Welcome to our Ethics review forum on L. Nandi Theunissen’s The Value of Humanity (OUP), reviewed by Kenneth Walden. Below, you’ll find a description of the book, as well as a condensed version of Kenneth’s review. Nandi’s response will appear in the comments. Please join Nandi and Kenneth in continuing the discussion!
From the book abstract:
L. Nandi Theunissen develops a non-Kantian account of the value of human beings. Against the Kantian tradition, in which humanity is absolutely valuable and unlike the value of anything else, Theunissen outlines a relational proposal according to which our value is continuous with the value of other valuable things. She takes the Socratic starting point that good is affecting, and more particularly, that good is a notion of benefit. If people are bearers of value, the proposal is that our value is no exception. Theunissen explores the possibility that our value is explained through reciprocal relations, or relations of interdependence, as when–as daughters, or teachers, or friends–we benefit others by being part or constitutive of relationships with them. She also investigates the possibility that we can be said to stand in a valuable relationship with ourselves. Ultimately, in The Value of Humanity, she proposes that people are of value because we are constituted in such a way that we can be good for ourselves in the sense that we are able to lead flourishing lives. Intuitively, a person matters because she matters to herself in a very particular sort of way; to appropriate a phrase, she is a being for whom her life can be an issue.
From Walden’s review:
The original question of moral philosophy is why I should care about anyone. A popular answer has it that I owe things to people, both myself and others, because they are valuable. Different kinds of value demand different kinds of response, and there is something about the value of humanity that forbids me from chopping up my neighbors and using their bones to make Halloween decorations. But if we want to say this, we face some obvious questions. Why is humanity valuable? And in what way is it valuable?
Nandi Theunissen’s excellent book is concerned with these questions. Her answers are situated in a more general thesis about value. There are lots of things that are good because they are of benefit—because they are good for someone or something. They have relational value. But according to some, there is also absolute value. Some things are good simpliciter, not because they benefit anything. Theunissen is interested in the hypothesis that all value is relational. […] Most of the book is aimed at overcoming two obstacles to the claim. The first obstacle concerns the value of humanity. If the relational view is correct, it can seem like we have to say that human beings are valuable just insofar as they are useful, and this seems like a poor foundation for morality […]. In response, Theunissen offers a more nuanced and attractive conception of the relational value of humanity. “People are of value because we are constituted in such a way that we are able to be good for ourselves in the sense that we are able to lead flourishing lives.” (2) Thus humanity is, like everything else, relationally valuable, but not in the exploitative way that serfs are good for their lords. […]The second obstacle is an argument purporting to show that relational value must be conditioned on absolute value lest we fall into a vicious regress. […] Theunissen […] ultimately concludes that this regress can be halted by something other than a foundation of absolute value. It can end in a small circle of reciprocal benefit.
More philosophy books should be like this. […] The theses themselves are creative yet well situated in larger traditions. The argument is detailed and rigorous without becoming tedious. […] I learned a healthy amount about an area I thought I knew well. […]
I’ll first address Theunissen’s treatment of the regress argument, which goes like this: (i) Some things are good because they can be good for something of value. (ii) The relation in (i) creates a chain of dependence between valuable things. (iii) Any such chain must terminate on pain of a vicious regress. (iv) In particular, it must terminate with an item x such that (a) some things are valuable because they are valuable for x and (b) x’s value does not depend on being good for something else. (v) Valuers meet the conditions (a) and (b). (vi) Therefore, valuers can terminate the regress on conditions of value. Theunissen […] offers a proposal on which the regress described in step (iii) is halted not by the absolute value posited in (iv), but by a pair of mutually dependent goods. The chain of dependence relations of ordinary relational goods (motor oil, aspirin, sprinkler heads) does depend on the value of valuers. But these valuers are not absolutely valuable: they are good for something else, namely their own lives, which, in turn, can be valuable for those very same valuers. So the regress terminates in a binary star system: the valuer and their life, each of which is capable of benefiting the other.
Theunissen’s proposal has the same geometry as other “virtuous circularity” approaches to stopping regresses—like coherentism about epistemic justification— and I think it invites versions of some objections to these views. […] [One] problem that arises in analogy with coherentism [is as follows]. We can imagine networks of mutually supporting beliefs that despite their internal coherence are, as a whole, completely unjustified because they are not suitably tethered to the world. Think of someone who takes the encyclopedic lore of some fantasy world to be literally true. Such a person’s beliefs seem unjustified despite their coherence. The analogous challenge for Theunissen is whether there are networks of items that stand in the value dependence relations she envisions but are worthless because they are “untethered” to genuine value. As an example, imagine a machine with many parts and accessories. The plookplook is good for the bongbong because the plookplook keeps the bongbong from drying out, and the bongbong is good for the doodlidee because the bongbong keeps the doodlidee lubricated, and the doodlidee is good for the dingalingdong because the doodlidee can help the dingalingdong stay in equilibrium. Now the dingalingdong itself is good for something else, namely the tinktinktink, which, in return, can be good for the dingalingdong. So the tinktinktink and the dingalingdong are mutually beneficial. What I have described here is a structure of value dependence that looks just like the one that Theunissen offers, only mine bottoms out in dingalingdongs and tinktinktinks instead of in valuers and their lives. […][W]hat distinguishes these things from the structure that involves valuers and their lives?
Theunissen does not address this question directly, but she discusses an adjacent one. One kind of objection to circularity relies on a “metaphysical principle”: “When an object, activity, or state of affairs, X, is instrumentally, constitutively, or in some other way good for a valuer V, and X is valuable, then the value of X must ultimately derive from the value of V.” (68) This principle tells against circularity because it requires that the value of such things be borrowed from a foundational good—an “arch lender”. On Theunissen’s view, there is no arch lender; there are just two relational goods whose benefit can be mutual. But the metaphysical principle is false, Theunissen says. […] I think [the] metaphysical claim about “borrowing” value is only one way to get at the larger issue I pressed in the previous paragraph. There can be systems of value dependence that look just like Theunissen’s but are, in whole and in part, quite worthless. Water does not magically appear when we install the pipes. So what makes the networks of value dependence involving me and you different? […]
I started this review by talking about the moral uses of claims about the value of humanity. There is something about the value of humanity that forbids me from using and abusing other people. But can Theunissen say this? […] Theunissen takes up this challenge in the final chapter. “As a valuer, a human being is a special instantiation of something that is of value because it can be good for human beings, and what is good for human beings generates reasons for human beings.” (127) What’s the practical significance of this value? “All valuable things should be responded to— acknowledged, treated, considered—as the valuable things they are. If people are of value because, as valuers, we are able to lead good lives, then we should respond to people as the value-bearers they are. In a word, we should respond to human beings as centers of a life to which they bear a special relation.” (128) And when it comes to other people, we “have a reason not to destroy, and more positively, to protect, other people’s capacity to value, and therewith, to live well.” (129) […]
Let’s unpack this argument by substituting another relational good. A can of motor oil is a special instantiation of something of value, so it generates reasons for human beings. […]All valuable things should be responded to as the valuable things they are, and motor oil is valuable because it lubricates our engines, so we should respond to motor oil as befits an engine lubricant. This gives us a reason not to destroy it, and more positively, to protect motor oil’s lubricative capacity. All these claims seem true, but they raise a worry about how much Theunissen has shown. It isn’t saying very much to say that I have a reason not to spoil the can of motor oil’s lubricative capacity. Of course, I do. But sometimes I have reasons to do the opposite. Painting my fence with motor oil would certainly spoil its ability to lubricate my engine, but with the right additives, it isn’t a bad primer, so there are times where I have more reason to spoil the lubricating potential of the oil by using it to paint my fence.
I don’t think anything Theunissen says forefends an analogous conclusion about the uses of humanity. Sure, I have a reason to protect your capacity to value, and using your bones as Halloween decorations would certainly interfere with that capacity. But I also have excellent reasons that favor doing just that[:] My house would look super spooky, and the winner of the decorations contest gets a ten-pound tub of candy corn. My concern with Theunissen’s attempt to restore the moral significance of personal value is that it shows too little. Reasons are plentiful and easily outweighed by other reasons. Common sense morality depends on showing that these reasons are especially stringent.
Theunissen has a way of doing this. The reasons mentioned above are reasons of non-destruction grounded in facts about what a valuable thing is for. Following Raz, Theunissen thinks that these reasons are non-defeasible and that flouting them is wrong. A chair is good for sitting in, and this gives me reasons not to smash it for my amusement. Not only that, it would be wrong to smash the chair for my amusement. The same goes for people: their purpose is to live a good life, and so destroying them for the sake of Halloween decorations would be wrong. Here my instincts diverge. Smashing a perfectly good chair may be wasteful, but it is only wrong under certain conditions, for instance if someone has a right to it. Likewise, destroying a person to turn them into Halloween decorations is not wrong because it is a misuse of their personhood, but because that person can make valid and binding claims on me (a power they have in virtue of their distinctive role in creating value).
In closing, I will repeat my assessment. I have enjoyed and benefitted from thinking about Theunissen’s challenging arguments. I am sure philosophers interested in value will be wrestling with them for years to come. […]
A warm thank you to Jordan MacKenzie for organizing this forum!
As I’ve had occasion to share with Kenny already, I am very grateful for his comprehending, sympathetic, and more than occasionally hilarious, review of my book. I am especially glad to have the opportunity to engage a card-carrying Kantian with skin in the game and beautifully worked out views. (I heartily recommend a forthcoming essay of Kenny’s, “Great Beyond all Comparison,” whose title states a familiar Kantian idea that is defended in a fascinating way.)
Down to business then! I’ll say a word about pookplooks and bongbongs; the borrowing conception of value; and reasons.
OF PLOOKPLOOKS AND BONGBONGS
Kenny describes a machine that is ostensibly structured by the same value dependence relations as the ones I envisage between objects and activities of value, valuers and their lives, only the machine is utterly worthless. If the evaluative relations are the same in the two cases, while the value of the relata is different, then the value cannot be a function of the relations.
My response is that the relations are not the same! The plookplook changes the state of the bongbong, as it changes the state of the doodlidee (and so on), so that there is a string of causes (with a structure of mutual causation at the end, something I leave to one side because it does not affect the present point). The “good for” in Kenny’s “the plookplook is good for the bongbong,” and for the doodlidees (etc.) that follow, means no more than changes the state of, and that is not yet an evaluatively significant relation.
In my model, by contrast, there are more than mere changes of state. The changes of state are evaluatively significant because they are relevantly related to states that are welcomed in their own right. In other words, what is instrumentally good for something depends for its evaluative significance on something that is non-instrumentally good for someone. In a toy example that illustrates a relevant sequence, the hammer is good for banging a nail in the wall, and banging a nail in the wall is good because it allows us to put the painting up, and having the painting up is good because it allows us to engage with it in appropriate ways, and that is something that is (or can be) itself good for us. Instrumental value depends on non-instrumental value (on what is good for someone in its own right), and I am also envisaging a dependence relation between non-instrumental value and final value. What I am denying is that non-instrumental or final value depends for its evaluative significance on absolute value.
In a way, the point really turns on the nature of instrumental value. If something is instrumentally valuable, it must conduce to something *valuable,* as I propose, something that is non-instrumentally good for someone. I emphasize this point because it is often overlooked, and that may be an artifact of Kant’s influential treatment of the non-moral good and the hypothetical imperative. Familiarly, Kant thinks that an outcome does not need to be good for it to be true that what conduces to it is instrumentally valuable, just as an end does not need to be valuable for it to be the case that we are under an obligation to take the relevant means. While we should not say that there is nothing rationally amiss about not forming an intention to bring about our ends, it is false that we have reason to take means to ends irrespective of their content. So too, what is conducive to something worthless or bad is not instrumentally valuable; the most we can say is that it brings it about.
This point bears on Kenny’s choice of example to illustrate a “use value” conception of the value of humanity against which my own relational proposal is being brought into positive relief. Kenny canvasses a view on which people are valuable in the “exploitative way serfs are good for their lords.” While recognizing the complexities in the example, I would not even be prepared to count this as a case of use value because I am not prepared to accept the implication that exploiting the serfs is good for the lords. In my view, a clearer example of a use value conception of human beings is the one envisaged by Sarah Buss when she proposes that we human beings are instrumentally valuable because we are able to pay homage to other valuable things, like works of art and the natural world. (See Buss’s “The Value of Humanity.”)
OF BORROWING & THE METAPHYSICAL PRINCIPLE
As Kenny brings out, questions about value dependence are at the heart of my rejection of a popular regress argument for the absolute value of valuers. I see proponents of the regress argument as mistakenly buying into a “borrowing” conception of value, one that is expressed in the principle that Kenny mentions. Very roughly, they suppose that for value to “get in” to a series of relations, it must “come from” something that does not “get it” from anywhere else, where valuers fit the bill. In his review, Kenny is probing my grounds for rejecting the borrowing conception, but he is also wondering whether it is really at the heart of things.
Why do I reject the borrowing model? To put the point somewhat differently from the way I put it in the book, while instrumental value depends for its evaluative significance on non-instrumental value, I think it is a mistake to extend the model of dependence further. It is false that when something is non-instrumentally good for a valuer, its value derives from the value of the valuer. That is to misunderstand the nature of non-instrumental relational value.
Let me say something about non-instrumental relational value as I understand it. When something is non-instrumentally good for someone, it is directly good for them, or good for them in its own right. Engaging with culture and the arts (in a common example) is good for people independently of whether it makes them interesting conversationalists—though that may be a welcome ramification. It is good for people in its own right in the sense that it necessarily involves the use of their imaginative, emotional, and intellectual powers, powers the exercise of which is a constituent of their good. The value *is* the alteration or transformation in the person that is marked when we say she is enriched, nourished, enlivened, moved, uplifted, consoled (etc.) by something. The relevant form of value is a *relation,* in a simple case, a dyad, between an object, a state, or an activity, and a person. Rather than being derived from something that is independently valuable, the value of what is non-instrumentally good for someone is explained by the valuable *relation* itself. The value *is* the enlargement of imagination, consciousness, and so on, that is a function of appropriate engagement. While there is a clear sense in which what is instrumentally valuable depends on something independently good, there is no comparable structure of dependence in the non-instrumental case. As I see it, the mistake made by proponents of the regress argument is to assume that different evaluative concepts have analogous structures of dependence.
So, I reject the borrowing conception because it misunderstands the nature of non-instrumental value. When we appreciate the truly relational character of non-instrumental value, we see that we do not need to “get value in” by having it “come from” something that is self-standingly valuable.
A confession. In the book I fall short of seeing the full implications of this point about non-instrumental value. That is to say, I partly miss my own point about the perils of borrowing. For in the book, I am prepared to accept that valuers must be of value in some way for other things to be valuable, I simply deny that they must be *absolutely* valuable, developing the suggestion that they may be *relationally* so. In new work (“Explaining the Value of Human Beings”), I reject even this. Valuers do not need to be of value for other things to be of value. A datum about our value does not fall out of the structure of evaluative explanation. To return to the example I used earlier, engaging with works of art can be non-instrumentally good for us—it can be by itself enriching for us—and that (suitably filled out) is a complete explanation of its value. At least, we do not need to invoke our being of value to explain the value of the work or the value of engaging with it.
REASONS
Kenny raises a natural question about whether my relational account of the value of humanity can support the robust reasons to which human beings give rise. The question comes up because I am giving up the Kantian idea that, to put it in contemporary terminology, the reasons that people generate are particularly stringent because the value of people is special: unlike the value of anything else. Kenny himself helpfully unpacks my response to this question, and he ends by marking a contrast with his own Kantian approach.
Let me simply add to what Kenny says here by providing a little context. I take a value-based approach to practical reasons, and that I see human beings as one among other values that ground practical concern. As I bring out in Chapter 1 of the book, I am here following Joseph Raz who situates ethical responses to human beings in a broader account of practical concern (though I radicalize Raz’s proposal in ways he would most certainly reject!) As I will put it, the basic response that is licensed by the good or the valuable is to see to it that it can serve its function as the valuable thing it is. As Kenny explains, these reasons are always stringent and flouting them is wrong. So there is a kind of formal parity for all that is good or valuable, while differences in the content of the reasons to which various values give rise turn on differences in the character of the values.
Kenny ends with a contrast between his Kantian approach to the value of humanity and my own. For Kenny, human beings are creators of value and for that reason especially valuable. For me, human beings are certainly agents, or as I put the emphasis, *valuers,* and this makes us valuable because it makes us able to live flourishing lives. The contrast that is marked here is between constructivism and relational realism, and this brings us into the heart of controversies in meta-ethics. Making the value of humanity a topic for meta-ethics is of course one of Christine Korsgaard’s key innovations and legacies in ethics, and we are all indebted to her.
Thanks to Nandi for recommending me as a discussion contributor, and to the organizers for this wonderful chance to do some value theory from the comfort of home.
The gist of my comment will repackage in a different key, I think, one of Kenny’s complaints: namely, that “Water [i.e, value] does not magically appear when we install the pipes [i.e., valuers for whom their life is an issue]” (810).
I take it that what motivates thinking that relational value must have a “source” in–that it must “come from”, “be borrowed or lended” by–something non-relationally valuable is a need *to explain* why there is relational value at all. Weight lifting is good for you, my doctor, Galen, who has been reading Nandi’s work, says to me. “Why”, I ask, only to be met with,
Galen: “Well, weight lifting is good for you because it helps you stay healthy, keeping your muscles strong and protecting you from the chance of developing heart disease and other such morbidities.”
Me: “Right, but why does being healthy, keeping my muscles strong, and protecting myself from the morbidities, etc, matter?”
Galen: “All of this matters because being healthy and well lets you exercise all those skills and cognitive and affective capacities and goods which only a healthy body enjoys. All of that matters because it is good for you, as the kind of valuer that you are.”
So far, we have an explanation of the instrumental relational value of exercise in terms of the instrumental relational value of health, which is explained by the non-instrumental relational value of exercising one’s skills and capacities. Schematically: Some instrumental relational value, R1, is explained by some other instrumental relational value, R2, which is explained by some final relational value(s), F.
But suppose I, gripped by philosophical curiosity, now ask: “But why does it matter that there are things good for me, as the kind of valuer I am? That is, what explains the final relational value of exercising my skills and capacities, etc–what explains F?”
The borrowing conception of value has an explanation, here. They explain F in terms of absolute value, and (for those like Korsgaard who give the valuer-dependence answer to Euthyphro) absolute value in terms of asymmetric explanatory powers of being a valuer. But it is hard for me to see how someone like Nandi, who thinks that the domain of value is relational “all the way down”, as it were, can explain F at all. Whereas a “borrower” about value can explain why pipes come filled with water, Nandi, it seems to me, is left with some variation of the thought that water/F is a dyadic relation, so where there are pipes/valuers for whom their life is a matter of concern there is water/F. And *that* it seems to me is not an explanation at all, as much as it is a reframing of the very thing that relational final value is.
And, so, the borrowing conception has an explanatory advantage, whereas rejecting that model of value, in the way Nandi does, seems to forego explaining F. That strikes me as a cost for Nandi’s and a reason to favor the borrowing model.
There are ways out of this, I think: perhaps by compartmentalizing the “what explains F” question as concerning realism/anti-realism, of objectivism about relational value vs subjectivism, as an orthogonal issue. But as I have tried to put things, the “what explains F” question is part of what motivates someone to become a borrower about value. And, so, if anything, perhaps telling against the borrowing model will call for a stance within this foundational debate.
Curious to hear your thoughts!
Grr, typos and leaving some stuff out!
Last paragraph should read:
There are ways out of this, I think: perhaps by compartmentalizing the “what explains F” question as concerning realism/anti-realism, objectivism about relational value vs subjectivism, which is an orthogonal issue. But as I have tried to put things, the “what explains F” question is part of what motivates someone to become a borrower about value. And, so, if anything, perhaps telling against the borrowing model will call for a stance within more foundational, meta-ethical debate.
Thank you, Abdul! And I am very glad you bring in Kenny’s nice phrase about the pipes — ha!
Exercising our capacities to value in the appropriate ways constitutes living well on my account, that is, it constitutes a good human life. You are asking, what is the explanation of the value of a good human life. As you bring out, my answer is that it is of value because (at least in the first place, but I think the first place is important) it is good for the person whose life it is. Now you imagine someone in a rather striking state of mind, someone who is not seeing the relevance of the state of his own life to himself. He asks, what is living well to me?
I think we should pause to ask how much force this question has. The ancient Greeks take it for granted that people care about their lives and that is why the good human life is the starting point in ethics for them. That it would affect the course of one’s life is thought to be a natural stopping place for a why question. (On these issues, I heartily recommend Katja Vogt’s _Desiring the Good_).
I am very sympathetic to this line of thought, though it is not actually the one I take in the book. There are philosophers for whom there is no basis with which to say that the person who does not care about the state of their life is making a mistake. If someone does not care about their own life, then they may not have reason to keep living. This is not the view that I develop in the book (so that the phrase you quote, beings for whom our life can be an issue, is not to be understood in an internalist way). I develop the view that living well is valuable, and that its value generates reasons independently of whether a person cares about their own life or not. My explanation of the value of living well is relational, but the relation is one of benefit understood in a realist way. Living well is good for human beings, and it generates reasons to live well that do not bottom out in our attitudes.
I hope this is responsive! Thanks again for your question.
Thanks for this response! I think it is fairly clarificatory of the view and its commitments, and I find your take on the “what is the value of F—the good human life” quite intriguing, especially because I am working–very early stages!–on whether it is intelligible to ask *why value my life*, and, if it is, what follows for ethical theory.
Anyhow, I think where we could do with some more clarity, which would also probably get to the heart of my comment, is your phrase, “the relation is one of benefit understood in a realist way”. Are you committed, here, to explaining what I called F, which you take to be the final relational value of a good life, realistically, such that the answer to the question what explains the relational value of a good life would be the realist’s/objectivist’s gambit: Such a life instantiates the good-for property, which, whatever that property’s nature (e.g., whatever we want to say about whether it reduces to something science-friendly or not ), does not need to be explained in terms of something else, particularly absolute value, of humanity or otherwise, in the way I suggested was an explanatory advantage held by the “borrowers”.
Put perhaps in another way: Is it your view that realism/Euthyphro objectivism about the good-for suffices as explanation-enough to the why-question I was highlighting?
Thanks again for the chance to engage your rich views!
Thanks to Nandi for a terrific book and her patient, probing, and stimulating comments on my review. I’ll have a few things to say. This first bit is about her first down to business post.
If I understand correctly, Nandi’s response to my silly machine is its parts stand not only in relational but also instrumental value relationships. But this makes it a non sequitur, since she agrees that instrumental value needs to be grounded in non-instrumental value. Fair enough. But this makes me want to ask about the relationship between instrumental and relational value. (I apologize if these points are addressed more fully in the book, reflect egregious ignorance of familiar literature, etc.) If I recall correctly, Nandi understands relational value in terms of benefit. Something is valuable for me if it benefits me. And something benefits me if I am better off with it than without it. This seems to make relational value very close to instrumental value, since it seems like any benefit can be understood as a *means* to my improved condition.
But wait, you say! That last inference doesn’t go through. There are standard exceptions involving when something is partly constitutive of the improved condition. A functioning digestive system redounds to my health but is not a means to health—it’s part of what being healthy consists in. All true, but it leads me to two questions about the approach.
1. Do you think there are exceptions to coincidence between benefit and instrumental value that I am trying to sneak into the debate beyond cases involving partial constitution?
2. I think we could imagine of a machine like the one I tried to describe where the parts are not a means to the benefit of other parts but partly constitutive of their well-functioning. What do you think? Would that raise the problems I suggested?
(There may be other kinds of examples, but I’m not sure Nandi brooks them. Suppose I want the Red Sox to beat the Yankees. We might say that a Red Sox victory is good for me simply because I value it, not because of the ends it begets. But it’s also not a *benefit* to me, so I don’t think Nandi wants say this. But maybe I’m wrong.)
It sounds like a great research project! I look forward to an occasion to hear more about it.
And the answer to your first question is, yes. You put it most helpfully. The life instantiates the good for property and good for does not need to be explained in terms of absolute value. Not only do we not need absolute value, I think an appeal to absolute value actually obscures the nature of value. And I do not think it is explicable in other terms either so that the relation _being good for_ is a primitive. I think we can elucidate the relation, for example, I think it is essentially affective in some of the ways I try to bring out in my remarks about the nature of non-instrumental value above. And while we can say that what is good for human beings is advantageous, beneficial, fitting, salubrious, enriching, and suitable, and we can say that what is good for human beings makes them flourish, these are ways of saying the same thing.
My earlier response was to Abdul.
This is a response to Kenny.
Kenny — you put your finger on some crucial issues. I think many people have difficulty finding room, conceptually speaking, for the category of what is non-instrumentally good for people. I think one reason for this is that it is common for people to define instrumental value as the value something has because of its effects. Then if one is presented with the notion of benefit which is essentially affective, not in the sense that it relates to states of feeling or emotion, but in the sense that it exerts an influence or has an effect, it is thought that we must have to do with instrumental value. I think the definition of instrumental value is misleading. Something can be valuable because of its effects, though it is not instrumentally valuable.
I find it helpful to draw the distinction between what is instrumentally good for people and what is non-instrumentally good for people in terms of what is indirectly good for people and what is directly good for people. What is indirectly good for people is what conduces to something that is (or is connected to something else that is) directly good for them. What is directly good for people is what is good for them by itself or in its own right. Engaging in philosophy (in appropriate ways – etc.), engaging with culture and the arts, having a meaningful conversation, sitting in meditation — these are all examples of activities that are directly good for people. These are activities that themselves change the state of our “soul” or mind in positive ways. We can see these activities as partially constitutive of the larger whole that is living well. And seeing it in that way can be illuminating. But I think that this move can obscure the category of the directly beneficial. This category is sometimes referred to as the “intrinsic non-moral good for people.” I think those terms “intrinsic” and “non-moral” make a lot of trouble, so I don’t use them. But this is a familiar phrase for a very important category in value theory and ethics.
I think you are completely right to treat health as a non-instrumental good. Or perhaps we might see health as both instrumentally and non-instrumentally good for human beings. It both allows us to engage in other activities that are good for us, and it is itself a dimension of our flourishing (for we mortal beings).
Regarding your machine. I think the parts of the machine can partially constitute its well-functioning. But a well-functioning machine is not enough for value to come on the scene. We need to know what the machine is good for, so we need to know why its well-functioning is important and worth seeing continue. And presumably that leads us to ask what it does for us.
I hope this is responsive! Thank you again for your thoughtful engagement!
Thanks to Nandi for the book, to Kenny for the review, and to all for the discussion thus far, it has been very interesting and illuminating.
I have two questions for Nandi, both continuing (and I hope not merely repeating) points made in the conversation in the comments thus far; one follows up on some questions/ideas from Kenny concerning how value dependence relations can terminate, and the other is about the reasons generated by human (and other) value. First to value dependence relations: as I understood some of the discussion above, Nandi’s view is that while human beings are valuable in the sense that we can be good for each other and for our own lives, appealing to this value is not necessary to explain the value of those things which are good for human beings. When we have a chain of value dependence (various instrumental goods that are good for other instrumental goods) all we need to end the chain and explain the value of the instrumental goods is a non-instrumental good (e.g. something directly good for human beings). But then, in response to Kenny’s example of the machine whose parts partly constitute its well-functioning, Nandi also mentioned that this would not be enough for value to come on the scene – we need to know what the machine is good for (the idea being, I think, that the parts would only be valuable if the machine was good for something further, for example if it was an instrument to something directly good for human beings).
I am still not quite clear on exactly why there is a difference between the case of the human being and the case of the machine. In the human case, being non-instrumentally/directly good for us is enough for value to come on to the scene, regardless of our value (regardless of whether we are good for anything). Now I would have thought that in Kenny’s example, the parts are not instrumentally but are rather non-instrumentally or directly good for the machine. That would mean that in the machine case, too, the value dependence chain ends with something non-instrumentally good for something (the machine) – but in this case, unlike in the case of human beings, this explanation is insufficient. So my question is, what explains the difference between the two cases? Why does value come on the scene when a value dependence chain ends with something non-instrumentally good for a human, but not when it ends with something non-instrumentally good for the machine?
I see at least two possibilities here: perhaps the parts are not directly good for the machine, even though they partly constitute its well-functioning (in which case the value dependence chain does not end with something non-instrumentally or directly good), but then it is not quite clear to me why they are not. One way to go about this would be to say that in order to truly be a beneficiary of direct or non-instrumental goods, something needs to meet certain conditions, conditions which the human being meets but the machine does not. The second route I see would be to say that the parts are directly good for the machine, but to say further that appealing to a direct benefit is not in itself sufficient for ending value dependence chains. Either way, perhaps here is where we bring in something special about organisms and benefitting organisms (something which Kenny in his review mentions Nandi suggesting).
My second question concerns practical reasons: Nandi mentioned in her comment about reasons that she opts for a view according to which (1) the good or valuable licenses the basic response of seeing to it that the valuable thing is able to serve its function, and (2) such reasons are stringent and flouting them is wrong. As I understand the view, this means that the beneficial licenses the basic response of seeing to it that the beneficial is able to benefit. My question is, why prefer this sort of view to a view according to which (1*) the good or valuable licenses the basic response of seeing to it that the valuable thing does serve its function (i.e., seeing to it that the beneficial actually benefit) and (2*) such reasons are proportional in strength somehow to the degree/significance of the benefit? I worry that the first view, in explaining the stringency of our reasons to treat human beings a certain way (and in explaining that mistreating them is wrong), might explain too much by claiming that other reasons are more stringent than they perhaps are. One of the typical desiderata of an account of human value is that such an account explain why the reasons we have for treating human beings well are stringent, but I take it that a closely related desiderata is that such an account explain why these reasons are also much stronger/more significant than most other practical reasons. But if everything that is valuable generates a reason to see to it that it can function as the valuable thing that it is, and if these reasons are always stringent and flouting them always wrong, then it seems as if accounting for that asymmetry will be difficult.
Thanks again to Nandi and Kenny for the excellent book and review, respectively.
Aaron —
Thank you for your engaging questions! (And it is good to “see” you here!)
Of dependence relations:
The machine is not a beneficiary of non-instrumental goods because a machine is designed by us to fulfill certain functions, so that we are the beneficiaries. And for the machine to be worth having, the purpose it serves for us must be a good one. This is a point about functional explanation. But I do fully accept your suggestion that “in order to truly be a beneficiary of direct or non-instrumental goods, something needs to meet certain conditions, conditions which the human being meets but the machine does not.” Here I would bring in the line from the book abstract that Abdul referred to in an earlier post: we are beings for whom our lives can be an issue. The difference between a machine and a valuer is that, as a matter of fact, a valuer has a perspective on the world from which it matters how things are going for it.
This point is beautifully elaborated and explained by Andrea Sangiovanni in his forthcoming essay, “Are We of Equal Moral Worth?” As I will adapt Andrea’s point, valuers are (ultimate) beneficiaries in virtue of the evaluative fact that valuers have a distinctive perspective on their lives — on what is going well or badly for them. (I put in the term “ultimate” to take account of what Andrea says about plants. In short, he says that plants have a flourishing, things can be good or bad for them, but they lack value in their own right. What I understand him to mean is that the value of the flourishing of plants is further explained in terms of the relations they bear to valuers, human and animal.)
Thanks Nandi! Good to “see” you here too. That was very helpful – the comparison with plants was particularly clarifying. I had been assuming that things could be directly good for plants, and I had been assuming that they could be so in a relevantly similar way to things being directly good for human beings, such that anytime a dependence chain ended with something being directly good for something (plant or person) that constituted a complete explanation. I now see that the explanation is only complete when the beneficiary is a valuer, a being who as you say “has a perspective on the world from which it matters how things go for it” (so I should either not be thinking of those things which are good for plants as directly good for them, or I should be thinking of being directly good for a person as dissimilar in important respects from being directly good for a plant).
Aaron — (sorry for the hiatus! I have been coming and going today.)
Of reasons:
You are asking about the difference between (1) the reason not to destroy, and more positively, to enable, something so that it _can_ fulfill its function, and (1*) the reason to respond to something so that it _does_ fulfill its function. I emphasize the former as the basic (in the sense of preliminary, but also in the sense of categorical) reason to which objects and activities of value give rise, and in doing so I am broadly following Raz’s distinction between reasons to respect and reasons to engage (in eg. _Value, Respect and Attachment_). The thought is that, generally speaking, everyone has reason not to spoil the forum. Not spoiling makes it possible for the forum to be engaging for people (and thereby to be beneficial). But not everyone must engage with the forum (and be benefited by it).
I accept some version of (2*) and you bring out some very important distinctions and qualifications. The Kantian notion of duty collapses various ideas that we might want to distinguish. Kantian duty is categorical in the sense that it does not depend on desire, and so it has a universal scope; it is inviolable; it is of supreme importance; and it is morally significant. The basic reasons described in (1) are categorical in the sense that they do not depend on desire and other subjective considerations, and accordingly, they apply to everyone. It is for this reason that flouting them is “wrong” (even as calling them “moral reasons” would be confusing because they are generated by things that are not usually thought to be morally significant, like the chair in Kenny’s example). The reasons are not inviolable, so while values in general give rise to a reason not to destroy and to enable, that reason might be defeated by other considerations in a particular case. In an example I use in the book, a general reason not to harm sentient beings may not hold in a case where rats, lets’s say, are in plague proportions. What is more, the reasons may be quite trivial or unimportant.
I think differences in violability and significance will turn on some of the distinctions I was drawing in my response to your earlier question (in the post above), when I was discussing machines, plants, and valuers. The fact that a valuer has a point of view on her life and how things are going for her makes a difference to the reasons to which she gives rise along these dimensions. I say valuer here, and in the book I allow that other animals can be valuers (valuers in the sense of having final ends in the way I discuss in ch. 4). So I do not make the further claim that the reasons human beings give rise to are necessarily more important than, or trump, those of other valuing animals. Korsgaard’s _Fellow Creatures_ is relevant here.
Again, I hope this is responsive. These are helpful, clarifying, and important questions!
Right — I do think that a plant is a direct beneficiary in a way that the machine in the example is not. But I think the complete explanation of the value of what is directly or non-instrumentally good for the plant involves a valuer, and that is why I have the valuer figuring as the final node in the chain of dependence. I like this formulation “being directly good for a person [is] dissimilar in important respects from being directly good for a plant.”
Thanks, Aaron!