Is it good for us if the different parts of our lives are connected to each other like the parts of a good story? Some philosophers have thought so (de Bres, 2018; Dorsey, 2015; Kauppinen, 2015; MacIntyre, 1981; Velleman, 1991), while others have firmly rejected it (Berg, 2023; Strawson, 2018). In this paper, I focus on the state-of-the-art anti-narrativist arguments Berg (2023) has recently presented in this journal. I argue that while she makes a good case that the best kind of lives for us do not revolve around a single project or theme, the best kind of narrativist views actually encourage us to pursue a variety of different projects, as long as they are mutually supportive. I claim that when interpreted in the most plausible way, prudentially good-making narrative coherence arises precisely out of this kind of unity in diversity. In course of making this case, I will introduce distinctions and arguments that should be of broader interest for debates about holism and atomism about well-being (Raibley, 2015), such as the difference between deliberative and good-making roles of holistic features of a life.

Let me spell this out a little bit more. Many people think that the value of a life for a person depends, in part, on its global features, and not just on the value of its parts considered in isolation. Here are two common contenders for value-relevant global features, formulated in broadest possible terms:

Coherence

Other things being equal, it is better for a subject if the different things they do over a lifetime somehow hang together so that they form a whole whose parts are mutually supportive, like the parts of a story that make most sense in the context of the whole.

Well-Roundedness

Other things being equal, it is better for a subject if the things they do somehow form a rich panoply that involves the use of diverse capacities in diverse activities.Footnote 1

On the face of it, there’s a tension between these two attractive theses, as vague as they are. Roughly speaking, Coherence seems to be linked with with focusing on one or a few related things, while Well-Roundedness seems to push for trying one’s hand at many things. So while both theses are prima facie appealing, perhaps one should give way for the other. Indeed, that is how I see Berg’s basic argument: given the appeal of Well-Roundedness, Coherence is of minor and local value at best. She presses the point that trying to maintain a narrative unity to our lives can be “stultifying and constraining” (2023, 638), making it too hard to change course and pushing us towards one-dimensional lives that lack refreshing variety. On the other hand, she also observes that if narrativists try to accommodate the value of well-roundedness by moderating their claims, the view threatens to become otiose, so that it “can’t tell us very much about how to live a good life that we can’t find in other theories of well-being” (ibid.).

This is a formidable challenge. Intuitions about the value of diverse goods in our lives are compelling. Can Coherentists make sense of them without giving up on their core claims about the value of having a prudentially good narrative structure in our lives? I will argue that they can. I’ll begin by sketching the version of narrativist Coherentism that is Berg’s main target, namely Relationism, and setting some ground rules. In Sect. 2, I’ll introduce Berg’s arguments. I emphasize that in addition to the evaluative thesis of Relationism, she implicitly also argues against a related decision procedure, Deliberative Narrativism, which says that we should consciously try to LIVE A GREAT STORY, as she reports being advised by an Instagrammer (2023, 637).

I’ll begin my response to her in Sect. 3 by accepting that Deliberative Narrativism is indeed problematic, but observe that Relationism is not committed to it, any more than consequentialism is committed to agents consciously trying to maximize the good. Instead, the (partial) prudential decision procedure that best fits Relationism is what we might call Indirect Relationism, which tells us to place deliberative weight on features of our lives that de re contribute to a good narrative structure. Given what sophisticated Relationist views say about good life histories, this means that an Indirect Relationist decision procedure will recommend not only considering existing commitments to things of genuine value, but also branching out towards valuable new opportunities that complement our existing pursuits in order to avoid repetition, stagnation, and missing out on valuable opportunities. Finally, in Sect. 4, I address Berg’s argument directly against Relationism. I point out that her thought experiments involve lives that differ not only in narrative terms but also in terms of other goods. If we construct a pair of more similar comparison cases, we can see that Relationist versions of Coherence remain plausible even if we take into account values like spontaneity and variety. In sum, Berg’s challenges can be met, because what makes a life positively coherent is not single-minded focus but suitable teleological connections among a diverse array of pursuits. Such connections add to the prudential value of the life over and above the values of its constituent parts considered in isolation.

1 Well-being and life stories

Let’s start by fleshing out Coherence a bit more to understand better how it might be in tension with Well-Roundedness. The version that Berg rightly focuses on appeals to narrative unity. After all, what stories do is relate a number of different events so that they form an intelligible whole whose parts make reference to each other. In the literature, there are two common ways to cash out this idea:

Recountism

People make their lives coherent by recounting them to themselves and others, thus forging a unity among disparate states and events, which may then guide their future choices in a direction that makes sense in light of the past. Coherence understood in this way is good for a person.

Relationism

People’s lives are coherent to the extent that the different things they do and experience at different times positively inform each other, so that their earlier pursuits contribute to later ones and later ones complete and complement earlier ones. If a subject recounts her life, she may find a narratable unity among disparate states and events. Coherence in this sense is in itself good for a person.

Something like recountism has been defended by Connie Rosati (2013) and Helena de Bres (2018) (from whom I borrow the labels). It is related to but distinct from the descriptive claim that people create an identity that provides unity and purpose for their lives by constructing a life story (e.g. Taylor, 1989; McAdams & McLean, 2013). Relationism has been defended by David Velleman (1991) and myself (2012, 2015). Berg’s target is the latter, so I will also exclusively focus on it as the relevant interpretation of Coherence. One reason to do so is that it’s not clear whether Recountists are really claiming that telling a coherent life story is in itself good for a person—both Rosati and de Bres highlight how such storytelling contributes to other goods that are plausible candidates for well-being, such as a sense of self-worth (Rosati, 2013, 45) or understanding and community (de Bres, 2018).

Relationalists, on the other hand, do hold that coherence in their sense is a welfare good among others, and thus something that allegedly benefits us apart from its possible consequences for other things. Consequently, they claim that it is sometimes better for a person to lead a life with a prudentially good story rather than a prudentially bad story, even if the latter contains more of other welfare goods.Footnote 2 To assess such claims, we evidently need some principles for prudentially (rather than, say, aesthetically) better or worse life stories—or perhaps better, life histories, since we’re talking about relationships among narratable events in a life rather than as narratives told about it.

What might such principles be? I will take it for granted here that stories are about what happens when agents pursue some relatively high-level aim or apparent good over time, as it is common for both philosophers and narratologists to think (Herman, 2004; Kauppinen, 2012; MacIntyre, 1981; Taylor, 1989). For short, stories are about the fate of our projects, which is always more or less vulnerable to luck. A fairly obvious evaluative principle is that an individual project goes well for the agent in story terms when the aim is worth pursuing and the pursuit is successful, and badly when either or both of these conditions are not met. But this is not a particularly controversial claim, and something like it is defended by many achievementists about well-being (such as Portmore, 2008). More distinctive and therefore interesting narrativist claims concern large-scale relationships among our various ground projects, the kind of pursuits that give meaning and direction to our lives. They do not, after all, necessarily link up to form a larger life story, as Galen Strawson (2004) emphasizes. Of course, in a looser sense, everybody’s life can be related in the form of a story, but this can involve a kind of distortion, retrospectively forging connections where they don’t exist. But sometimes they do. Given that stories are about what happens in the pursuit of aims, for earlier and later projects to genuinely form parts of a larger story, they must stand in some sort of teleological relationship—maybe failure in one pursuit is the stimulus for another, for example, or the pursuit of one aim facilitates the pursuit of another. That way a fitting account of the person’s life won’t be just a list of the various things they’ve done (“first he was a soldier, and then became a teacher”) but has the shape of a proper story (“first he was a soldier, and because of what he experienced, he decided to become a teacher”).

Here it is important to bear in mind that even if one’s various pursuits do link up teleologically, the resulting story need not be a prudentially good one: maybe something stupid I did in the past causes me to get fired now. It doesn’t seem that I’m any better off than if I’m fired for some other reason. Therefore, Relationalists should not and do not say that simply having any kind of narrative cohesion in a life is prudentially good (although Berg seems at times to attribute this view to themFootnote 3). The kind of coherence that they defend is what we might label ‘positive coherence’, which I gestured at by talking about ‘mutual support’ among earlier and later parts of our lives in my definition of Coherence earlier.

So what Relationalists do can be described as spelling out principles for specific teleological relationships among projects that make for a positively coherent life. Here are three examples of such principles that will be relevant to evaluating Berg’s arguments. First, here’s a commonly endorsed principle:

Make the Past Serve the Future

Other things being equal, it is better for a subject if her later success is because of earlier efforts, whether or not they were successful on their own terms.

In support of this principle, Velleman discusses the case of a politician whose years in the wilderness eventually pay off in terms of an election victory, contrasting it with an alternative scenario in which their change in fortune comes about as a result of winning the lottery (1991, 53–54). He also talks about how learning from an earlier misfortune redeems it and thus enhances its value for one’s whole life well-being, so that one is better off than learning the same lesson independently of one’s past failure (ibid., 54–55). In terms that MacIntyre (1981) or Taylor (1989) might use, there’s an overall movement of one’s life in the direction of the good. In Kauppinen (2012), I put this by saying that both of these patterns contribute to a prudentially valuable progressive teleological shape in one’s life.

If reward and redemption involve making the past mean something positive for the future, the second common Relationalist principle tells us not to undermine the significance of what we’ve already done:

Don’t Tarnish the Past

Other things being equal, it is better for a subject to stick to a commitment to something of genuine value rather than switch to pursuing some other good, when doing so would defeat the purpose of their past efforts.

A good example of the value of sticking to a commitment is Elizabeth Anderson’s case of a couple who hold onto a family restaurant they’ve gradually built up rather than selling it to a big company that would franchise it all over the country. While selling would give them the financial security that would allow them to undertake other valuable projects, it would also “undermine the point of their lives’ personal investments and struggles, which were aimed not just at making money but at creating an alternative to the humdrum, homogenized, and predictable chain restaurants taking over the area” (Anderson, 1993, 34).

The third Relationalist principle for positive coherence emphasizes that there are many different values with respect to which we can make narrative progress:

Diversity Beats Single-Mindedness

Other things being equal, it is better for a subject if her life contains mutually supportive projects in pursuit of many different valuable aims that build on one another than if it contains a single project in pursuit of one valuable aim.

While a few narrativists, such as MacIntyre (1981), do speak of a good life story in terms of a singular quest, others, in particular myself (2012), highlight the value of having many mutually supportive projects. I argue that as long as there is a variety of valuable goals whose achievement requires different capacities, dedicating everything to a single purpose makes for a less than ideal life story (2012, 367–368). A life dedicated to a single aim can, to be sure, be positively coherent in one way: the various actions and events that comprise it can be unified by way of promoting the realization of the one grand aim in one way or another. But at the same time, it will, trivially, lack the kind of coherence among different ground projects that is possible when they’re mutually supportive. For example, serving in the military may involve acquiring skills and undergoing experiences that later improve one’s success as a teacher, generating large-scale positive coherence in addition to possible positive coherence among actions and events within each component pursuit. Diversity Beats Single-Mindedness can thus be seen as an application of the principle of organic unity that Robert Nozick defends for intrinsic value: the value of a complex object like a life is enhanced both by diversity of the material that is unified and by diversity of the unifying relations themselves (Nozick, 1981, 415–416).

Further support for Diversity Beats Single-Mindedness can be derived from the idea that a good narrative shape of a life involves overall progress towards the good, if we accept the plausible claim that there are many valuable aims and many kinds of value. Again, one way to make such progress is moving toward the single valuable aim one is devoted to. But if it is possible to move toward a variety of valuable ends at a time or over time in a coherent fashion, one can make even more valuable overall progress towards the good. This helps make sense of why it’s bad in terms of a life story to keep repeating more of the same kind of thing (even if one is successful), or to persist in stagnant projects that promise to yield little further payoff, or to miss out on valuable opportunities that one could have pursued without giving up one’s existing commitments. In all of these cases, one could make more overall narrative progress (and thus enhance positive coherence) by engaging with new valuable aims that complement earlier ones rather than sticking with the old (even though it might maximize local positive coherence, as it were).

Finally, before discussing Berg’s anti-relationalist arguments, I want to establish some ground rules that derive from the other-things-being-equal nature of the claims about the value of global features. After all, nobody claims that (positive) Coherence or Well-Roundedness is the only good-making feature of a life. For example, it’s no counterexample to Well-Roundedness if a really successful life centered around the opera is better than some more well-rounded life, since other goods may outweigh the value of having diverse pursuits. Of course, if just any amount of other goods can outweigh Coherence or Well-Roundedness, they’re of little theoretical interest and of no practical relevance, since no two actual lives will be exactly the same in other respects. So the theses should really say that if two lives are roughly equal in other good-making features, a more coherent or well-rounded one is better. This gives rise to what I’ll call.

The Standard Operating Procedure (SOP)

To test whether a feature makes a meaningful difference to the value of a life for a subject, compare two scenarios that differ in terms of that feature, but are otherwise roughly equal in terms of good-for-making (and bad-for-making) features.

The reason I highlight SOP is that Berg, surprisingly, doesn’t directly employ it, as I will argue. This is going to be relevant to assessing some of her arguments.

2 Berg vs. Narrativism

So let’s turn to Berg’s paper. Officially, her target is what she calls Narrativism about well-being, or the thesis that “The narrative structure of a person’s life plays a major role in determining their well-being over time” (Berg, 2023, 638). Since she construes narrative structure in Relationist rather than Recountist terms, I will continue to use Relationism as the label for her main target. However, although she’s for the most part not explicit about it, she also has a secondary target, the role of narrative in prudential decision procedures. I will characterize this target as follows:

Deliberative Narrativism

When making life choices, we should consider their impact on narrative coherence (de dicto) and give weight to it in our deliberation.

Towards the end of her paper, Berg does distinguish between narrativism as an evaluative criterion and as a decision procedure (Berg, 2023, 655). While she unambiguously rejects both, she doesn’t explicitly say which of her arguments target Relationism and which target Deliberative Narrativism, so I will have to engage in some rational reconstruction below.

To begin with Berg’s case against Relationism, then, I will focus here solely on her most elaborate thought experiment and her analysis of it, since she acknowledges that Relationists may have a way around some of her other cases (e.g. Berg, 2023, 642). The thought experiment features three characters, Andy, Becca, and Crystal, whose lives differ from each other in many ways, containing different goods in addition to having a different degree of coherence. (I’ll be addressing this departure from SOP soon.) Andy’s life is the most strongly narratively unified, while Crystal’s is the least unified and most diverse; Becca is somewhere in between. I will set Becca aside, since the comparison between Andy’s and Crystal’s lives is decisive. Here’s how Berg describes them:

Andy

Andy works on the docks from the time he’s young, starting as a stevedore and moving on up. Over time, he becomes involved in union politics, eventually becoming the president of his local. As president, he develops a reputation for candor and incorruptibility; he also develops an independent interest in the history of the labor movement, which helps him become a savvy political strategist. He marries the president of the city’s teachers’ union, and together they work together to negotiate a new, fairer contract with the city. Because of this success, he’s eventually elected president of the national longshoremen’s union, and he uses this position to advocate for a new workers’-rights bill, which passes Congress just before he retires. Andy spends his retirement contentedly reflecting on his hard work and enjoying spending time with his grandchildren. (Berg, 2023, 650–651)

Crystal

In Crystal’s view, the main point of life is to try everything once. Over the years, she has a number of jobs, from social worker to barista to car salesperson. She’s talented enough at all of them, and she appreciates the distinctive skills she learns at each (appreciating others’ lives and struggles, knowing how to roast coffee perfectly, honing her powers of persuasion), but she doesn’t want to be tied down to just one career. Crystal has a few consistent hobbies; she always has a cat or two at home, and she loves to pick out a tune on her banjo. She watches documentaries about all sorts of things: astronomy for a while, then philosophy, then art history. She’s dedicated to each pursuit for as long as she does it; she thinks it’s important to throw herself into all that life has to offer. Early in her life, Crystal decided marriage and kids weren’t for her; she’s a dedicated serial monogamist, entering a variety of short- and long-term relationships over the years. These relationships have their ups and downs, as all relationships do, but Crystal is generally happy in them. (Berg, 2023, 651–652)

Berg’s argument is straightforward. According to her, first, Relationists are committed to saying that Andy’s life is clearly better for him than Crystal’s is for her, since her life is “very low on thematic consistency and cohesion”, even though it’s “higher on diversity” (Berg, 2023, 652). Second, that comparative judgment is false: Crystal’s life can be just as good for her as Andy’s is for him, considering that Andy’s life is “missing adventure, novelty, and spontaneity” (ibid., 653), which hers contains. More generally, Berg emphasizes that while both lives contain many of the same objectively valuable goods, such as personal relationships and concern for others, “people structure those goods in many different ways, and narrativism unduly constrains which lives count as good ones” (ibid., 652).

Berg’s first challenge to Relationism, then, is that it elevates narrative coherence above a kind of well-roundedness, leading a life that contains a variety of different goods (and different tokens of goods like relationships). But as I suggested, some of her arguments are best understood as targeting a different quarry, namely Deliberative Narrativism. Consider her case of Michael the chemistry teacher:

Michael

Michael just graduated from college and started a job as a middle-school science teacher. Now he has to figure out how he’s going to spend the rest of his time. He wants to help out in his community, and he’s deciding between tutoring high-schoolers or working at the soup kitchen. He wants to do more than just watch TV in the evenings, and he’s deciding between reading up on bioethics or getting into graphic novels. He wants to meet new people, and he’s deciding between joining a chemistry-themed meetup group or trying out adult kickball. (Berg, 2023, 643–644)

Here Michael faces a series of choices about using his free time, each of which involves doing either something chemistry- or teaching-related or something unrelated. Berg argues that if Michael makes choices on the basis of what he expects to make for “thematic consistency” (Berg, 2023, 644) in his life, he is likely to end up with a one-dimensional life that lacks important goods like diversity, novelty, and resilience, because his concern for narrative leads him to make a chemistry-related choice each time. Basically, if he aims to make his life be about chemistry, and thus more narratively unified, he makes his life worse. As she puts it in her introduction to an online newsletter discussing her paper, “If we try to turn our lives into good stories, we may find ourselves making choices that are bad for us”Footnote 4 (my emphasis). Hence, she concludes, Deliberative Narrativism is a bad prudential decision-making procedure.

3 Indirect Relationism as a prudential decision procedure

Although Relationism is Berg’s explicit primary target, it will prove useful to begin responding to her critique by focusing on Deliberative Narrativism, since doing so involves clarifying what Relationism is committed to. I will grant straight away that Berg is right in observing that Deliberative Narrativism is problematic. But what I want to emphasize, first, is that it is logically independent from Coherence, and more specifically Relationism. Relationism is simply a claim about prudential value, not decision procedures. As such, it doesn’t say anything about how Michael or anyone should go about making his mind—it doesn’t say that people should consciously try to LIVE A GREAT STORY (Berg, 2023, 637). Indeed, Relationists have noted that consciously trying to live a great story is in fact likely to lead people astray if their conception of a prudentially good life story is mistaken, as happens when Madame Bovary’s romantic dreams come crashing down (Kauppinen, 2021, 105).

To be sure, even if this is granted, it’s fair for Berg to demand that Relationalists offer a plausible model of good prudential deliberation that is responsive to narrative value without explicitly thinking in these terms. And she’s right in thinking this challenge hasn’t been met so far. My own view is that the situation is parallel to Act Consequentialism and decision procedures (see Railton, 1986 for the classic account): we should employ whichever method leads us to give just the right weight to narrative features. Consciously thinking about what would make for the best story is unlikely to be that method. If so, Relationism will favor some form of Indirect Relationism as a decision procedure, just as many consequentialists favor indirect consequentialism. Such a method will involve sensitivity to what makes for a good (or great) story de re—that is, the kind of features that actually contribute to Coherence in one’s life—rather than de dicto, or trying to lead a life one thinks would make for a prudentially good story (under that very description).Footnote 5

What would Indirect Relationism tell us to take into account, then? My aim in this paper is not to give a full account of it, but I will mention a few elements that will plausibly play a role in it. First, recall that Relationism says that prudentially good life histories are shaped by how we exercise our agency in pursuit of something objectively good and how later pursuits build on or complete earlier ones, as Make the Past Serve the Future and Don’t Tarnish the Past emphasize. Given such principles, one deliberative strategy that is likely to promote positive coherence is giving weight to existing commitments and values that go beyond what we think would yield positive experience for us. Since commitments and values will have been shaped by our past choices and experiences, we don’t need to explicitly think about what our future actions might mean for the significance of past ones. We will nevertheless be less likely to abandon our partners or ground projects at the first hint of trouble, for example. Of course, since Relationism doesn’t say that the right kind of narrative structure is the only good thing in life, the decision procedure it recommends doesn’t tell us to pursue these values any cost, or to never give up commitments, but only to take special care before doing so.

At the same time, the most plausible forms of Relationism are wary of repetitiveness and single-mindedness, as I noted above. To take into account Diversity Beats Single-Mindedness, Indirect Relationism will presumably say that we should be on the lookout for opportunities to pursue new valuable aims, at least when doing so is consistent with existing commitments. To avoid stagnation, it will even recommend giving up pursuing aims that have shaped our life up to now, if there’s little room for further progress left and new alternatives are available. And even if one is in a good position to realize some type of valuable aim once again, Indirect Relationism will recommend considering something new that builds on what one has been doing, so that one avoids repetition.

To go back to Berg’s Michael, how would Indirect Relationism have him deliberate about his free time? Well, it will tell him to give a lot of weight to his commitments, and ensure that whatever he does during his free time will not undermine his ability to do a good job as a teacher, now that he has trained for years for that very purpose, which has a lot of mileage left in it. This, as such, doesn’t rule out either tutoring high-schoolers or working at a soup kitchen, for example. It might slightly tell in favor of the chemistry-related activities, since engaging in them somewhat increases the payoff of his past studies. But at the same time, anti-single-mindedness considerations tell in favor of Michael trying out something new—branching out to new projects that are consistent with his commitments, but complement what he has already done or been doing. This, as such, doesn’t entail that Michael should take up work at the soup kitchen or try out adult kickball, since it’s a claim about a kind of pattern that one should take into account in one’s choices, rather than particular choices. But it does favor Michael giving priority to some non-chemistry related options—if he’s going to choose tutoring high-schoolers in chemistry over working at the soup kitchen, maybe he should choose kickball over the chemistry meet-up to avoid his story becoming too repetitive or narrowly focused.

Thus, Indirect Relationism seems to yield the right recommendation for Michael. Indeed, it gives the sort of guidance that non-narrativist views that ignore the overall pattern in one’s choices across time cannot give. So while Berg is right to reject Deliberative Narrativism, Relationism offers a compelling rationale for a replacement decision procedure that yields intuitively plausible results.

4 Testing for the value of coherence

Let us then turn to Berg’s main target, Coherence as an evaluative criterion. Does the case of Andy and Crystal show that Coherence (more specifically, Relationism) is false? Here’s why I don’t think so. First, and most obviously, the mere fact that Crystal can lead a good life with low narrative coherence (which it indeed does have, on a natural reading) doesn’t mean that coherence isn’t a significant good-making feature, as long as enough of other good-making features are present in her life. Second, when comparing Andy’s and Crystal’s lives, were not just comparing lives with different levels of coherence, but, by Berg’s stipulation, lives that contain different kinds of other goods, like adventure, novelty, and commitment. But this is where SOP is really important. Without following it, we can’t isolate the potential effect of coherence, so we can’t conclude that it doesn’t make a difference (or that it does). Note also that high coherence as such doesn’t entail the absence of adventure and novelty—think of the life of an explorer or a movie director or an author like Ernest Hemingway or Martha Gellhorn. Such lives can contain a lot of novelty and adventure while nevertheless hanging together nicely. So we need to look beyond Berg’s descriptions of Andy and Crystal to really decide what we should think of Relationism.

To achieve this, let us turn to the Standard Operating Procedure. Berg is definitely correct in holding that Relationism entails that Crystal’s life would be better for her if it were more like Andy’s life, while being otherwise roughly similar—that is to say, if it contained pretty much the same other goods, but had a better narrative structure by plausible Relationist criteria. To settle whether this is indeed correct, we need to compare two Crystals, whom I’ll call Scattered Crystal and Coherent Crystal. Because Berg’s scenario is so long and complex, I’m going to focus on a couple of aspects of their lives and simplify things as far as possible. Here we go:

Scattered Crystal

Scattered Crystal has worked as a social worker, barista, and then as a car salesperson. She learned distinctive skills she learns at each (appreciating others’ lives and struggles, knowing how to roast coffee perfectly, honing her powers of persuasion), but what she learned and did as a social worker in no way informed what she did as a barista – she might as well never have been one – and doesn’t make any difference to how she sells cars. Still, she’s pretty good at each separate activity. She also watches documentaries about all sorts of things: astronomy for a while, then philosophy, then art history. She’s dedicated to each pursuit for as long as she does it, but then completely forgets about it. Having done philosophy gives her no insight into art history, nor does it lead her to rethink anything about astronomy. The different times in and aspects of her life are thus wholly independent and isolated from each other.

Coherent Crystal

Coherent Crystal has worked as a social worker, barista, and then as a car salesperson. She learned distinctive skills she learns at each (appreciating others’ lives and struggles, knowing how to roast coffee perfectly, honing her powers of persuasion), and what she learned and done as a social worker significantly informed what she did as a barista – she was able to relate to certain customers better and served them better as a result – and also helps her make conscientious choices about when to make a hard sell in the car trade. She watches documentaries about all sorts of things: astronomy for a while, then philosophy, then art history. She’s dedicated to each pursuit for as long as she does it, but each is also shaped by what she’s done before and the other things that are going on in her life right now. Having done philosophy gives her insights into art history, and her later interests lead her to appreciate new things about astronomy. Different times in and aspects of her life thus mutually inform each other, so that earlier pursuits shape later ones and later ones complete and complement earlier ones.

It seems clear that Coherent Crystal is doing better than Scattered Crystal. Why? Relationism gives a simple answer: the diverse pursuits in her life are positively coherent in virtue of bearing fruit for each other. If this is the best explanation of why Coherent Crystal is better off, considering variations of Berg’s original case supports rather than undermines Relationism. To be sure, we must be careful to bear in mind that it’s easy to misread these vignettes so that Scattered Crystal does worse than Coherent Crystal at the activities they engage in. So we should imagine that while Coherent Crystal does better as a social worker in virtue of having been a barista than she would otherwise have, Scattered Crystal does just as well as a social worker for some different reason—though her past activities don’t have any impact on her social work, she may have, say, more innate talent for it to compensate for the lack of inter-project connection. Even so, it seems to me that Coherent Crystal is better off than Scattered Crystal. This is parallel to how the person whose past failure teaches them a hard lesson is better off than someone whose failure teaches them nothing, but who learns the same lesson by testimony, as Velleman (1991) noted in defending what I’m calling Make the Past Serve the Future.

Here it is natural to ask for a deeper explanation of why it would be in itself good for us for our lives to be coherent rather than scattered. I offer my favorite kind of explanation in Kauppinen (2020), where I argue along Aristotelian perfectionist lines that what’s good for us depends on our nature, and that it’s an important aspect of our nature that we’re temporally extended agents rather than a series of momentary selves. When our earlier and later pursuits are mutually supportive, we fulfill our nature more successfully than we would if they were scattered, in which case they might as well be performed by distinct selves. Coherent activities are good for me, not just for me-right-now. Velleman’s (1991) alternative take is based on his distinction between diachronic and synchronic well-being (see note 2 above). For him, the well-being of any creature is relative to their own point of view, and it is part of our nature that we take two different perspectives on our lives, a momentary one and a whole-life-one. He would presumably hold that while Coherent Crystal and Scattered Crystal are equal in terms of synchronic well-being, Coherent Crystal rates higher on diachronic well-being in virtue of her life having higher Coherence, and is thus overall better off.

With this in mind, we can return to Berg’s original Andy and Crystal. Berg’s judgment was that in spite of narrative differences, neither life is clearly prudentially better. One lesson of my Coherent Crystal and Scattered Crystal scenarios is that it’s not obvious what Relationism would say about Andy and the original Crystal. If Berg’s Crystal is like Coherent Crystal, which she might well be, her life story will rate fairly high on Coherence, even if not quite as high or for the same reasons as Andy’s. In that case, Relationism doesn’t entail that Andy’s life is clearly prudentially better. (As Berg acknowledges, it might be better at the end of the day, but not clearly better.) If, on the other hand, Berg’s Crystal is like Scattered Crystal, we might well feel that what Andy’s life has a significant good-making feature that her life lacks. Still, let’s recall that Berg stipulates that her life has “adventure, novelty, and spontaneity” (2023, 653), which Andy’s life lacks. Insofar as we think that these things have non-narrative value, they might compensate, at least in part, for reduced narrative value. Thus, even if Relationism is true, even Scattered Crystal’s life might not be clearly worse than Andy’s. This naturally depends on whether Andy has some other goods in his life that compensate for the absence of adventure. One plausible key feature is how Andy himself feels about his life—while Berg describes Crystal as “happy” in her many relationships, all that she says about Andy’s psychology is that he’s “contented” in his retirement. If we suppose that far from being bored with his life’s work in the union, Andy is gripped and excited by it throughout and proud of his achievements, it certainly begins to look to me like he's doing better than Scattered Crystal, all things considered, and indeed even clearly better.

But again, all that Relationism is committed to is that if Andy’s and Scattered Crystal’s lives are roughly equal in terms of other goods, Andy’s life goes better for him than Crystal’s for her. In contrast, on Berg’s view, Andy’s and Scattered Crystal’s lives are on a par, prudentially speaking, even if Andy’s significantly more coherent life contains a roughly equal amount of happiness, adventure, and novelty. Here we should again bear in mind that positive coherence is compatible with having diverse and exciting projects—after all, Andy’s work in the union and in national politics might take him to all sorts of places doing all sorts of things. For Berg, Andy still wouldn’t be doing clearly better, because for her positive coherence has very little or no value. I submit that the more scattered we make our description of Crystal and the more equal the description of non-narrative goods, the less plausible this no-difference claim is.

5 Conclusion: the unity of unity and diversity

It is very much to Berg’s credit that she forces narrativists to explain why their view doesn’t push us towards a one-dimensional and impoverished life in which everything revolves around a single theme. But I’ve argued that this challenge can be met. Coherence and Well-Roundedness only seem to be in tension with each other, especially if we understand the former in narrativist terms. As I noted above, some narrativists, like myself (2012), have explicitly argued against repetitive and single-minded pursuits while nevertheless highlighting how diverse activities can complete and complement each other. To put the point differently, the kind of narrative coherence among ground projects that confers additional prudential value to a life is unity in diversity, which is better promoted if there is unity among diverse ground projects as well as within each. It’s the kind of thing that Nozick highlights in his discussion of organic unity when he says that “The more diverse the material that gets unified (to a certain degree), the greater the value” (1981, 416). I believe that once we understand Coherence along the lines that contemporary Relationists have defended, the attractiveness of a well-rounded life doesn’t count against the importance of a prudentially good life story, but rather provides further support for it.