A New and Improved Whole Life Satisfaction Account of Happiness Suikkanen, J. (2019). The advice models of happiness: A response to Feldman. International Journal of Wellbeing, 9(2), 8-13. doi:10.5502/ijw.v9i2.837 Jussi Suikkanen University of Birmingham jussiphil@gmail.com Copyright belongs to the author(s) 8 ARTICLE The advice models of happiness: A response to Feldman Jussi Suikkanen Abstract: In his critical notice entitled “An improved whole life satisfaction theory of happiness?” focusing on my article that was previously published in this journal, Fred Feldman raises an important objection to a suggestion I made about how to best formulate the whole life satisfaction theories of happiness. According to my proposal, happiness is a matter of whether an idealised version of you would judge that your actual life corresponds to the life-plan, which he or she has constructed for you on the basis of your cares and concerns. Feldman argues that either the idealised version will include in the relevant life-plan only actions that are possible for you to do or he or she will also include actions and outcomes that are not available for you in the real world. He then uses examples to argue that both of these alternatives have implausible consequences. In response to this objection, I argue that what is included in the relevant life-plan depends on what you most fundamentally desire and that this constraint is enough to deal with Feldman’s new cases. Keywords: happiness, whole life satisfaction, Fred Feldman, conditional fallacy 1. Introduction It seems clear that happy people are satisfied with their lives – they must surely appreciate how well their lives are going. There is a long philosophical tradition of developing this simple observation into a full-blown theory of happiness.1 According to the resulting “Whole Life Satisfaction” theories (hereafter “WLS theories”), being happy is, in some way to be specified further, a matter of judging that your life compares well with your life-plan and especially a matter of feeling satisfied as a consequence. In an earlier article and in his recent book on happiness, Fred Feldman makes a powerful objection to the WLS theories that challenges their very fundamental structure (Feldman, 2008, 2010, ch. 5). It has the form of a dilemma. The objection begins from the idea that either we have to accept an actualist or a hypotheticalist version of the WLS theories. It then argues that, whichever way we choose, we are led to implausible conclusions concerning how happy different agents are. The most basic actualist versions of the WLS theories make the happiness of a person depend on an actual judgment which that person must make about how well her life is going compared to her life-plan. The problem here, however, is that many of us are so immersed in our daily activities that we have little time to form any life-plans or consider how well our actual lives would compare to them. Yet we do not want to claim that we are necessarily for this reason unhappy. 1 For a list of references, see Suikkanen (2011, p. 150 fn. 6). Advice models of happiness Suikkanen 9 To avoid this problem, the WLS theorists could suggest that perhaps our happiness is instead a function of how well we would judge our lives compared to our life-plans if we were to form such plans and considered how well our lives matched those plans. Yet, Feldman’s insight is that these kinds of hypotheticalist accounts are flawed too. A spontaneous person who is forced to form life-plans and compare her new reflective life to it in the counterfactual condition is likely to judge that her life is now not going according to her plan, as she had no intention to form life-plans or engage in self-reflection in the first place. The point of my previous article in this journal (Suikkanen, 2011) was to investigate whether the WLS theories could be structured in a way that would enable them to avoid the previous dilemma. My diagnosis of Feldman’s wonderful objection was that it revealed that the hypotheticalist WLS accounts commit the so-called conditional fallacy.2 These views attempt to capture happiness in terms of subjunctive conditionals: what you would judge, if you were in certain circumstances. However, putting you into the relevant counterfactual conditions changes your life in the relevant respects, and so the judgments you make about your own life in those circumstances have little relevance with respect to how happy you are in your actual life. At this point, I noticed that similar conditional fallacies have recently been committed elsewhere in ethics, for example, in the debates concerning practical reasons where it has been suggested that we should understand reasons in terms of what we would desire if we were fully rational.3 The lesson of these debates has been that this doesn’t work: making you fully rational changes your situation in a way that makes your desires in that new situation irrelevant to the reasons you actually have in the real world where you are not as rational (Smith, 1995, p. 111). These analogical debates gave me hope. In that context, Michael Smith has found a way of avoiding the conditional fallacy. In order to give an account of what reasons you actually have, Smith (1995, sec. 1) formulated a distinct evaluating perspective from which an idealised version of you has certain desires with regards to what the real version of you is to do in the evaluated actual world where you remain the same. In this case, the idealising changes made to the desires of your idealised version cannot change the features of the evaluated actual situation and so the conditional fallacy is avoided. I used a similar strategy to construct a new structure for the WLS views that could enable the resulting views to avoid the conditional fallacy horn of Feldman’s dilemma (Suikkanen, 2011, secs 5-6). This strategy relied on drawing the distinction between the actual person in the real world whose happiness is being evaluated and the idealised version of this person who is a theoretical construct responsible for forming an ideal life-plan on the basis of the person’s cares and concerns, and making a judgment of how well that person’s real life conforms to the hypothetical life-plan. This enables the WLS views to avoid the problem of the happiness of spontaneous people, as the individuals who are happily immersed in their activities can continue to do just that. All the view requires is that their evaluating version comes to judge that there is a match between the life-plan they construct out of the real person’s cares and concerns and the real person’s actual life. These kinds of advice model versions of the WLS theories then have the right structure to avoid Feldman’s powerful objection that seemed fatal to the more traditional WLS theories. I still think that this is exactly right, and nothing in Feldman’s critical notice gives me reason to think otherwise (see Feldman, 2019). The objection he puts forward does not challenge the idea that the advice model views have the right structure to avoid his previous 2 This type of fallacy was first made famous by Shope (1978). 3 See, e.g., Brandt (1979, ch. 1). Advice models of happiness Suikkanen 10 objections to the hypotheticalist versions of the WLS theories. In fact, as we will see below, Feldman’s new objections do not even turn on the advice model structure, as similar objections can be made to all versions of the WLS theories. I also want to emphasise that, in the original article, I presented the advice model only as a flexible framework in which different versions that have different extensions can be formulated (Suikkanen, 2011, sec. 6.1). In this respect, I was inspired by the recent debates about consequentialism in which it has been suggested that different versions of consequentialism share the same structure even if they can come to very different conclusions about which actions are right as a result of adopting different theories of value.4 In the same spirit, I thought that different versions of the advice model WLS theory, which have different extensions, can be formulated by specifying the idealised version of the person whose happiness is being evaluated in different ways. Perhaps I was not clear enough about this in my original article. This is shown by the fact that Feldman’s objections make certain substantial assumptions about how exactly I would specify the idealised evaluator and the resulting version of the advice model. It was, however, never my intention to be so ambitious in my article. I assumed that finding out which version of the advice model turns out to be true should be done at a later stage by relying on the reflective equilibrium method.5 We could begin from certain intuitive assumptions, according to which the idealised evaluator has a more coherent version of the evaluated person’s set of desires, goals, intentions, cares and concerns, and he or she is also perhaps more informed. I thought that we can then compare the consequences of this basic view’s verdicts of which agents are happy to our own intuitions about different cases. If we find conflicts between the two, we can then both go back to re-evaluate our views of the idealised evaluator and also fine- tune our intuitions about the cases until we reach a reflective equilibrium. It is then fortunate that exactly the kind of cases which Feldman’s describes in his critical notice allow us to apply the previous method and thus learn more about the advice model and happiness. In these cases, Feldman specifies agents who are intuitively either happy or unhappy. His intention is to present these cases as ones in which my advice model version of the WLS theory comes to wrong conclusions. However, the way I understand these cases is that they enable us to get a sense of how we should best understand the idealised evaluator. I will conclude this response to Feldman by illustrating how this works. 2. Feldman’s new cases Consider Smoky first (Feldman 2019, sec. 3): Smoky smoked like a chimney for many years. Eventually it caught up with him and his health has been ruined. Now, as he lies on his deathbed at t, Smoky is miserable. He does best to cope with his fatal illness, but he cannot undo the mistakes of his youth… He is miserable. He deeply regrets the now unalterable errors of his past. If we are to have a plausible version of the advice model, it better not entail that Smoky is happy. Feldman, however, seems to suggest that my view will entail that he is. One reason for this is that Feldman assumes that the relevant life-plan created by Smoky+ will consist only 4 For an overview of these consequentializing debates, see Portmore (2009). 5 See Rawls (1971, 19-21 and 46-51). Advice models of happiness Suikkanen 11 of actions which Smoky can do in his present situation. In that case, Smoky+ might well judge that Smoky’s life is going according to his life-plan. This, however, never was a part of the view. I took the idea of life-plans from John Kekes’s brilliant 1982 article “Happiness.” As Kekes puts it, a “life-plan is the hierarchical ordering of the first-order wants” (Kekes, 1982, p. 364). According to this view, we have first- order desires for different outcomes – for our lives to have different general qualities. We then have to decide what role these wants are to have in our lives – we must make a commitment. These decisions then together amount to a life-plan that orders one’s cares and concerns in terms of their importance. This is what also happens when an actual individual constructs a life-plan and what I assumed likewise happens when the ideal versions of us construct life- plans for us on the basis of the first-order desires they have inherited from us. At this point, I believe we can assume that some of the fundamental wants the real Smoky has are health, long life, and taking part in the activities that he likes but cannot do when ill. So, if Smoky+ constructs a life-plan for Smoky by ordering all his wants in terms of how much they matter to Smoky, that life-plan will presumably give a central place to having those very things which Smoky deeply cares about. Because of this, Smoky+ is unlikely to judge that Smoky’s life matches well with his life-plan, and so it looks like there are versions of the advice model that come to the correct conclusions about this case. Feldman’s next case is the following (Feldman, 2019, sec. 4): Luckless… was born with a congenital condition that has made his life miserable. Suppose Luckless has done his best to cope with it, but to no avail. … But of course he is not [outstandingly happy]. In this case too, Feldman suggests that the advice model theories will imply in an objectionable way that miserable Luckless is outstandingly happy. This is because he speculates that, when Luckless+ makes a life-plan for Luckless, he will probably leave the unalterable components of Luckless’s actual life in place, and so the ideal life-plan for Luckless will contain all the misfortunes of his actual life. This is why Luckless’s life will compare well with the ideal life- plan and so the view will deem him happy. Here I do not believe that plausible versions of the advice model will have this consequence, for the same reasons as in Smoky’s case. I assume that many of Luckless’s most fundamental and deepest wants, cares and concerns have to do with living a life free of pain, being able to do things that he knows he would enjoy and so on. This is because a very good explanation of why Luckless feels so miserable is that he cannot get what he most desires. If these are Luckless’s most fundamental wants, then presumably they will play an essential role in the life-plan which Luckless+ constructs for him. This is why I believe Luckless+ would not come to judge that Luckless’s life is going according to his life-plan and so there are versions of the advice model that agree with Feldman about what we should say about this case too. Feldman argues, however, that the previous response makes the advice model unable to deal with his final case, Mr. Chipper (Feldman, 2019, sec. 4): Suppose Mr. Chipper wants to be healthy, wealthy, and wise. Suppose he knows that he has always been healthy, wealthy, and wise. Suppose, as a result, that he is satisfied with his life as a whole. Mr. Chipper might like to be healthier, wealthier, and wiser … [b]ut he knows that in virtue of his extraordinary health, Advice models of happiness Suikkanen 12 wealth, and wisdom, no improvement in these areas is possible. He is as healthy, wealthy, and wise as a person could be… He is very happy. Feldman claims that the advice model cannot lead to this conclusion because in the previous example Luckless+ includes things in Luckless’s life-plan that are not possible for him to obtain. So, Feldman suggests that Mr. Chipper+ will have to do the same. He too will form a life-plan for Mr. Chipper that will include taking the imaginary supplements that make him even more healthy, making the imaginary investments that make him amazingly wealthy, and taking the brain enhancing pills that make him even wiser. On the basis of this life-plan, Mr. Chipper+ will then come to judge that Mr. Chipper’s life is not going according to plan. This would make him an unhappy person, according to the advice model, which he is not. Here the challenge is to explain in a principled way why Mr. Chipper+ would not include these impossible actions in Mr. Chipper’s life-plan. This explanation must also still enable us to continue to claim that Luckless’s life-plan will include things that he just cannot obtain. This challenge does not seem too difficult to me. If we consider Luckless, it is intuitive that many of his most fundamental and deepest actual cares and concerns are directed at living a pain-free life and also a life that allows him to take part in many of the important human activities that are not currently available for him. It is then plausible to think that Luckless+ takes the fundamentality and centrality of these cares and concerns into account in formulating the life-plan for Luckless. Yet, the way in which Feldman describes Mr. Chipper makes it clear that his most fundamental cares and concerns have absolutely nothing to do with being even healthier, wealthier, and wiser. As Feldman puts it, Mr. Chipper merely “might like to be healthier, wealthier and wiser” (Feldman, 2019, sec. 4). If these desires are in this way mere wishes or peripheral fleeting desires, then it’s not clear to me that they would play a role in Mr. Chipper+’s hypothetical life-plan for Mr. Chipper. This is exactly because the whole point of the life-plan is to structure Mr. Chipper’s actual wants in terms of how central a role they play in his psychological make-up. As a result, Mr. Chipper+ would not judge that Mr. Chipper’s life is not going according to his life-plan merely because these further wishes are not satisfied in his actual life.6 This makes me think that the solution to these new cases does not really have anything to do with whether the idealised agents should only include possible actions and outcomes in the life-plans or also things which the actual agents cannot do or achieve. Rather, the life-plans are better understood in terms of what kind of outcomes the agents themselves want most – what kind of generic qualities the agents fundamentally want their lives to have. And, as the last two examples show, it is important to emphasise the fact that, when the ideal versions of the evaluated agents make life-plans for those agents, they will take into account how central a role the different cares and concerns out of which the life-plans are constructed play in the agents’ psychologies. Furthermore, the ideal versions of the agents will also, in judging how 6 It could be objected that a version of Mr. Chipper could actually ultimately desire more health, wealth, and wisdom than he currently has and what he believes is achievable for him. In this case, his ideal version might include these seemingly impossible goods in his life-plan and judge that the actual life of Mr. Chipper does not match up to that plan. In this case, the proposed view would entail that this version of Mr. Chipper is unhappy. However, given that the ultimate desires of this Mr. Chipper are frustrated, and he is aware of this, this sounds like a plausible conclusion to draw. Of course, it is true that many agents tend to desire less those things that they believe are impossible for them to achieve (but this is not always the case – see Smoky’s case above). This seems to fit the way in which many empirical life satisfaction studies have found evidence of adaptation (see, e.g., Luhmann et al. (2012)). Advice models of happiness Suikkanen 13 well an agent’s life matches up to her life-plan, focus on whether the agent’s life satisfies her core cares and concerns. If the more peripheral elements of the agent’s motivational set are left unsatisfied, this will not make the agent’s ideal version judge that the agent’s life is not going according to her life-plan.7 It seems to me that this is all we need to say in order to defend the advice model against Feldman’s new challenge. Despite this, I am still extremely thankful for Feldman’s new cases, as they have helped us to get closer to how we should understand the ideal versions of the agents in the advice model and what their role is. Author Jussi Suikkanen University of Birmingham jussiphil@gmail.com Publishing Timeline Received 15 January 2019 Accepted 17 January 2019 Published 9 July 2019 References Brandt, R.B. (1979). A theory of the good and the right. Oxford: Oxford University Press. Feldman, F. (2008). Whole life satisfaction concepts of happiness. Theoria, 74, 219-238. Feldman, F. (2010). What is this thing called happiness? Oxford, England: Oxford University Press. Feldman, F. (2019). An improved whole life satisfaction theory of happiness? International Journal of Wellbeing, 9(1), 1-7. doi:10.5502/ijw.v9i2.762 Kekes, J. (1982). Happiness. Mind, 91(363), 358-376. Luhmann, M., Hofmann, W., Eid, M., & Lucas, R. E. (2012). Subjective well-being and adaptation to life events: A meta-analysis on differences between cognitive and affective well-being. Journal of Personality and Social Psychology, 102(3), 592-615. Portmore, D. (2009). Consequentializing. Philosophy Compass, 4(2), 329-347. Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University Press. Shope, R. (1978). The conditional fallacy in contemporary philosophy. The Journal of Philosophy, 75(8), 397- 413. Smith, M. (1995). Internal reasons. Philosophy and Phenomenological Research, 55(1), 109-131. Suikkanen, J. (2011). An improved whole life satisfaction theory of happiness. International Journal of Wellbeing, 1(1), 149-166. 7 Of course, this still leaves open many interesting questions about the proposal. For example, we can construct different versions of the view by varying exactly which fundamental desires should be taken into account by the ideal version of the agent. Some agents change their fundamental desires, and so the question is whether their ideal versions should, in the construction of the life-plan, only take into account their actual version’s current desires, or, rather, some set of the ultimate desires both before and after the change. Views that take a different stand on this issue will disagree about which agents are happy. The hope would be to use the kind of reflective equilibrium methodology illustrated here to discover what the most plausible version of the view is in this respect too. But carrying out this work will have to be left for another occasion.