a mere thing is morally even worse than being treated merely as a means. At least, being treated as a mere means allows for the possibility of having re-usable merely extrinsic moral value, hence as having at least some positive moral value, even though it’s merely extrinsic and it’s also strictly morally impermissible. By contrast, mere things have no intrinsic moral value whatsoever, whether absolute or relative. Furthermore, as disposable, mere things do not even have a re-usable merely extrinsic value. Therefore, insofar as they have merely extrinsic moral value, mere things have only negative merely extrinsic moral value. This can happen in several different ways. Mere things may be simply obstinately useless and need to be washed or swept away, like dirt or dust. Or mere things may be disgusting or noxious, and need to be exterminated, discarded, or flushed away, like pestilence, garbage, or offal. Then to treat someone as a mere thing would be to treat that real person as if she were nothing but fodder, fuel, dirt, dust, pestilence, garbage, or offal. It is misnamed “dehumanization” because it is in fact de-personalization. More specifically it is human-real-person-degrading, and if its victims do survive, it also produces in them the very worst kind of suffering, the suffering of degradation.[xi] In any case, this sort of treatment of human real persons is pretty much the most horrible thing in the world. It’s how the Nazis actually and systematically treated millions of people, and horrifyingly, of course, they are not the only ones to have done so since the mid-1930s, even on comparable scales of magnitude. To take just one example, it is clear that Americans and Japanese certainly regarded each other, and also more or less systematically treated each other, in this very way during the brutal Pacific War from 1941-1945.[xii] And the list of such abuses since that time goes on and on and on.[xiii]
In this connection, I should also note that it’s perfectly consistent with the near-satanically evil mindset of those who treat other people as mere things in either the obstinately useless or the disgusting, noxious senses, that the victims of this treatment be used up, washed or swept away, exterminated, discarded, or flushed away by so-called “humane methods” involving anesthesia, highly efficient pest-control techniques, extreme cleanliness, or “best practices”—as in the hideously sanctimonious sign over the entrance to Auschwitz, and other death-camps, Arbeit macht frei, “work makes you free,” or in the equally hideous term, “ethnic cleansing,” used by near-satanic fellow travellers of the Nazis well after 1945, and now well into the 21st century.
The other idea that needs further explication here is the notion that ineluctably contingent contextual differences in proximity, distance, and causation, as well in egocentrically-centered emotional and social relations, can partially determine the content of first-order substantive ceteris paribus objective moral principles in the contexts in which they are chosen or acted upon. These contextual differences in proximity, distance, temporality, and causation, as well in egocentrically-centered emotional and social relations, are not, in and of themselves, morally relevant differences, no matter how naturally or personally important they might otherwise be. But they do pick out and trigger morally relevant differences, that is, they pick out contextual differences that trigger the specific application of first order substantive ceteris paribus objective moral principles and the concrete determination of moral duties.
This is closely analogous to the phenomenon of “essential indexicality” in the philosophy of language and in the theory of mental content,[xiv] which I have also explicated elsewhere in terms of essentially non-conceptual content.[xv] The fact of essential indexicality is in play whenever the semantic content of a term or judgment displays inherent, systematic, and non-reducible context-sensitivity. Thus, for example, I can be the referent of “I” in some cases, and the referent of “he” or even “that” in others, and it is only contextual differences in spatial or temporal position, varying from case to case, that trigger semantic reference.
There are many interesting truths of the logic and semantics of essential indexicality. For example, necessarily, I am the only person in the world who can say “I” and mean me; and necessarily, each distinct speaker who uses “I” refers to a different individual person, namely that very speaker. Moreover, even though the terms “I,” “he,” and the demonstrative use of “that” systematically shift reference depending on the context, they mean different things at the level of what is known as the character, or general semantic function,of those expressions.[xvi] Even though, necessarily, the contextual referent of “I” is me, the character of “I” is, roughly, whoever is here and now using this token of the word “I.” Correspondingly, the character of “he” is, roughly, whoever is the male minded animal indicated by the speaker here and now, And the character of “that” is, roughly, whatever is now over there in the place indicated by the speaker. So too with proximity, distance, temporality, and causation, as well as egocentrically-centered emotional and social relations: they vary by context, but they are necessary features of every moral situation. In turn, it follows directly from these points that proximity, distance, temporality, and causation, as well as egocentrically-centered emotional and social relations, are all systematically contextually morally relevant and cannot be explained away. So let’s call this moral essential indexicality. Moral essential indexicality entails the necessary presence of essentially non-conceptual content in at least some and perhaps all objective moral principles and moral judgments.
Moral essential indexicality has one other crucial feature, namely, what’s called egocentric centering. This means that the moral relevance of spatial, temporal, and causal factors, as well as emotional and social relations, via what [Michelle] Maiese and I call “affective framing,”[xvii] in a given context or situation, always depend on establishing a subjectively experienced center or origin-point in that context/situation. Relatively to that subjectively-experienced center or origin-point (that is, a spatiotemporal and causal point of view, or affective frame), the contextually/situationally determined factors of proximity, distance, time, and causation, as well as emotional and social relations (I vs. thou, Us vs. Them, and so-on), can then all be suitably morally calibrated. So actual and possible moral contexts or situations are an example of what are called centered possible worlds. But the moral calibration can vary from context/situation to context/situation. What counts as morally relevant proximity in one context/situation (say, being two feet away from someone who is standing in front of a track-switching device, or standing next to a loved one, friend, or family member, as opposed to a stranger), may or may not count as morally relevant proximity in another. Each new context/situation needs to be morally re-calibrated.
By way of wrapping up this sub-sub-section, let’s come back now to The Trolley Problem of Saving Lives insofar as it’s treated according to The Morality De Re Solution. The crucial moral difference between Bystander at the Switch and all relevantly similar cases on the one hand, and Fat Man and all relevantly similar cases on the other hand, including Thomson’s loop variant, Parfit’s and Kamm’s cases, and the shoving-the-small-person-aside variant, is that they’re all essentially indexically determined by the sacrificed/trolley-killed one’s brute spatial distance or proximity, temporal overlap, and causal relatedness to the ongoing mortal-threat-situation, together with the sacrificer/switch-throwing bystander’s not treating anyone as a mere means or mere thing and without his actual or possible rational consent, by forcing him or her from an otherwise “causally buffered,” or relatively causally inaccessible, spacetime position into the relvant ongoing mortal-threat-situation, thereby killing him or her. This leads to the following principle:
Other things being equal, it’s morally permissible to kill one innocent human real person in order to save five other innocent human real persons if the one is already a participant in an ongoing mortal-threat-situation, provided that (i) no innocent bystander is being treated as a mere means or mere thing by being forced into that mortal-threat-situation from an otherwise causally buffered spacetime position, and (ii) no innocent bystander is being treated without her actual or possible rational consent by being forced into that mortal-threat-situation from an otherwise causally buffered spacetime position.
I’ll call this The Specific Morality De Re Trolley Principle. The Specific Morality De Re Trolley Principle adequately explains the morally relevant difference in all the pairs of specific 1-person- vs.-5-person cases covered by Thomson, Parfit, Kamm, and others, including the cases that count against Thomson’s theory—that is, including the shoving-the-small-person-aside variant—and all other relevantly similar cases. More generally however, I’m saying:
Other things being equal, it is morally permissible to kill a few innocent human real persons in order to save significantly more innocent real persons if the few are already participants in an ongoing mortal threat situation, provided that (i) no innocent bystander is being treated as a mere means or mere thing by being forced into that mortal threat situation from an otherwise causally buffered spacetime position, and (ii) no innocent bystander is being treated without her actual or possible rational consent by being forced into that mortal threat situation from an otherwise causally buffered spacetime position.
I’ll call this The Generalized Morality De Re Trolley Principle.
Here’s one possible objection to The Generalized Morality De Re Trolley Principle. It says that in treating the one in Fat Man as a human real person with dignity, we must respect the autonomous choices of the one. We might then plausibly think that it would be morally permissible and even highly morally praiseworthy, although supererogatory, for the Fat Man to choose to throw himself down onto the tracks in order to save the five. But if so, then how could the Fat Man fail to be able to give his explicit or implicit rational consent to my pushing him down onto the tracks? My reply to this objection, on behalf of broadly Kantian nonideal dignitarian moral theory, would be that it’s one thing for the Fat Man to consider giving his explicit or implicit rational consent to his jumping down onto the tracks in order to stop the trolley, and quite another thing altogether to consider giving his explicit or implicit rational consent to his being pushed down onto the tracks in order to stop the trolley. The formerhe indeed can rationally consent to, as a morally great and heroic, sinner-saint-like, although non-obligatory, autonomous act of self-sacrifice—but the latter he cannot explicitly or implicitly rationally consent to. More precisely, the Fat Man cannot explicitly or implicitly rationally consent to his being treated as a mere means or mere thing, whether by himself or by others, simply because it’s absolutely impermissible to treat anyone as a mere means and mere thing, even if the causal and moral value consequences of his jumping down onto the tracks and his being pushed down onto the tracks are exactly the same. The Fat Man’s being pushed down onto the tracks directly violates The Formula of Humanity as an End-in-Itself in its self-directed/reflexive application, and this remains the case “whatever the consequences.”
NOTES
[i] P. Foot, “The Problem of Abortion and the Doctrine of the Double Effect,” in P. Foot, Virtues and Vices (Berkeley, CA: Univ. of California Press, 1978), pp. 19-32.
[ii] See also P. Foot, “Killing and Letting Die,” in S. Cahn and P. Markie (eds.), Ethics: History, Theory, and Contemporary Issues (3rd edn.; New York: Oxford Univ. Press, 2006), pp. 783-788.
[iii] See J.J. Thomson, “Killing, Letting Die, and the Trolley Problem,” in J.M. Fischer and M. Ravizza (eds.) Ethics: Problems and Principles (New York: Harcourt, Brace, and Jovanovich, 1991), pp. 67-77; and J.J. Thomson, “The Trolley Problem,” in Fischer and M. Ravizza (eds.) Ethics: Problems and Principles, pp. 279-292.
[iv] Thomson, “The Trolley Problem,” p. 284.
[v] See, e.g., D. Parfit’s “George-the-gangster” case in his On What Matters (Oxford: Oxford Univ. Press, 2011), vol. 1, ch. 9; F. Kamm, Intricate Ethics (Oxford: Oxford Univ. Press, 2006), ch. 5; and F. Kamm, Morality, Mortality, 2 vols.(Oxford: Oxford Univ. Press, 1996), vol. II, chs. 6-7.
[vi] Thomson, “The Trolley Problem,” p. 285.
[vii] Thomson, “The Trolley Problem,” p. 288.
[viii] For an earlier version of this case—approached in three steps—see R. Hanna, “Morality De Re: Reflections on the Trolley Problem,” in J.M. Fischer and M. Ravizza (eds.), Ethics: Problems and Principles (New York: Harcourt, Brace, and Jovanovich, 1991), pp. 318-336., at pp. 325-326. When I wrote this paper (in 1988 or 1989), I didn’t recognize that this case is also an apparent counterexample to the strict impermissibility of treating people merely as a means.
[ix] Rawls’s veil of ignorance is of course simply a sub-routine within a larger rational deliberative mechanism for choosing moral principles. But to the extent that any higher-level or Kantian real human person is asked to provide actual or possible rational consent to some treatment, she puts herself behind a self-imposed veil and morally considers her own situation as if it were that of any other member of The Realm of Ends.
[x] According to the de re vs de dicto distinction, a proposition is de re if it is referentially committed to the existence of a certain unique actual object that also, as it happens, bears a certain descriptive profile, but de dicto if that proposition refers to any object whatsoever that bears that descriptive profile, whether or not such an object actually exists.
[xi] See R. Hanna, Kantian Ethics and Human Existence: A Study in Moral Philosophy (THE RATIONAL HUMAN CONDITION, Vol. 3) (New York: Nova Science, 2018). section 4.3.
[xii] See, e.g., J. Dower, War Without Mercy: Race and Power in the Pacific War (New York: Pantheon Books, 1986).
[xiii] See, e.g., Human Rights Watch, available online at URL = <http://www.hrw.org/>.
[xiv] See, e.g., J. Perry, “The Problem of the Essential Indexical,” Noûs 13 (1979): 3-21; and R. Hanna, “Direct Reference, Direct Perception, and the Cognitive Theory of Demonstratives,” Pacific Philosophical Quarterly 74 (1993): 96-117.
[xv] See, e.g., Hanna, Cognition, Content, and the A Priori: A Study in the Philosophy of Mind and Knowledge (THE RATIONAL HUMAN CONDITION, Vol.5), ch. 2.
[xvi] See, e.g., D. Kaplan, “Demonstratives: An Essay on the Logic, Metaphysics, Semantics, and Epistemology of Demonstratives and Other Indexicals” and “Afterthoughts,” both in J. Almog et al. (eds.), Themes from Kaplan (New York: Oxford Univ. Press, 1989), pp. 481-563 and 565-614.
[xvii] See R. Hanna and M. Maiese, Embodied Minds in Action (Oxford: Oxford Univ. Press, 2009), section 5.3; M. Maiese, Emotions, Embodiment, and Cognition (London: Palgrave Macmillan, 2011), esp. ch. 5.
Against Professional Philosophy is a sub-project of the online mega-project Philosophy Without Borders, which is home-based on Patreon here.
Please consider becoming a patron!
