Kinds of values, Kinds of attitudes

Kinds of values, kinds of attitudes 7th March 2013 Olivier Massin Department of Philosophy, Geneva University/CISA Abstract I shall present three fundamental distinctions among kinds of values discussed by contemporary axiologists and the way they relate to kinds of affective attitudes. (i) Thin vs. thick values. Some things are good; some things are delicate, fair, graceful... Relatedly, some emotions are thin, aimed only at the goodness/badness of things (liking/disliking); some are thick, directed at the thick values of things (pride, contempt, fear. . . ). (ii) intrinsic vs. extrinsic values. Some things are good in themselves, some others are good in virtue of being related to others things of value. Relatedly, some things are appraised/liked/desired... for their own sake, some others only extrinsically. (iii) Personal values vs. impersonal values. Some things are good for some persons only. Some things are good simpliciter. Relatedly, one should distinguish between impersonal affective attitudes and "for someone’s sake" attitudes. 1 Thin vs. thick 1.1 The axiological distinction thin values: good, bad thick values: courageous, elegant, ugly, unjust, rude, ridiculous, confuse, appalling, marvelous, cruel, interesting, generous, pleasant, delicate, disgusting, tactful, comic, fair, clever, holy, polite, charming, vulgar, silly, sublime, useful, comic, cool, awful, etc. (distinction introduced by Williams, 2006, pp. 129 sqq. about ethical concepts). Thick values are “thicker” in the sense of having more “content” than thin ones: we learn more about x when we discover that x is admirable that when we just learn that x is good. • If x has a thick value, then x has a thin value (pro tanto): if x is unjust, then x is bad. • Whether the reverse holds is controversial (e.g. the value of knowledge does not appear to entail a further thick value; see Tappolet (2004) vs. Mulligan, 1989). The main debate among axiologists on this issue concerns which kind of value is prior to the other. The standard view (Moore, Ross, Sidgwick, Broad) has long been that thin values wear the trousers, but under the influence of Williams, many think now that thick values are the most fundamental ones. Fundamentality, with respect to that distinction, is mostly understood in terms of reduction: 1 axiological thinism: thick values boil down to thin values + some non-axiological property (Burton, 1992, Elstein and Hurka, 2010). ex: x is courageous=df x is good in virtue of being disposed to overcome his fear when facing a danger. axiological thickism: thick values are irreducible to thin ones. (Williams, 2006, pp. 141 sqq.). According to thinists, thick values stand to thin values in the same way that species stand to genus (species =df genus + some differentia). Thickists, on the opposite, standardly conceive of the relation between thin and thick values in terms of the relation between colour and redness (determinable/determinate). • one thinist argument: if thick values are not values in virtue of being constitutively good or bad, they must be values in virtue of belonging to a resemblance order. But there is no resemblance order among all thick values. • one thickist argument, “the shapelesness” objection: it very difficult to get at the nonaxiological property constitutive of thin values. e.g. there is no single non-axiological property that all unjust acts share. 1.2 The psychological distinction thin attitudes: love/hate (alternatively: liking/disliking; positive/negative attitude...). thick attitudes: admiration, fear, guilt, disgust, being offended, indignation, joy, envy... Thick attitudes are “thicker” in the sense of having more “content” that thin one. If we learn that Paul admires Mary, when learn something more than if we just learn than she has a positive attitude towards her. • If x has a thick attitude, then x has a thin attitudes (pro tanto): if x is disgusted, then x has a negative attitude. • Does the reverse hold? It is possible to have a positive attitude towards something without admiring it, envy it, cherishing it, being moved by it, etc. ? A debate akin to the debate between axiological thinists and thickists can be transposed among thin and thick attitudes: attitudinal thinism: thick emotions boils down to thin emotions plus some non-emotional difference. ex: S admires x =df S likes x because S believes that x is admirable. attitudinal thickism: thick emotions cannot be reduced to thin ones. Thinists think that emotion is a genus, than can be specified by different non-affective differentiae for each particular emotion. Thickists on the other hand favour a bottom-up approach: the are infinitely many affective episodes that are emotions in virtue of resembling each other, and that affectively differ from each other. 2 2 Intrinsic vs. extrinsic 2.1 The axiological distinctions Some things are good in themselves, for their own sake. Some things are good in virtue of the goodness of some other things. One often calls the value of the former intrinsic, and the values of the latter extrinsic. But this terminology is unfortunate for intrinsic, applied to property means something different. A better, though less common terminology, distinguishes fundamental from derivative values. derivative value: x has a derivative value relative to y=df x is (dis)valuable because y is (dis)valuable (x 6=y). fundamental value: x has a fundamental value relative to y=df x is (dis)valuable and x is not (dis)valuable because y is (dis)valuable (x 6=y). Some derivative values are more fundamental than others. If good chocolate cakes are good in virtue of giving pleasure, and if a recipe is good in virtue of leading to a good chocolate cake, the derivative value of the chocolate cake is more fundamental than the derivative value of the recipe. The most discussed type of derivative value is instrumental value. Clear cases of instrumental values are the values that causes possess in virtue of the value of their effects. Some might be willing to define instrumental values in that narrow causal way. But we shall here be interested in a broader concept of instrumental value. First, things may have instrumental value even if they do not actually cause some effect, but could do so (a knife remains instrumentally good when it does not actually cut, see Rønnow?Rasmussen (2002, pp. 29-30)). The relation of conduciveness is often introduced to widen the concept of causation in that way: conduciveness is not an actual causal relationship but is a dispositional property, so that entities that are not actually causing value-exemplifications might still retain some instrumental value. An instrumental value is then defined as the value that an entity has in virtue of being conducive to other valuable entities. This is better, but still too narrow. Some things might have instrumental value, in the broad sense, even if they do not enter into any actual or possible causal relationship with things of more fundamental value: • a formal method might be good in virtue of the validity of its results. • a political procedure might be good in virtue of the goodness of the decision it leads to. • a declaration of war might be good in virtue of the goodness of the war it leads to. • a reason might be good in virtue of the goodness of the action it leads to. • a one dollar note might be good in virtue of the acquisition of goods it allows. Such relations are not obviously causal relations, but they share two important features with causal relations: 1. the second relatum depends on the first: the result depends on the method, the decision depends on the procedure, the war depends on the declaration, the action depends on the reason, the acquisition depends on the dollar bill. 3 2. their relata are wholly distinct from each other: the methods and the result, the procedure and the decision, the war and the declaration, the reason and the action, the dollar note and the acquisition, do not share any part or constituent. instrumental value: x has a instrumental value relative to y=df (i) x and y are wholly distinct (ii) x is (dis)valuable because y, which is (dis)valuable, depends on it. Though instrumental values are by far the most discussed type of derivative values, and legitimately so, there are three other types of derivative values based on determination relations between their bearers. The second type of derivative value refers to the type of value that a part has in virtue of the value of the whole of which it is part. The typical example is the value of a single note of a nice melody. One standardly speaks here of contributory value (Lemos, 1995, pp. 41 sqq.; Tappolet, 2000, p. 24). contributory value : x has a contributory value relative to y=df (i) x is a part of y and (ii) x is (dis)valuable because y, which is (dis)valuable, depends on x. A third, neglected though important, type of derivative value is the value that entities have in virtue of the values of the wholly distinct entities they depend on. An example is the value that might attach to an effect in virtue of the value of its cause: some effects might be good because their causes are good. It is sometimes claimed that things created by God are good in virtue of the goodness of God; or that things created by an artist, whatever they are, are good in virtue of the genius of this artist. Let us call this kind of value consecutive value: consecutive value: x has a consecutive value relative to y=df (i) x and y are wholly distinct (ii) x is (dis)valuable because it depends on y, which is (dis)valuable. Consecutive values play a central role in important ethical and political theories: 1. Standard virtue ethics holds that actions are virtuous in virtue of being performed by virtuous agents. The fundamental moral value is here the virtue of the agent (including his character traits), and actions are good because they externally depend on virtuous agents: actions are then consecutively good. 2. Kantian ethics, on some readings, has it that actions are (morally) good in virtue of resulting from the (morally) good will. The goodness of an action is, likewise understood as having consecutive value: actions are good because they externally depend on good will. 1 3. The concept of consecutive value is sometimes used with respect to justice-injustice. One central thread of Hayek’s criticism of social justice (Von Hayek, 1973, bk 2) is that justice, as a fundamental value, applies to people or behavior, and can only derivatively qualify situations that result from just/unjust person or behaviors. Situations that result from unjust behaviors (such as robbery) are consecutively unjust. It is therefore meaningless, Hayek argues, to speak of the injustice of situations that do not result from unjust people or behavior. 1 Note that if the agent or the will is included among the constituent parts of the action, the value of the action is no longer a consecutive but a constitutive value, as will be discussed below. 4 4. symbolic value is the value that attaches, for instance, to a flag symbolizing a country. The value of the flag derives from the value of the country it symbolizes and the flag depends on the country it symbolizes. If the flag is not a part of the country, or the reverse, the flag’s symbolic value is a kind of consecutive value. (see Scheler, 1973, p. 104 for a defense of realism about symbolic values). The last category of derivative value to be mentioned here is the converse of contributory value: wholes sometimes have values in virtue of the values of their constituents. An alloy of gold and lead might be precious in virtue of the gold it contains. Following Ehrenfels (see Smith, 1986) one might speak here of constitutive value (see also Schroeder, 2008). constitutive value: x has a constitutive value relative to y=df (i) y is a part of x (ii) x is (dis)valuable because y, on which it depends, is (dis)valuable. In summary, instrumental, contributory, consecutive values and constitutive values are all derivative values that are grounded in ontological dependence relations between their bearers. Instrumental and contributory values are grounded in the dependence of their bearers upon bearers of (more) fundamental value. Consecutive and constitutive values are grounded upon the dependence of bearers of (more) fundamental values upon their bearers. Instrumental and consecutive values are grounded in relations of dependence between wholly distinct entities. Contributory and constitutive values are grounded in dependence relations between overlapping entities. These four different kinds of derivative values are presented in table 1. relations between x and y, y being of basic value y depends on x x depends on y x wholly distinct from y x overlaps y x has instrumental value with respect to y x has consecutive value with respect to y x has contributory value with respect to y x has constitutive value with respect to y Table 1: Derivative values grounded in dependence relation between their bearers This classification of derivative values is not intended to be exhaustive. This classification is exhaustive only with respect to a certain kind of derivative values, namely derivative values that are grounded in dependence relations between value bearers. One can think of at least two other kinds of derivative values that are not grounded in any dependence relations between their bearers. 1. The first kind of derived value that is not grounded in dependence relations among valuebearers might be called fetishist value. This is the value that, according to some, pertains to things that have been in the proximity of some other things of value without depending on them or grounding them in any way. Put something neutral in contact with something good, and it acquires a fetishist value. The value of Lady Diana’s dress might be an example. The dress does not appear to depend on Lady Diana: it could have existed, and be the dress it is, without having been the dress of Lady Diana. Still, just having been worn by Lady Diana confers some value on the dress. 5 2. The second kind of derivative value not grounded in dependence relations among value bearers is the case of the overall value of a value-bearer that has several values of different kinds (Zimmerman, 2001, p. 258). Overall value is not to be confused with the total constitutive value of a whole. First, simple things that have no parts lack constitutive values, but might have overall value. Second, things that have constitutive values, might also have other types of values: instrumental, contributory, etc., that should enter in the overall value. They might also have fundamental value of different kinds: something might have both a fundamental aesthetic value and a fundamental moral value. 2.2 The psychological distinction It is common to distinguish between intrinsic and extrinsic desires. Intrinsic desires bears on facts that one desires for their own sake; extrinsic desires, on the other hand bears on facts that one desire for the sake of something else (see Feldman, 1997, p. 89). One might try to capture the distinction between intrinsic and extrinsic desires as follows: extrinsic desire: S extrinsically desires that x occurs=df S desires that x occurs and S desires that x occurs only because [S desires that y occurs (x 6=y) and S thinks (feels, expects...) that the occurrence of y depends on the occurrence of x ]. intrinsic desire: S intrinsically desires that x occurs=df S desires that x occurs and S does not desire that x occurs only because [S desires that y occurs (x 6=y) and S thinks (feels, expects...) that the occurrence of y is related to the occurrence of x ]. A similar distinction can be drawn for affective attitudes: one might like something for its own sake, or one might like it because one likes something else which is related to it. extrinsic liking: S extrinsically likes x =df S likes x only because S likes y (x 6=y) intrinsic liking: S intrinsically likes x =df S likes x and S does not likes x only because S likes y (x 6=y) One might then distinguish: instrumental liking: liking money, because one likes buying goods. contributory liking: liking a tile because one likes the mosaic. consecutive liking: liking the world because one likes its creator. constitutive liking: liking the cake because one likes the chocolate in it. fetishist liking: liking Michel Sardou’s shirt because one likes Michel Sardou. overall liking: liking a thing one the whole because one likes it under several respects. (useful, beautiful...) 6 values fundamental derivative 7 not grounded on dependence relations between value-bearers fetishist overall grounded on dependence relations between value-bearers ... dependent on something good on which something good depends consecutive constitutive instrumental contributory effect of good cause whole with good part cause of good effect part of good whole Figure 1: Fundamental and derivative values 3 Personal vs. impersonal 3.1 The axiological distinction Some things are good simpliciter, or impersonally, but some things are good for some persons in a way they are not for others. Personal values have been of particular interest to recent value theory (see e.g. Sumner, 1996, pp. 20 sqq., Darwall, 2002, Feldman, 2004, pp. 135-6, (RønnowRasmussen, 2007, 2011), Rosati, 2006, 2008, Zimmerman, 2009). • some positive personal values: the value of children for their parents, the value of a compliment for the complimented, the value of medicine for the sick person, the value of the victory for the winner, the value of a person’s health for that person, the value of the satisfaction of one’s need, the value of a person’s life for that person, the value of a state for its citizens, the value of a person’s pleasure for her... • some negative personal values: the disvalue of one’s death, the disvalue, for us, of the death of our friends, the disvalue of danger for the endangered, the disvalue of the offence for the offended, the disvalue of the insult for the insulted, the disvalue of a person’s illness for herself, the disvalue of a punishment for the punished person, the disvalue of a person’s pain for that person... The central intuition underlying these examples is that some things (facts, episodes, substances...) are valuable for some people in ways they are not for other persons. Mary’s recovery is good for her in a way it is not for Paul. Warning: to say that x is personally good for P is NOT to say (i) that P believes, knows, feels that x is good (ii) that P likes x (iii) that only P can access x ’s goodness (iv) that P owns x which is good. If x is personally good for P, then x is of benefit to P, but P might not know x, might even dislike x, and might not own x in any strict sense. One way to figure out what personal values are it to appeal to Feldman’s crib test: Imagine that you are filled with love as you look into the crib, checking on your newly arrived firstborn child. The infant is sleeping peacefully. You might think of various ways in which the baby’s life could turn out. What schools will he attend? What career will he choose? What sort of personality and intellect will he have? Will he someday have children of his own? Your concern for the baby might express itself in the hope that, whatever he does, things will turn out well for him. You might hope that this baby gets a good life—a life good in itself for him. (Feldman, 2004, pp. 8-9) 3.2 The psychological distinction Echoing the distinction between impersonal and personal values, one finds a distinction between impersonal and personal attitudes. One might like something simpliciter ; or one might like something for someone’s sake (Rønnow-Rasmussen, 2011). Concern, care, or romantic love are positive personal attitudes of this kind. Moral emotions, on the other hand, such as indignation, are not personal: when one is confronted with an action that seems unjust, we are not indignant for somebody, but indignant tout court. Relatedly, some emotions are made correct by values simpliciter. Indignation is correct if its object is unjust. Injustice is not a personal value. Not all emotions, however, aim at impersonal values. Love –sentimental love– is not made correct by the impersonal values of the beloved. If it 8 were, and if indifference to the valuable is incorrect, then not only Romeo, but everybody, ought to love Juliet. So Romeo’s love is not made correct by Juliet’s goodness tout court, but by Juliet’s goodness for him. 4 Conclusion The three main distinctions between kinds of values and kinds of attitudes mentionned above (thin/thick; fundamental/derivative; impersonal/personal) can be crossed with each other. Some values might be thick, consecutive and impersonal, some other thin, fundamental and personal, etc. And likewise for attitudes: one might instrumentally and personally like chocolate (because it gives us pleasure), one might consecutively and impersonally love the world (because it has been created by a perfect God), etc. We get a bijection between kinds of values and kinds of psychological attitudes: • thick values ↔ thick attitudes • thin values ↔ thin attitudes • fundamental values ↔ fundamental attitudes • derivative values ↔ derivative attitudes • instrumental values ↔ instrumental attitudes • contributory values ↔ contributory attitudes • consecutive values ↔ consecutive attitudes • constitutive values ↔ constitutive attitudes • fetishist values ↔ fetishists attitudes • overall values ↔ overall attitudes • impersonal values ↔ impersonal attitudes • personal values ↔ personal attitudes What exactly is the relation between each kinds of value and each kinds of affective attitudes? One might rely on this equivalences to reduce values to attitudes. Two main ways of doing so: subjectivism about values: x has a value of kind K =df x is the object of an attitude of kind K. ex: being personally good is being personally loved. buck-passing about values:: x has a value of kind K = S a an appropriate attitude kind K towards x. Note that one might be reductionist about some values (e.g. economic values) but not others. On the opposite, on might try to reduce affective attitudes or emotions to apprehensions of values: perceptualism about emotions: S has an emotion of kind K =df S has a non-emotional attitudes (perception, believes, feelings; apprension) towards an x of kind K. Alternatively one can refuse any reduction, and claim that kinds of attitudes are made correct by kinds of values, without any of them boiling down to the other. 9 References Burton, S.L. 1992. “’Thick’ Concepts Revised.” Analysis 52:28. Darwall, S. 2002. Welfare and Rational Care. Princeton: Princeton University Press. Elstein, Y., Daniel, and T. Hurka. 2010. “From thick to thin: Two moral reduction plans.” Canadian Journal of Philosophy 39:515–535. Feldman, F. 2004. Pleasure and the Good Life: Concerning the Nature, Varieties, and Plausibility of Hedonism. Oxford University Press. —. 1997. Utilitarianism, Hedonism, and Desert: Essays in Moral Philosophy. Cambridge University Press. Lemos, R. 1995. The nature of value: axiological investigations. University Press of Florida. Mulligan, K. 1989. “Wie verhalten sich Normen und Werte zueinander?”, pp. . Rønnow-Rasmussen, T. 2007. “Analysing personal value.” J Ethics 11:405–435. —. 2011. Personal Value. Oxford: Oxford University Press. Rønnow?Rasmussen, T. 2002. “Instrumental Values: Strong and Weak.” Ethical Theory and Moral Practice 5:23–43. Rosati, C. 2006. “Personal Good.” In T. Horgan and M. Timmons, eds. Metaethics after Moore. Oxford University Press, USA, pp. 107–131. Rosati, C.S. 2008. “Objectivism and relational good.” SOY 25:314–349. Scheler, M. 1973. Formalism in Ethics and Non-Formal Ethics of Value. Evanston: Northwestern University Press, trad. M. Frings, R.L. Funk. Schroeder, M. 2008. “Value Theory.” Stanford Encyclopedia of Philosophy, pp. . Smith, B. 1986. “The theory of value of Christian von Ehrenfels.” Christian von Ehrenfels: Leben und Werk , pp. 150. Sumner, L. 1996. Welfare, happiness, and ethics. Oxford: Clarendon Press. Tappolet, C. 2000. Emotions et valeurs. Paris: Presses universitaires de France. —. 2004. “Through thick and thin: good and its determinates.” Dialectica 58:207–221. Von Hayek, F. 1973. Law, legislation and liberty. Univ. of Chicago Press. Williams, B. 2006. Ethics and the Limits of Philosophy. London: Routledge. Zimmerman, M. 2001. The nature of intrinsic value. Rowman & Littlefield Pub Inc. Zimmerman, M.J. 2009. “Understanding What?s Good for Us.” Ethic Theory Moral Prac 12:429– 439. 10