Show Less
Open access

Essays on Values and Practical Rationality

Ethical and Aesthetical Dimensions


Edited By António Marques and João Sàágua

The essays presented here are the outcome of research carried out by members of IFILNOVA (Institute for Philosophy of New University of Lisbon) in 2016.

The IFILNOVA Permanent Seminar seeks to show how values are relevant to humans (both socially and individually). This seminar is the ‘place’ where different research will converge towards a unified viewpoint. This includes the discussion of the following questions: What is the philosophical contribution to current affairs and decisions that depend crucially on values? Can philosophy make a difference, namely by bringing practical reason to bear on these affairs and decision? And how to do it? Which are our scientific ‘allies’ in this enterprise; psychology, communication sciences, even sociology and history?

This volume shows the connection between practical rationality and values and covers the dimensions ethics, aesthetics and politics.

Show Summary Details
Open access

Understanding Value ‘Relativity’ (Jamie Buckland)

← 12 | 13 →

Understanding Value ‘Relativity’


In this paper I want to question the feasibility of the distinction between agent-neutral and agent-relative values, and explore an alternative method of understanding value in terms of Alan Thomas’s distinction between non-relational and relational intrinsic value. I shall then briefly suggest that these observations can lend support to value primacy: the idea that talk of value cannot be reduced to talk of the reason-giving force of other (perhaps non-evaluative) properties.1 I shall begin by explaining how Thomas Nagel drew the distinction between the agent-neutral and agent-relative in terms of both reasons and values.

1.  Agent-Relativity and Agent-Neutrality: A Value-Based Theory of Reasons

The clearest statement of the distinction between the agent-neutral and agent-relative can be found in Nagel’s account of reasons for action in The View from Nowhere:

If a reason can be given a general form which does not include an essential reference to the person who has it, it is an agent-neutral reason. For example, if it is a reason for anyone to do or want something that it would reduce the amount of wretchedness in the world, then that is a neutral reason. If on the other hand the general form of the reason does include an essential reference to the person who has it, it is an agent-relative reason. For example, if it is a reason for anyone to do or want something that it would be in his interest, then that is a relative reason. (Nagel 1986: 153–154) ← 13 | 14 →

In a nutshell, agent-neutral reasons are said to be reasons for anyone and everyone, for instance, the general reasons there are for us all to donate money to charity or look after the environment. Such reasons are said to ‘depend on what everyone ought to value independently of its relation to oneself’ (Nagel 1991). If some act, event, or circumstance is said to have agent-neutral value, then anyone has reason to promote its occurrence, or at least desire that it happen. Agent-relative reasons, on the other hand, are said to be reasons only for particular individuals or groups; the special reasons there are for each of us to look after our own interests, or the interests of our family and friends, etc. If some act, event, or circumstance is said to have agent-relative value, then only some particular agent or groups of agents have reason to want and pursue it because it is related to them in the right way – it is valuable for that individual. For instance, everyone’s life is considered to possess both agent-relative and agent-neutral value. We each have agent-relative reasons to care about our own lives (because our lives are valuable for us) as well as having agent-neutral reasons to care about the lives of others (because the lives of all individuals have intrinsic or absolute value). In this sense one-and-the-same act, event, or circumstance can often be understood as having both agent-relative value for someone in particular, or as having agent-neutral value simpliciter, or absolutely.

2.  Understanding Agent-Neutral Value

Christine Korsgaard noted there is some ambiguity surrounding this notion of agent-neutral value or good-absolutely. In her examination of Nagel’s distinction, she considers two approaches to understanding agent-neutral value. The first approach she calls ‘objective realism’:

An agent-neutral value might be a value that is not relative to what agents actually value. According to this interpretation, the goodness of, say, my happiness, has what G. E. Moore called an intrinsic value, a property that is independent either of my interest in promoting it or yours. It provides a reason for both of us the way the sun provides light for both of us: because it’s out there, shining down. And ← 14 | 15 → just as the sun would exist in a world devoid of creatures who see and respond to light, so values would exist in a world devoid of creatures who see and respond to reasons. (Korsgaard 1996: 278)

On this Platonic or Moorean view, one comes to value something when one perceives or discovers its value empirically. Consequently, agent-neutral values are understood as fundamental; agent-relative values are generated by, or derived from agent-neutral ones (Nagel 1970: vii; Korsgaard 1996: 278). On a second approach, agent-neutral value is understood in terms of its intersubjectivity:

[A]gent-neutrality does not mean independence of agents as such, but neutral with respect to the individual identities of agents. On this reading values are intersubjective; they exist for all rational agents, but would not exist in a world without them. (Korsgaard 1996: 278)

Given this intersubjective account, agent-relative value is fundamental; agent-neutral values are constructed out of agent-relative values when agents recognise and come to share each other’s ends.

Before considering where Nagel’s position fits in with these approaches, it is worth looking at G. E. Moore’s form of objective realism in some more detail, particularly his argument against what is now understood as agent-relative value. Moore infamously rejects the idea of agent-relative value qua some sui generis evaluative property:

What, then, is meant by ‘my own good’? In what sense can a thing be good for me? It is obvious, if we reflect, that the only thing which can belong to me, which can be mine, is something which is good, and not the fact that it is good. When, therefore, I talk of anything I get as ‘my own good,’ I must mean either that the thing I get is good, or that my possessing it is good. In both cases it is only the thing or the possession of it which is mine, and not the goodness of that thing or that possession. […] In short, when I talk of a thing as ‘my own good’ all that I can mean is that something which will be exclusively mine, as my own pleasure is mine (whatever be the various senses of this relation denoted by ‘possession’), is also good absolutely; or rather that my possession of it is good absolutely. (Moore 1903: 99)

Now, some take this position to amount to the claim that talk of what is good-for or valuable-for some particular agent or another is, itself, ← 15 | 16 → nonsensical (Kraut 2007: 70). However, the more plausible interpretation of Moore’s argument is that the idea of some kind of sui generis agent-relative evaluative property is nonsensical, for goodness, itself, is not a relational property. Consequently, talk of agent-relative value, or ‘good-for x’ can only ever be talk of good-absolutely which stands in a certain relation to a particular agent. Goodness or value is a metaphysically simple, non-natural property which belongs to things which are good.

Moore’s view aside, Korsgaard’s favoured position is intersubjectivism. Nevertheless, she insists there is some ambiguity as to whether Nagel shares this position or is, rather, committed to some kind of objective realism (Korsgaard 1996: 279–282). However, when understood correctly, it is clear that Nagel’s position is more akin to the intersubjectivist position, whereby values are understood as objective in the sense that they are normatively real.

As we saw in section 1, Nagel advances a value-based theory of reasons where the presence of a reason for action is explained by the relation in which an action stands to some valuable state of affairs, yet for Nagel the idea of some kind of Platonic or Moorean objective realism is misguided. For Nagel, evaluative judgements depend on judgements about the presence of particular reason-constitutive considerations, and the objectivity and reality of value is fundamentally grounded in the conception of a practical agent. Thus, values arise from and are analysed in terms of the conception of a good reason for action:

Values are judgments from a standpoint external to ourselves about how to be and how to live. Because they are accepted from an impersonal standpoint, they apply not only to the point of view of the particular person I happen to be, but generally. They tell me how I should live because they tell me how anyone should live. (Nagel 1986: 135)

What appears odd, however, is that Nagel’s objection to reductive accounts of normativity more generally is stated explicitly in terms of values rather than reasons:

If values are objective, they must be so in their own right, and not through reducibility to some other kind of objective fact. They have to be objective values, not objective anything else. (Nagel 1986: 139) ← 16 | 17 →

Nevertheless, for Nagel this does not amount to a reduction of values to reasons; the relationship he envisages between reasons and value is one of direct correspondence or equivalence; there is an asymmetrical dependence relation between value and the grounding of reasons in the sense that valuable acts, events, or states of affairs are said to provide agents with a reason to promote them. It is in this sense that values are said to be normatively objective, for to objectify them under a non-normative criterion would be to reduce them to some other objective fact, for instance, a psychological fact such as a desire which is said to explain the presence of a reason.2 Consequently:

Normative realism is the view that propositions can be true or false independently of how things appear to us, and that we can hope to discover the truth by transcending the appearances and subjecting them to critical assessment. What we aim to discover is not an aspect of the external world, called value, but rather just the truth about what we and others should do and want. (Nagel 1986: 139 [emphasis added])

Agents, then, do not discover reasons for action that exist independently of their pre-existing subjective motivational states and interests, but rather, taking up an objective standpoint allows objectivity to bear on their will, which can alter and constrain those motives that are already present. These considerations are, as such, the seeds of our moral theorizing. Objectivity and realism in ethics, then, is not analogous to the sense found in theoretical reasoning or empirical Moorean realism. Nagel offers a practical account of objectivity and a normative realism disanalogous to theoretical reasoning in the sense that we do not arrive at new beliefs that include ourselves as components; rather we arrive at an extended set of values and normative judgements from a centreless and impersonal standpoint of objectivity.3 ← 17 | 18 →

3.  Three Kinds of Agent-Relativity?

Nagel’s aim in The Possibility of Altruism was to show that all personal agent-relative reasons had to have an agent-neutral counterpart4 (Nagel 1970: vii). Nevertheless, he eventually succumbed to the idea that there are agent-relative reasons (and values) that are too idiosyncratic to be subsumed under a suitable agent-neutral counterpart, so he allowed for the existence of certain agent-relative reasons (and values) so long as they were tolerable from an agent-neutral perspective. These agent-relative reasons were said to fall into three categories. Firstly: reasons of special obligation. These are reasons stemming from the value of the personal/familial relationships we have with others, for instance, the agent-relative reasons I have to look after my own family and friends. Secondly: reasons of autonomy. These reasons are grounded in the value of an agent’s personal projects or goals, for instance, my ambition to climb Kilimanjaro. And, thirdly: reasons of deontology. These are reasons stemming from an agent’s special concern with his or her own actions, for instance, reasons not to kill innocent people, tell lies, etc.

Now, although orthodox, the idea that deontological reasons are agent-relative is somewhat controversial (cf. McNaughton & Rawling 1991; Portmore 2013). Remember, the idea is that agent-relative reasons track agent-relative values, yet it is far from clear how the disvalue of, say, killing an innocent person is ‘agent-relative’ in the required ← 18 | 19 → sense. Granted, there is a sense in which not killing an innocent person is good-for-me, but it is simply odd to say that this is what makes the killing of innocents bad. Nevertheless, it certainly seems fitting that the value of one’s personal/familial relationships, personal projects, or goals can fittingly be described as ‘agent-relative’, i.e. ‘good-for’ particular individuals. Though, again, as Korsgaard has noted, this seems to mischaracterise the phenomena.

4.  Korsgaard on Ambitions and Special Obligations

Korsgaard refers to reasons of autonomy as ‘ambitions’. For Nagel, the agent-relative value of one’s ambitions provide agents with agent-relative reasons to do things, yet the normative force of these reasons does not extend beyond the agent for whom they are ambitions: no one else has a reason to, say, help me climb Kilimanjaro, and I have no reason to help you achieve your ambitions (Nagel 1986). Korsgaard rightly insists this is mistaken:

Suppose it is my ambition to write a book about Kant’s ethics that will be required reading in all ethics classes … Following Nagel’s analysis, we will say that this ambition is agent-relative … But this way of describing the situation implies a strange description of my own attitude. It suggests that my desire to have my book required is a product of raw vanity, and that if I want to write a good book, this is merely a means to getting it required [in all ethics classes.] … So the structure of this ambition is not:

(1) I want my book to be required reading (where that is an agent- relative end);

(2) therefore: I shall write a good book (as a means to that end); but rather:

(1) Someone should write a book on Kant good enough that it will be required reading (where that is an agent-neutral end);

(2) I want to be that someone (agent-relative motive). (Korsgaard 1993: 287–288)

The same observation extends to reasons of special obligation: ← 19 | 20 →

The structure of reasons arising from love is similar to that of reasons of ambition. I think that someone should make my darling happy, and I very much want to be that someone. And others may have good reasons to encourage me in this. But if I try to prevent someone else from making my darling happy, or if I suppose that my darling’s happiness has no value unless produced by me, that is no longer an expression of love. Again, it is a very familiar perversion of it. (Korsgaard 1991: 211)

The structure of ambitions and obligations is more complex than Nagel allows, then. Rather than a self-interested desire for an object, or a desire to realize something you think is good-for-you, an ambition is the desire to stand in a special relationship to what is good, agent-neutrally (intersubjectively). It may well be some type of ‘agent-relative’ element that has the motivating force, i.e. it is my ambition to climb Kilimanjaro, and my daughter whose care I must prioritize, but this desire is not the source of my reason. It is here that Korsgaard’s intersubjectivism becomes apparent. In offering you my reasons I am offering you the ‘familiar voice of humanity, not the voice of alien idiosyncrasies’ (Korsgaard 1996: 290). Subsequently, the value of our ambitions and personal obligations is not, as the objective realist would have it, intrinsic, but rather an expression of the interest in other agents or humanity. Qua intersubjectivist, an agent first understands himself as simply an agent among many, and then attempts to understand and share the ends of others. The objective realist works in the opposite direction. He must first see if he can share another’s ends, and then decide what relationship he wants to have with others. Korsgaard regards this as a mistake: ‘We should promote the ends of others not because we recognize the value of those ends, but rather out of respect for the humanity of those who have them’ (Korsgaard 1996: 279).

5.  A False Dichotomy?

With regard to the nature of value, for Korsgaard there are only two possibilities: either all value is accounted for in terms of some mind-independent Moorean objective realism, placing it in no relation to the ← 20 | 21 → subjective interests of agents whatsoever (objective realism), or it is explained in terms of intersubjectivism, where the interests of rational agents enter into the very analysis or content of value itself in the sense of rendering all values relational, i.e. the value of an object of moral concern obtains ‘in its relation to the subject’ (Korsgaard 1996: 279). However, this leaves Korsgaard committed to the implausible idea that all values are extrinsic values, as their existence depends solely on a process of construction by rational agents.

Following Alan Thomas, we can refer to this importation of relationality into the very content of value as relationality in value (Thomas 2006: 48). It is this peculiar notion which seems to fuel the idea of agent-relative value: that an agent’s apprehension of value somehow constitutes it, i.e. the idea that the value of my ambition to climb Kilimanjaro is essentially mine is what constitutes it being valuable. This idea seems confused. Indeed, it is more than plausible that you (or anyone) can grasp the value of my ambition to climb Kilimanjaro in the same manner as you can grasp the value of your ambition to, say, play the piano. For Thomas, the very notion of agent-relative value points towards a more interesting distinction between relational and non- relational intrinsic values.

A value is both intrinsic and relational when it stands in an asymmetrical constitutive relation to a subject. In this sense, the value of my ambition to climb Kilimanjaro can only be accounted for by making reference to whom the project is valuable. However, unlike Korsgaard’s intersubjectivism, these relational values can be both relational and intrinsic:

[Relational intrinsic] values do not derive their value from the relations to any object outside their nature (extrinsically); it is, rather, that it is in their nature to be values or disvalues in so far as they stand in a certain constitutive relation. (Thomas, unpublished)

The key point to note is that this kind of relationality is not ‘agent-relativity’ in the Nagelian sense. Of course, the value of my climbing Kilimanjaro is essentially relational, but this is not explained by the fact that it is valuable for me, Jamie Buckland. Analogously, take the disvalue of pain. Of course the pain I feel when I stub my toe on the coffee table is bad in so far as it is my pain, qua my mental state, but this ← 21 | 22 → is not what constitutes its disvalue; ‘mineness’ cannot function as the metaphysical grounding of the value of pain qua mental state. The relationality concerns the metaphysical relationality of value (the value is constitutively the value of a subject), as opposed to Korsgaard’s peculiar phenomena of relationality in value.

At this point it might be objected that Thomas’s notion of relational intrinsic value is simply a truism. But this is avoided by the fact that there are also non-relationally valuable states of affairs whose value obtains simply by virtue of their intrinsic properties, i.e. the value of these non-relationally intrinsically valuable states of affairs can be explained without making reference to the relational states of a subject. For instance, Thomas is happy to identify the notion of a non-relational intrinsic value with Moore’s idea that a beautiful world obtains without sentient subjects. However, unlike Moore’s view, Thomas’s idea is merely a micro-level claim because, as a whole, value itself is related to human interest via a doctrine of presupposition. Again though, this raises the concern that non-relational intrinsic values simply collapse back into relational intrinsic value or even extrinsic value.

The issue is complex, but a response ties into the wider sensibility theorist’s position that values are anthropocentric, but, nonetheless, real, so the relation between value and human interest is one of presupposition: ‘there is a sense in which value as a whole stands in relation to human interest. But this relation does not enter into the analysis of value itself’ (Thomas 2006: 48). The relationality of inherently relational (but intrinsically valuable) values, then, must be distinguished from the deeper, presupposed perspectival relationality found within our evaluative judgements:

Our relationship to value exhibits different kinds of relativity. One obvious case is that evaluative judgements deploy concepts that exhibit perspectivalness. Our peculiarities as a class of judgers enter into the possession conditions for the concepts that are deployed in our evaluative judgements. The metaphysical idea is that certain classes of judgeable contents are only available to judgers of a certain kind. But it is important to realise that in such an account this deep relativity to our metaphysical point of view does not enter in to the very content of the judgements we make, making them explicitly indexical in their very content. The relativity is rather presupposed, so that we can, from a standpoint of engagement with our conceptual scheme, make judgments that are, from our perspective, plainly true. (Thomas 2006: 47–48) ← 22 | 23 →

The idea that our relation to value is presupposed sheds further light on the distinction between metaphysical debates about the relationality of value, and the separate phenomenon of the relationality in value that besets Korsgaard. The problem with Korsgaard’s analysis is that the relationality of value has become part of the very nature of value itself – it has become relationality in value. In keeping this distinction separate the door is opened for those who wish to defend a metaphysical account of certain intrinsic values as constitutively relational without them being either relative or extrinsic or even instrumental.

6.  Value Primacy

How, then, does the foregoing lend support to the primacy of value; the idea that talk of value cannot be reduced to talk of the reason-giving force of non-evaluative properties?

As we saw above, for both Nagel and Korsgaard the idea of Moorean objective realism is misguided: evaluative judgements depend on judgements about the presence of particular reason-constitutive considerations. In this sense, the objectivity and reality of value is grounded fundamentally in the conception of a practical agent; values arise from and are analysed in terms of the conception of good reasons for action. This idea can be understood as a precursor to what has now become known as the ‘buck-passing’ theory of value, namely because Nagel’s normative realism ‘passes the buck’ from evaluative talk to the grounding of practical reasons for action, i.e. goodness is not a reason-grounding property itself, but is reducible to the reason-giving force of other properties, to quote Scanlon:

Goodness is not a single substantive property which gives us reason to promote or prefer the things that have it. Rather, to call something good is to claim that it has other properties (different ones in different cases) which provide such reasons. (Scanlon 1998: 11)

However, the sensibility theorist’s account of value offered above explicitly rules out passing the buck because evaluative attitudes cannot ← 23 | 24 → be characterized without ‘ineliminable reference to the values that these attitudes bring into view’ (Brewer 2006: 157 [emphasis added]). In following Thomas by understanding the metaphysical status of value as both anthropocentric, yet real, value is cited ‘irreducibly and ineliminably in the best explanation of our formulation of moral beliefs’ (Thomas 2006: 48 [emphasis added]; Wiggins 2002: Ch. IV, V). This is, essentially, a phenomenological appreciation, and irreducible to the recognition of reasons for action.


Brewer, T. (2006). The Retrieval of Ethics. Oxford: Oxford University Press.

Korsgaard, C. (1996). Creating the Kingdom of Ends. Cambridge: Cambridge University Press.

Kraut, R. (2007). What is Good and Why? The Ethics of Well Being. Cambridge (MA): Harvard University Press.

McNaughton, D., & P. Rawling (1991). ‘Agent-Relativity and the Doing-Happening Distinction’. Philosophical Studies, Vol. 63, No. 2: 167–185.

Moore, G. E. (1903). Principia Ethica. Cambridge: Cambridge University Press.

Nagel, T. (1970). The Possibility of Altruism. Oxford: Oxford University Press.

(1986) The View From Nowhere. Oxford: Oxford University Press.

___(1991) Equality and Partiality. Oxford: Oxford University Press.

Portmore, D. (2013). ‘Agent-Relative vs. Agent-Neutral’, in LaFollette, H. (ed.), International Encyclopedia of Ethics. Oxford: Blackwell.

Scanlon, T. M. (1998). What We Owe to Each Other. Cambridge (MA): Harvard University Press.

Thomas, A. (2006). Value and Context. Oxford: Clarendon Press.

Wiggins, D. (2002). Needs, Values, Truth (3rd edition, amended). Oxford: Clarendon Press.

1 I want to make it clear from the outset that I have no intention of settling anything in this paper. The idea is simply to outline a position I find plausible, and to demonstrate how such a theory lends support to the primacy of value.

2 Though one may agree that values are not reducible to the non-normative without thinking that values = reasons.

3 Importantly, the distinction between agent-neutral and agent-relative value does not map onto the intrinsic/extrinsic distinction. Nagel stresses that there is a separate metaphysical question concerning the manner in which reasons vary in their externality or independence from human concerns. He acknowledges the fact that most of the apparent reasons visible from our personal standpoint are intimately connected with our first-personal interests and desires, yet it seems that many of the objects of such interests and desires have an intrinsic value or a goodness in themselves, independently of how it is valued, or the satisfaction an agent may derive from valuing such objects, i.e. that some values are not reducible to their value for anyone. Admittedly, Nagel is unsure how to establish the existence of such values, but seems to regard the aesthetic value of the Frick Collection as a prime example of an object of external interest. Nevertheless, contra to Moore’s beautiful world, which retains its value without sentient beings, Nagel wants to avoid the ‘implausible consequence’ that the Frick Collection somehow retains its ‘practical importance’, even if humanity were destroyed (Nagel 1986: 153).

4 Incidentally, Nagel regarded the argument of The Possibility of Altruism as intimately related to Moore’s infamous argument against the egoist’s notion of agent-relative value or personal goodness although, on Nagel’s understanding, ethical egoism is a theory primarily concerned with reasons for action rather than the good (Nagel 1970: 86).