_______________________________________________________________________________
In “Naturalism, Normativity, and the Open Question Argument” Connie Rosati challenges new naturalism’s claim to have closed off any open questions regarding its definition of ‘good.’ Focusing on Peter Railton’s naturalist account, Rosati argues that one can still wonder whether what is ‘good for a person’ on the naturalist view is in fact good for a person (Rosati 65). If this is the case, then Railton’s definition will not have effectively captured ‘good for a person’ in purely natural terms. After developing Rosati’s argument, I will argue that it can be resisted in several ways and that the new naturalism is safe from her objections.
The new naturalist’s goal is to capture the normativity of goodness in purely descriptive and natural terms. He or she will wish to “construct a descriptive meaning for ‘good’ that secures its recommending and expressive functions simply in virtue of the proposed descriptive content” (Rosati 46-7). In keeping with this goal, Railton’s ‘synthetic naturalism’ defines ethical terms such as ‘good’ in terms of natural properties and relations to which they are identical a posteriori (Rosati 48, 65). The view is developed by first noticing that, as expressivism has shown us, ethical judgments are inherently motivating or recommending. As such, it is “a necessary condition on something’s being good that it be capable of motivating” (Rosati 49). This is the thesis of existence internalism. Second, if something is capable of motivating an action, then it must be something that we care about (Rosati 50). However, individuals obviously care about many different things, so what may be ‘good’ for one person may not be ‘good’ for another in the sense that one person may care about X while another person will not care about X. Thus, if something is to be considered good for a person, it must be “made for” or “suited to” that person (Rosati 50).
When something is good for a person, that person will be motivated by it under certain relevant conditions (Rosati 50). These relevant conditions, as Railton has developed his view, concern our cognitive status. We recognize that our desires can be misguided due to faulty or incomplete information or processes and so they, and we, may not fare well in the world (Rosati 51). Consequently, “the normativity of goodness [for a person] involves not only an internal connection to the individual whose good it is, but justification in the sense of epistemic warrant” (Rosati 51-2). One must be justified in believing that a course of action will indeed be good for oneself. Thus, the new naturalist arrives at a definition of goodness for a person: “X is good for a person A iff A would want her actual self to want it were she fully informed and rational and contemplating the position of actual A as someone about to assume her position” (Rosati 52). As Rosati explains, “a person’s good is indicated by what she would want her actual self to want were she fully informed about her nature and circumstances and completely rational, suffering no cognitive defects and committing no errors in instrumental reasoning” (Rosati 53).
Rosati believes that Railton’s definition fails because it makes a person’s motivational system determine what counts as ‘good’ for that person, and it may be this very thing that is being questioned by that person. Consider Rosati’s supposed counterexample in which Sally is concerned that she is living her life in an overly cautious, orderly, and restrictive manner. She dislikes the fact that she likes order so much, and admires her friend Madelyn’s spontaneity. Worried that she might be missing out on important things in life, she wonders whether she should become more spontaneous, and seek therapy in order to accomplish this (Rosati 53). Given this concern, she will wonder if “what someone like herself would want for her if fully informed and rational is indeed good for her” (Rosati 54). If Sandy can wonder this, then there is still an open question related to Railton’s account of ‘good for a person.’
What exactly is the worry here? It seems to be that our personalities may be biased in such a way that a fully informed and rational individual with our personality may not make the best recommendation about what we should do. Rosati does not explicitly define what she means by “full rationality,” but she does say that a completely rational person would (1) suffer no cognitive defects and (2) would make no errors in instrumental reasoning (Rosati 53). The biases cannot be from difficulties in reasoning properly and so must come from one’s desires. Sandy wonders if she ought to desire different things (e.g., spontaneity). If she was a different sort of person and had different character traits, she would desire different things, and so she wonders if she should be a different sort of person. Hence, the type of person she currently is cannot play a role in helping her decide what type of person she ought to be, for this is the very thing she is questioning.
Before moving on, it is important to note that Rosati seems to be assuming a Humean view of practical rationality in which our actions issue from a belief-desire pair that determines our actions. If two individuals had the same information, the same beliefs, and the same desires, they would make the same decisions. Consequently, any two fully rational and informed individuals will differ in their recommendations of action only if they have different desires that are relevant to the recommendation.
I believe the new naturalist can resist Rosati’s argument in several different ways. First, one can take issue with her understanding of ‘full information.’ What ‘full information’ amounts to is never defined by Rosati (probably because she believes that the notion is “not clear or coherent”) (Rosati 67). Still, she uses it in such a way that it is limited to information concerning facts about one’s actual surroundings (i.e., circumstances) and facts about one’s actual physical and psychological status (i.e., nature). Notice that important modal information is missing regarding what would happen if Sandy were to follow different recommendations and what another person with different character traits and desires would recommend.
Rosati seems to have assumed that if one does not have the proper desires regarding what to do, then one cannot know what to do. But this is false. If an idealized individual can have modal knowledge, we can then say that an idealized individual will know which desires and actions are best to have in order to satisfy Sandy’s most fundamental desires and can advise Sandy about how to shape these desires through action, even if the idealized individual does not have these certain character traits. Sandy’s idealized self will know how a less rigid person will assess Sandy’s situation and will know if being less rigid will lead to satisfying her fundamental desires. Thus, Sandy’s idealized self can make the best recommendation for Sandy and consequently there will be no open question about whose advice to follow.
Second, notice that I am assuming that Sandy has fundamental desires that she identifies with. These desires are unchangeable. Take Sandy for example. In order to make sense of her wondering whether to be another sort of person, we have to attribute to her some feature that will survive any change. Not all of her desires can change, particularly those that are fundamental to who she is. Instead, she seems to be concerned with modifying some of her desires and motivations, and only those that she does not identify with. Rosati seems to be presupposing that one can critically assess each of one’s desires, as though we can put them all aside and look at them. She claims that Sandy can consider “which desires to embrace or reject, which motivations and traits to retain or transform. Through her reflection, she comes to dissociate herself from some of her desires, motivations, and traits, while identifying with others” (Rosati 61). This is true with regards to first order desires, but is more difficult and perhaps impossible with higher order and more fundamental desires. Indeed, how should we critically assess our desires if there were no guiding or overriding desires to assess them by? Thus, I believe that Rosati’s notion that we “self-invent” is too strong, for we can only “self-invent” within limits.
Rosati may respond that perhaps some of our desires are fixed, but this gives rise to the very problem she has pointed out, for idealized advisors will have different fundamental desires and therefore will recommend different things. However, it seems to me that any two sorts of people will have some fundamental desires in common, and if so, then all idealized advisors will have some common fundamental desires that can ground and unify their recommendations. Hume believed that there are universal sentiments in humans that come from our nature. Consequently, humans will have some fundamental desires in common. So it is not true that “an account of a person’s good will fail to be suited to persons as self-inventors insofar as it treats a person as identical with certain motivations or traits, for this is to accord her current features a normative authority that they lack” (Rosati 62). There are current motivations or traits that are given to us by nature and hence do have normative authority. For our purposes, we only need to identify an individual as identical to her most fundamental desires that make her who she is and which are common to all others (Rosati 62).
For example, it seems to me that at the very least, no matter what sort of person Sandy could become, she will always care about her own happiness (however one wishes to spell this out naturalistically). Sandy is wondering whether she would be happier if she were more spontaneous and less rigid like Madelyn. Happiness appears to be intrinsically good and so one cannot claim that persons like us (capable of emotion) could possibly not desire happiness in some form. Furthermore, any idealized individual with full information and full rationality will know how to maximize this happiness in any given situation, and so it seems to me that any two idealized individuals will agree in their recommendations if desire-satisfaction is part of what makes a choice better than another choice (Rosati 57).
Rosati could object that “our concern is not simply with happiness, for instance, but with what sort of happy person to be” (Rosati 60). The idea behind this objection seems to be that there are many ways to be happy, and Sandy is wondering which way she should take. The new naturalist can respond in two ways. First, perhaps there are many ways to be happy, but only one way to be maximally happy. If this is true, then any two idealized individuals will recommend the same route to becoming maximally happy. Second, perhaps there are many ways to becoming maximally happy, and if so, then any of these choices is equally correct. Rosati herself allows for such a possibility: “a person’s good may be indeterminate in some cases, either because there is no determinate answer as to what her good consists in, or because several options are equally acceptable” (Rosati 67). This is a form of openness that is not problematic for the naturalist.
Third, even if humans do have common and fundamental desires, Rosati worries that in order to become fully informed and rational we must lose our capacity to judge what is good by becoming cold and calculating. If so, “the reactions of persons with these characteristics would hardly be recommending to most of us, regardless of how fully informed and rational they might be” (Rosati 57). The new naturalist can make two responses here. First, there is no reason to assume that full rationality and the possession of full information are incompatible with feelings and good character. For example, the notion that God is omnibenevolent and omniscient does not seem to be inherently contradictory. Second, suppose that the process does make an individual cold and calculating. Does that individual need to feel in order to make good recommendations? It does not seem so. We can imagine a computer that has full information and rationality and possesses all facts about feelings and character traits although it possesses none of them. Why should we reject its recommendations, especially if following them has always proven to turn out for the best? We can still accept its judgments without wanting to be like it in all respects, because we value it as an advisor regarding our actions and not as a model of what sort of person we should be. Thus, either way, it would be wise to follow the advice of a fully informed and fully rational advisor’s recommendation concerning our good.
Fourth and finally, Rosati is assuming a Humean view of practical rationality when one might instead adopt a Kantian view. Hume believed that our desires were not subject to rational criticism while Kant believed that at least some of them were. Railton agrees with the Kantian view and believes that we can criticize our desires on rational grounds (Railton 271). Indeed, how can we understand Sandy as wondering whether she should desire something else unless she takes some of her current desires to be open to criticism in some way? If this is true, then at least some of our desires will be excluded because they are irrational, and it seems likely that some desires will be demanded by rationality. Consequently, a fully rational and informed individual like us will have some of the same desires as another fully rational and informed individual like us in virtue of the fact the she is fully rational. As long as these common desires control the deliberative process, then two fully rational and informed individuals will make the same recommendation and the open question is avoided.
Given these four available responses, I believe that the new naturalist can resist Rosati’s argument. Armed with full rationality, full information, and fundamental yet universally held desires that may themselves be demanded by rationality, a person’s idealized advisor will undoubtedly make the correct recommendation concerning whether something is ‘good for a person.’ Thus, any open questions have been effectively closed.
Works Cited and Consulted
Darwall, Stephen. “Morality and Practical Reason: A Kantian Approach.” The Oxford Handbook of
Ethical Theory. Ed. David Copp. New York: Oxford, 2006. 282-320. Print.
Railton, Peter. “Humean Theory of Practical Rationality.” The Oxford Handbook of Ethical Theory.
Ed. David Copp. New York: Oxford, 2006. 265-281. Print.
Rosati, Connie. “Naturalism, Normativity, and the Open Question Argument.” Nous 29.1 (1995): 46-
70. JSTOR. Web. 25 Aug. 2011.
Sturgeon, Nicholas. “Ethical Naturalism.” The Oxford Handbook of Ethical Theory. Ed. David
Copp. New York: Oxford, 2006. 91-121. Print.
No comments:
Post a Comment