INTRODUCTION
Theories of welfare attempt to tell us what is good for a person. One such family of theories is Hedonism. Hedonist theories, broadly construed, are those theories stating that happiness is the only thing that is ultimately (non-instrumentally) good for a person. The happier a person is the better off they are; the better a life they live. While there are many objections to hedonism, arguably none have been as influential as Nozick's Experience Machine:
Suppose there were an experience machine that would give you any experience you desired. Superduper neuropsychologists could stimulate your brain so that you would think and feel you were writing a great novel, or making a friend, or reading a book. All the time you would be floating in a tank, with electrodes attached to your brain. Should you plug into this machine for life, preprograming your life's experiences?Footnote 1
Would you plug into an experience machine? If hedonism were true, you would, or should, plug in. But according to Nozick, many, if not all, would refuse the opportunity. Since we refuse to plug in, Nozick concludes that hedonist theories of welfare are incorrect. Nozick's point seems to be this: we intuitively recognize reality as being more prudentially valuable (i.e. good for us) than happiness. Call the intuition to prefer reality to the experience machine the anti-hedonist intuition. But he is not alone in drawing this conclusion. His experience machine thought-experiment (hereafter EMTE) has been so influential that it is often only mentioned in passing as a refutation of hedonism.Footnote 2
Recently however, a growing number of philosophers question Nozick's EMTE.Footnote 3 They argue that the vast majority of people do not share this anti-hedonist intuition, often citing empirical data involving variant EMTEs to support their point. Rather than finding consensus that can be plausibly understood as a ‘vast majority’, intuitions actually conflict. Because there is such conflict amongst intuitions, I call this the conflicting intuitions problem.
The hedonist argument against Nozick runs thus. Empirical data show that the vast majority of people do not share the anti-hedonist intuition. Since people do not prefer reality to experience, Nozick's EMTE is found wanting. But the conflicting intuitions problem is a problem for hedonists too, as our intuitions do not side with hedonism either. Hedonists then attempt to debunk anti-hedonist intuitions as being ill formed because of psychological and evolutionary cognitive biases. So what do we learn from EMTEs? And are conflicting intuitions really a problem for them?
This article offers a novel solution as to why there are conflicting intuitions: people are making different cost-benefit analyses about where the welfare pay-off lies for any given EMTE. My solution is available if we adopt a two-dimensional characterization of welfare. This bipartite distinction draws a line between the dimension called ‘doing-well’ (mental states) and another dimension called ‘going-well’ (states-of-affairs). By adopting the view that welfare judgements are formed within a two-dimensional space, we can make sense of (a) the changes of intuitions amongst sample-sets, and (b) the disparity between them. In doing so, we do not have to make controversial claims that all anti-hedonist intuitions are misguided.
This article will proceed as follows. In section I I sketch out the experience machine challenge to hedonism. In section II I give a detailed explanation of how hedonists have attempted to cut Nozick's argument down at its root, by attacking the assumption that the vast majority of people prefer reality. In section III I start from recent work by Lin, who utilizes comparative welfare judgements.Footnote 4 Here, I sketch out a preliminary reason for why welfare might be best construed as two-dimensional, namely through comparative welfare judgements. I then show in section 4 how such a characterization can provide intuitive solutions to the empirical data provided by hedonists against experience machine thought-experiments.
I. NOZICK'S EXPERIENCE MACHINE
Would you plug into Nozick's experience machine? Nozick believes our answer, like his, is an irrevocable no. According to Nozick, our anti-hedonist intuition reveals that, ‘something matters to us in addition to experience’.Footnote 5 And that something, again according to Nozick, ‘is an actual connection with reality’.Footnote 6 He gives three anti-hedonist judgements that fit with both the intuitive rejection of hedonism and provide support for the value of reality:
First, we want to do certain things, and not just have the experience of doing them . . . A second reason for not plugging in is that we want to be a certain way, to be a certain sort of person . . . Thirdly, plugging into an experience machine limits us to a man-made reality, to a world no deeper or more important than that which people can construct.Footnote 7
But how do the anti-hedonist intuition and anti-hedonist judgements fit together to form an argument? It is not entirely clear. As Barber points out, such judgements could be anti-hedonistic reasons by themselves, thus rendering the anti-hedonist intuition non-substantial.Footnote 8 Yet this seems mistaken: the most pervasive interpretation of Nozick's argument is that our intuition does play a substantial role. That is, the anti-hedonist intuition elicited is itself supposed to count as evidence against hedonism. Indeed, when turning to the literature those who both accept or reject Nozick's conclusion do so with only the anti-hedonist intuition in mind, paying no heed to the three anti-hedonist reasons provided above.Footnote 9
Yet even after accepting the intuition as evidence against hedonism, there is some division as to how to construct the argument around it. According to Weijers and Schouten, we can interpret the argument as being either deductive or abductive.Footnote 10 In either case, both interpretations take it to be the supposed consensus of intuitions that gives the experience machine its strength. If the vast majority of people report preferring reality to a life in an experience machine, then hedonist theories of welfare are false (deductive version), or it is rational to believe that hedonist theories of welfare are false (abductive version). Both of these interpretations seem to capture the most orthodox understandings of Nozick's argument, although which of these Nozick himself would endorse is unclear. My opinion is that Nozick most likely meant the deductive version; simply for the fact that Nozick seems incredibly confident that his conclusion/position is correct.
Nevertheless, the deductive version is particularly problematic. Baber and Tännsjö both accuse the deductive version of presupposing some form of desire-satisfaction theory of welfare, and thus begging the question against hedonism.Footnote 11 Feldman and Hewitt criticize the move from ‘matters to us’ to ‘has value’, as what people value has little to no consequence for questions about what has intrinsic value.Footnote 12 Barber and Feldman also believe we could be mistaken in thinking that hedonism is mistaken.Footnote 13 And finally, ‘no careful philosopher should believe that widespread agreement between reasonable people on the matter of value deductively proves any particular conclusion’.Footnote 14
To circumvent these problems I shall ignore the deductive version. From here on, we will understand Nozick's argument as the abductive version: if the vast majority of people prefer reality to a life in an experience machine, then it is rational to believe that hedonist theories are false. Weijers and Schouten formulate the abductive version of Nozick's argument as follows:
AP1. In terms of the internal aspects of our experiences, an experience machine life would be much better than a life in reality (stipulated in thought-experiment).
AP2. When instructed to ignore their responsibilities to others, the vast majority of reasonable people report preferring reality over a life in an experience machine (empirical claim).
AP3. The best explanation for AP2 is that reality matters intrinsically to the vast majority of reasonable people.
AP4. Inference to the best explanation: If a hypothesis is the best explanation of an observation, then it is rational to believe that hypothesis is true (standard methodological premise).
AC1. Therefore, it is rational to believe that reality matters intrinsically to the vast majority of reasonable people (modus ponens AP3, AP4).
AP5. The best explanation for reality mattering intrinsically to the vast majority of reasonable people is that reality has intrinsic prudential value.
AP6. Inference to the best explanation (standard methodological premise).
AC2. Therefore, it is rational to believe that reality has intrinsic prudential value (modus ponens AP5, AP6).
AP7. If internalist prudential hedonism is true, then the internal aspects of pleasure and pain are the only things of intrinsic prudential value (or disvalue) in a life (stipulated definition).
AC3. Therefore, it is rational to believe that internalist prudential hedonism is false (modus tollens, AC2, AP7).Footnote 15
We can see from the described formulation that there are several ways in which one might undermine Nozick's argument. In this article, I am concerned only with AP2: the claim that the vast majority of people report preferring reality to a life in an experience machine.
II. THE CONFLICTING INTUITIONS PROBLEM
In understanding the anti-hedonist intuition as the linchpin to Nozick's argument, one particularly blunt way hedonists have responded is by simply denying the intuition. By choosing to plug in to the experience machine, Nozick's argument is severely undermined. After all, if the intuition itself is the evidence then without it there is no argument. One philosopher who has flatly rejected the ‘normal’ intuition is Tännsjö, who claims he would choose to plug in to the experience machine over staying in reality:
The point now is rather that, when confronted with [Nozick's] experience machine, we are supposed to think: it would not be wise to plug in . . . But if this is how we are to understand Nozick's thought-experiment, then I have to confess that I do not share Nozick's intuition about the case. I have already indicated that many people seem to be prepared to plug into Nozick's experience machine. I'm personally no different from them.Footnote 16
Tännsjö additionally points out that many people take drugs to avoid or escape reality, and that this is the ‘pharmacological equivalent’ of connecting to the experience machine.Footnote 17 But the abductive argument described in section I is resistant to such a challenge. The reason for such resilience is found in the clause ‘vast majority of reasonable people’, as seen in AP2. Such a clause avoids outliers, since those who do not personally share the intuition (such as Tännsjö) have ‘incorrect’ intuitions, as it seems reasonable to assume that one's personal intuition, if in conflict with the vast majority, is likely to be mistaken.
Yet even this more robust clause has been called into question: do the ‘vast majority of reasonable people’ really prefer reality to experience? Nozick merely assumes our intuition will be like his (anti-hedonist), but whether we actually do is, as noted by Weijers and Schouten, clearly an empirical claim. As such, a number of philosophers in recent times have conducted several surveys in order to see whether (at least according to their participants) Nozick's assumption maps reality. Unfortunately for Nozick (and his sympathizers), the results are not those desired. Rather than finding consensus that could be plausibly understood as a ‘vast majority’, intuitions actually conflict. The fact that intuitions conflict over EMTEs is used as ammunition against AP2.
Surveying how people respond to Nozick and other EMTE variants brings about the conflicting intuitions problem. The reasoning for variant EMTEs is something like this. If Nozick is right about what we value, then we ought to see people choose reality over an experience machine regardless of context. As Weijers and Schouten put it, ‘if other versions of the experience machine thought-experiment can equally or better isolate this prudential value comparison, and people's judgments about them significantly diverge, then either AP2 or AP3 might be false’.Footnote 18
Kolber, De Brigard and Weijers have all created variant EMTEs in their attempts to demonstrate the falsity of AP2.Footnote 19 For example, Kolber revises the ‘experience machine question’ (i.e. would you plug into an experience machine?) by asking, ‘would you get off an experience machine to which you are already connected?’.Footnote 20
De Brigard, along the same lines, describes a reverse experience machine in which a Mr Smith tells you that you have been mistakenly plugged into an experience machine:
All your experiences thus far have been nothing more than a computer simulation designed to give pleasurable experiences. You are then given a choice: you can either remain plugged into the experience machine or you can opt out and return to your old life.Footnote 21
But De Brigard does not stop there. He offers several vignettes that offer different information about life outside the experience machine. The negative vignette is that outside the machine you are in a maximum-security prison in West Virginia. The positive vignette is that outside the machine you are a multimillionaire living in Monaco. In the neutral vignette you are given no information about life outside the machine, whilst in the second neutral vignette you are only told that life outside the machine is nothing like life inside the machine. The reported results were as follows:
• Negative: 87% remained (13% disconnected).
• Positive: 50% remained (50% disconnected).
• Neutral: 46% remained (54% disconnected).
• Second neutral: 59% remained (41% disconnected).
Weijers gives a significantly detailed EMTE variant that I call the astable experience machine.Footnote 22 Here, Weijers explains that throughout Boris's life Boris has been switching between a real and a computer-simulated reality, spending half of his life in the machine and the other half outside of it. All the bad times that have occurred in Boris's life have happened while he was in reality, and all the good times happened inside the experience machine. Boris is then offered a permanent spot in an experience machine. Weijers explains:
If Boris accepts the spot, then he would stay in an Experience Machine permanently. If he rejects the spot, then he would never be offered a spot again. Boris’ life would be the same length in an Experience Machine as it would otherwise have been. Ignoring how Boris’ family, friends, any other dependents, and society in general might be affected, and assuming that Experience Machines always work perfectly, what is the best thing for Boris to do for himself in this situation?Footnote 23
Weijers reports that 55 per cent of participants (42/77) thought Boris ought to stay connected to the experience machine for the remainder of his life.Footnote 24 The reports from both De Brigard and Weijers demonstrate that there are conflicting intuitions amongst groups.
Yet different conclusions are drawn from the above data. For while De Brigard and Kolber seem to take it as a decisive defeater for Nozick's anti-hedonist position, Weijers takes a more modest approach, suggesting that experience machines ‘should no longer be considered to provide conclusive or even strong evidence that hedonism and all other internalist mental state theories of well-being are false’.Footnote 25 Whatever the case might be, the simple fact that there is no consensus of intuitions (let alone a consensus large enough to be described as a ‘vast majority’) means conflicting intuitions really are a problem for Nozick and his anti-hedonist sympathizers.
III. THE BIPARTITE DISTINCTION
In response to the criticisms of Kolber, De Brigard and Weijers (amongst others), Lin has argued that the experience machine gives us a pro tanto reason to reject hedonism.Footnote 26 He reasons that comparative welfare judgements about experientially identical lifetimes can demonstrate this:
Consider a case in which two lives are experientially and thus hedonically identical instead of a case in which the experience machine life is hedonistically superior . . . A spends his life in the real world, whereas the subject of B is plugged into an experience machine for his entire life . . . do A and B contain the same total amount of welfare?Footnote 27
Lin believes that many would share his comparative judgement that, ‘A is at least somewhat higher in total welfare than B’.Footnote 28 Further, such an intuition is corroborated by other concepts that link up with welfare, such as pity.Footnote 29
While I agree with Lin, I think comparative judgements can do more work for us. In this section, I suggest that comparative judgements reveal to us that welfare is two-dimensional. That is, we can intuitively carve up our welfare judgements along two distinct lines: how a person is ‘doing’, now that same person's life is ‘going’. I call this the bipartite distinction. This bipartite distinction can give us access to a linguistic framework that can help explain our welfare judgements on, not just comparative judgements, but the intuitions across EMTE cases. First, let us take a step back from experience machines and consider two people, both of whom are in reality. Consider the following:
Peter: Peter is happy, upbeat and in a good frame of mind.
Paul: Paul has a loving family, job security and lives comfortably.
Who is better off: Peter or Paul? No answer is immediately obvious. It seems reasonable to assume many would be conflicted about describing either as being better or worse off than the other. Even if we were to provide an answer, we would be both hesitant and unsure of it. The reason we feel conflicted, it seems, is because we do not have all the relevant facts. For that reason, the most obvious thing to do is simply refuse the cases as they stand. Suppose we are given more information:
Peter: Peter is happy, upbeat and generally in a good frame of mind. Yet events in Peter's life are poor, e.g. family death, relationship breakup, demotion at work.
Paul: Paul has a loving family, job security and lives comfortably. Yet Paul lacks fulfilment. Paul is not happy or satisfied.
I do not think this new information makes comparative welfare judgements any easier. But why? In one sense, who is better off seems indeterminate: we cannot say with any confidence who is better off than the other. Yet in another, it seems right to say that both are better off, and worse off, than the other, just in different ways. Adopting a two-dimensional characterization of welfare will give us a space in which we can make plausible and accurate welfare judgements. In the case of the second version we might say:
• Even though Peter's life is not going well he is still doing well.
• Paul's life is going well, but he is just not doing well.
Such judgements accurately capture our intuitions about what is good and bad for both Peter and Paul. Such judgements make sense to us because we comprehend both what is being said and described with doing and going. Our comparative welfare evaluation is that both are better off and worse off than the other in different respects. The bipartite distinction gives us the linguistic framework to make such judgements along two dimensions of assessment. For the purposes of this article, we need only a rough approximation of the bipartite distinction:
Doing-well: an individual is doing well if and only if, then only in so far as, their subjective experience and attitudes towards their lives are of a particular set or kind.
Going-well: an individual is going well if and only if, then only in so far as, a particular set or kind of life-circumstances obtain for them.
If we return to Lin's example, we now have a principled explanation why we take A to be better off than B: while both are doing just as well, A's life is going better than B. So A's welfare increases along both dimensions of welfare, whereas B's is only along one.
The labels I am using – ‘doing’ and ‘going’ – are often objected to because they strike many as confusing or counter-intuitive. Why not opt for describing each part as merely subjective/objective or internal/external? After all, doing-well seems to be about subjective or internal experiences, whereas going-well appears to track objective facts external to the agent. However, my choice to use these labels was deliberate and with good reason.
The first reason is that those more intuitive labels (e.g. subjective/objective, internal/external, etc.) are loaded terms. By loaded, I mean that people can have very specific understandings of those words and it can be difficult (if not impossible) to uncouple their preconceptions. So by using doing-well and going-well, I hope to avoid such preconceptions.
The second reason is that the bipartite distinction, as sketched here, is (as said above) only a rough approximation. Such a preliminary understanding is open to differing refinements, which all differ in their significance. For example, Kim draws a distinction between two different concepts of well-being, ‘narrow’ (just mental-states) and ‘wide’ (things that lie outside the mental scope).Footnote 30 But Kagan has argued for a distinction between ‘me’ (my mental states and bodily states) and ‘my life’ (states of affairs, life story, etc.).Footnote 31 Michalos, however, describes the difference as one between ‘actual conditions of that life’ and ‘how the conditions are perceived, what is thought and felt about those conditions, what is done, and finally, what consequences follow from all these inputs’.Footnote 32
Clearly the views of Kim, Kagan and Michalos all share similarities; they all divide well-being (or what is traditionally taken as well-being) into two separate parts. Yet what each dimension amounts to according to each refinement is different. The argument I make in this article does not depend on any particular refinement, so I need not commit to any of them to make my point. By keeping with the bipartite distinction and the related terms, we can abstract away from specific refinements and maintain a neutral stance.
Finally, how I am using doing and going is not without precedent, and there is extensive use of the terms within the literature. Kagan writes, ‘the thought that it comes to the same thing to ask how well a person is doing, and how well that person's life is going, is a quite familiar one’.Footnote 33 Sumner notes that, ‘common sense tells us that a person's welfare, or wellbeing, is a matter of how well she is doing, or how well off her life is going, or how well of she is’.Footnote 34 Ferkany, in following Sumner, uses the same language: ‘If we are not doing well (we are miserable or taking things badly), our life is going in a way badly for us; if things are going badly (e.g. our plans are being frustrated), this usually makes us unhappy. Nevertheless on commonsense understandings they can come apart.’Footnote 35 And finally, Alexandrova employs questions of how one is doing or going when putting forth a theory of welfare contextualism.Footnote 36
So, my reasons for employing the bipartite distinction and the labels doing and going are not spurious. Not only are there good reasons for avoiding the usual labels, but also precedence within the literature for use of my favoured terms.Footnote 37
Before continuing, I want to address one final concern. Upon first glance, one might take the bipartite distinction to be merely a form of an objective-list theory of well-being. I want to clarify that the bipartite distinction is not a theory of well-being. Rather, the bipartite distinction is a characterization of welfare. Since the bipartite distinction is not a theory of welfare itself, it is compatible with several different types of theories.
By characterization, I mean it sketches out the structure, whereas a theory describes the constituent, or constituents, therein. While particular types of objective-list theories (those adding mental states to the list) would be compatible with the bipartite distinction, they are not the only theories that can account for the two-dimensional view sketched out. A desire-satisfaction theorist might claim that the relevant states of affairs or mental states are those desired by the individual. A perfectionist, on the other hand, could argue that perfection requires both dimensions to be adequately fulfilled, or that the dimensions require the right kind of relationship.
Note, however, that the particular types of theories above do not belong to the same family of theories, yet share the same view about the structure of well-being (namely, that welfare is two-dimensional).Footnote 38 So it is possible to share a characterization of welfare while disagreeing on which theory is correct. The bipartite distinction is not an objective-list theory. Additionally, the distinction itself is compatible with different categories of theories. One can accept the bipartite distinction view of welfare without committing to a specific theory (such as objective-list theories).Footnote 39
IV. TRADE-OFFS AND PAY-OFFS
The bipartite distinction can do more for us than merely explain our intuitions about two examples involving comparative welfare judgements. In this section, I shall explain how the bipartite distinction can resolve the conflicting intuitions problem. Let us begin by employing the same linguistic framework I used earlier in section III, which is available only through the bipartite distinction. With that in mind, we can capture (a) what both intuitions map onto, and (b) what is correct about both pro-hedonist and anti-hedonist intuitions: those who get into the experience machine will be doing well even if their life is not going well, while those who refuse to enter the machine have a greater chance of going well, even if they might have sacrificed how well they do.
Such an expression offers a welfare judgement that neatly captures both pro-hedonist and anti-hedonist intuitions while maintaining the integrity of each. When we say a person in an experience machine will be doing well, we are saying they will have the right mental states while in the machine. And when we say their life will not be going well, we are admitting that, contrary to their personal perception, the person in the experience machine will only be experiencing as if states of affairs obtained, even when in reality they do not. After all, these are the reasons people choose to plug in (i.e. they want to do well) or not (i.e. to go well). The fact we are drawing a distinction between two different dimensions in welfare is enough evidence to demonstrate we are considering multiple inputs when making welfare judgements about EMTEs.
How does the bipartite distinction help resolve the conflicting intuitions problem? I suggest the argument that results from conflicting intuitions is not a problem concerning the nature of well-being itself, but rather a disagreement about where the greater welfare pay-off lies. That we weigh trade-offs between dimensions gains credence when we consider how sensitive welfare evaluations are to contextual factors, such as what is happening, or is most likely to happen, in the life outside or inside an experience machine. In welfare contexts in which we calculate a projection of increased welfare along both dimensions we will, in all likelihood, choose to stay unplugged. But in circumstances where such increases are unlikely or likely to go worse in the future, we will most likely plug in.
While Nozick's EMTE lacks detail about the welfare context, he does seem to be sensitive to the possibility of a simulated life being preferable to particular kinds of lives in reality (for example, a life of torture).Footnote 40 And that a simulated life can be better than a real life, depending on contextual factors, seems true. We can easily imagine circumstances that would dictate pro-hedonist judgements. For example, Belshaw points out cases where a person might be suffering a terminal illness or that a person's remaining years look somewhere between incredibly dull to excruciatingly painful. In such cases, Belshaw concludes, plugging into an experience machine does not seem unreasonable.Footnote 41
Conversely, we can easily motivate anti-hedonist intuitions when considering a person whose welfare looks particularly promising. Suppose someone is about to marry the person of their dreams, and get a promotion (and substantial pay raise) with their current employer. Under such circumstances, why would they get in an experience machine at all? The bipartite distinction provides a principled explanation due to its two-dimensional nature. If chances of increasing welfare along both dimensions are high, our intuitive judgement in all likelihood will be to not plug-in. When chances are low, we see in experience machines an opportunity to increase one dimension, and so our intuitive judgement, in all likelihood, is to plug-in.
But there is a third option that is constantly ignored in the literature: indecision. By indecision I mean those who cannot choose (whether they ought to, or even would) one way or another. Such indecision often results in one of three outcomes: (1) abstain, (2) reluctantly choose, or (3) choose at random. Why this group is ignored I do not know, so I can make no empirical claims about what kind of percentage might fall into this third category. Yet it seems to me the indecisive category exists and quite plausibly would take numbers away from the two primary options.Footnote 42
The bipartite distinction offers two explanations why people might choose this third position. First, people can be unsure as to what those estimated welfare projections might be, resulting in intuitive judgements stalling. Second, EMTEs ask us to increase one aspect of our welfare at the potential cost of another. But if increasing welfare results in only increasing one dimension and curtailing the other, then it is unclear that our welfare is increasing at all if we plug-in.
If we increase our doing-well at the expense of going-well, and have the option of increasing our going-well but not our doing-well, then both options result in approximately the same level of well-being. So indecisiveness exists because, in some cases, there is no right answer (or if there is one, it is entirely unclear what it is). The conflicting intuitions problem is resolved once we understand (i) the nature of welfare judgements and, (ii) how those judgements map onto the two-dimensional nature of welfare. Our welfare evaluations are context sensitive and their outcomes will be dependent upon how we calculate our future welfare gains (or loses) depending on those same contexts.
V. CONCLUSION
In this article I have argued that the bipartite distinction does serious explanatory work by showing why people have different intuitions on what they would do, or ought to do, with regard to EMTEs. I explained how comparative welfare judgements help give preliminary reason for thinking there is a difference of dimensions (doing-well and going-well), and that such a hypothesis gave a principled explanation of such judgements. From here, I argued conflicting intuitions about EMTEs occur because of, (a) where individuals see the greater welfare pay-off, (b) how we calculate welfare outcomes, and (c) the welfare context in which welfare evaluations are made. The conflicting intuitions problem was deflated, as it was not a conflict about the nature of welfare but rather a conflict on how people calculate welfare pay-offs.
On a final note, I would like to consider further research opportunities in light of the arguments made in this article. First, the next step for this project would be to test my hypothesis empirically. Adding a third option, as well as space for how certain participants’ intuitive judgements are, will shed further light on whether or not EMTEs still have relevance as an argument against hedonism. Second, as noted in section III, there are several refinements of the bipartite distinction, all of which differ in their significance. Further still, philosophers have drawn very different conclusions about what such a distinction means for the philosophy of well-being more generally. So we need to find the most plausible refinement of the two-dimensional characterization and carefully consider these upshots. And since I have only given a cursory argument for the bipartite distinction, a more robust explication and defence of it is required.Footnote 43
