EVOLUTIONARY PSYCHOLOGISTS’ ACCOUNTING of the costs of social change in the form of happiness and freedom is, we have seen, both incomplete and biased. But there is a deeper problem. The very conceptions of happiness and freedom in their discussions of the costs of change are narrow ones that fail to take account of many important aspects of human well-being and autonomy (Kelman 2005).
The obvious place to start in assessing people’s happiness in a particular society is to ask them how happy they are. But this approach has well-known weaknesses. Critics of welfare economics have noted the problem of “adaptive preferences”—that people modify their preferences and their standards of happiness to accord with their expectations (Gasper 2005; Giovanola 2005; Teschl and Comim 2005). Thus, people with lives that seem to be severely deprived by any objective standard may report that they are fairly happy and that their preferences are being met quite well simply because they have relinquished desires that they have come to regard as unrealistic. (People whose lives are exceptionally well provided, on the other hand, may be frustrated by their inability to fulfill overblown expectations and so report themselves to be unhappy.) Describing the phenomenon this way suggests that people whose lives are objectively deprived know more or less what they are missing though they have given up on it. But sometimes things are even more difficult for those seeking a happiness tally, and people’s life circumstances are such that they really have no conception of what other, better lives might be possible for them.
One response to the problem of adaptive preferences is to replace the measures of subjective happiness or preference satisfaction with objective measures of people’s ability to do things. This “capability approach,” developed in different forms by economist-philosopher Amartya Sen and philosopher Martha Nussbaum, provides a way of assessing the quality of people’s lives even when their own ability to judge their life quality has been compromised (Alkire 2002; Comim 2005; Gasper 2005; Giovanola 2005; Nussbaum 1995; Sen and Nussbaum 1993). The response perspective on development and behavior adds a further twist to the puzzle, however. It suggests that we need to consider not just the way that people may “downgrade” their preferences quantitatively in keeping with diminished expectations but also the likelihood that the different environments people inhabit may elicit qualitative variation in the desires that they come to possess. People sometimes want less because they expect less, but they may also want differently because they live differently.
Such differences could themselves be part of an evolved responsive human nature. When Sen talks about “adaptive preferences,” he means that shifting preferences help people to adapt their attitudes to their circumstances, but shifting preference may be adaptive in another—evolutionary—sense as well. Some environmentally triggered differences in people’s preference profiles may be evolved adaptations to particular kinds of environments. For example, Gowaty’s “sexual strategies” hypothesis suggests that environmental factors may affect the extent to which both men and women develop behavioral tendencies and preferences that match common gender stereotypes (Gowaty 2003, 2008, 2011; Gowaty et al. 2007; Gowaty and Hubbell 2013). Just how strongly both women and men want to spend time with their children (or to pursue careers in the sciences) may depend on subtle environmental indicators of women’s reproductive autonomy. Indeed, how strongly women want reproductive autonomy itself may depend on the indicators of that autonomy.
Other instances are equally suggestive. New thinking about the adaptive role of “risky behaviors” suggests that environmental factors may cue the adoption of life-cycle strategies that affect people’s preferences regarding risk taking of many kinds (Chisholm 1999; Ellis et al. 2012; Wilson and Daly 1997), while recent work on the adaptive significance of bullying suggests that the kinds of peer relationships that people seek may be cued by background features of their social environment (Volk et al. 2012). It is worth looking at these instances a little more closely.
Risky behaviors are those that are likely to lead to harm for those who engage in them—typical examples include drinking and drug use, sexual promiscuity, petty crime and gang membership, and the like. It is well known that some social environments are much more conducive to risky behaviors among young people than are others. The conventional explanation of this pattern, the “developmental psychopathology” model, is a conservative inter-actionist explanation. It assumes that the normal developmental trajectory is one that leads via a well-adjusted childhood and adolescence to a responsible adulthood but that pathologies such as antisocial behavior or premature sexual activity can result when conditions in the family and local social environment do not meet the conditions required for normal development. The evolutionary explanation, instead, sees children as responding to their environments by “choosing” one of two available developmental strategies (although no conscious choice, of course, is presumed to be involved). If the environment provides cues indicating good prospects for a long and healthy life, the child develops according to a strategy that can take advantage of those conditions—deferring reproduction in order to achieve healthy maturity, high social status in adulthood, and advantageous mating, thus optimizing the prospects for successfully rearing high-quality off spring. But if environmental cues indicate that the prospects for a long and healthy life are poor, the child develops instead according to a “live fast, die young” strategy evolved to promote reproductive success despite environmental challenges—pursuing early mating and high social dominance in adolescence irrespective of the negative longer-term effects that are likely to follow (Amin and Thompson 2001; Chisholm 1999; Ellis et al. 2012). What kinds of cues are relevant? Research on the age at which women first become pregnant shows a wide range of factors. Girls who were separated from their mothers for long periods in infancy, whose fathers are absent, or who moved from one home to another repeatedly in early childhood tend both to mature earlier and to become pregnant earlier than their peers do (Nettle, Coall and Dickins 2011). Similarly, the conventional explanation for bullying understands it either as a pathological result of a lack of social skills brought about by a social environment inadequate to the needs of normal development or as learned behavior impressed upon developing children by bullying parents. The evolutionary explanation sees bullying instead as an adaptive response to particular cues in the social environment: a means of achieving social dominance in adolescence under conditions that make this strategy likely to succeed. Conventional explanations in both these cases see some desires—or some kinds of happiness—as proper expressions of human nature, others as distortions produced by substandard environments. The desire to find happiness by getting married and raising children in a safe neighborhood is part of a proper developmental path; the desire to find happiness by using illegal drugs and joyriding is a distortion. Evolutionary explanations, by contrast, see both desires as equally adaptive responses to different environmental challenges.
Environments thus seem to play two different but entangled roles in determining people’s happiness: they help determine how capable people are of achieving the fulfillment of their desires, but they also help determine what desires people have. Is happiness a matter of the satisfaction of one’s actual desires or of satisfaction of the right desires? Aren’t some desires better than others, more likely to lead to true flourishing if they are fulfilled?
Sen and Nussbaum disagree somewhat about this—Sen considers that what matters most is the extent to which one has the capabilities needed to realize one’s own preferences, whatever these might be, while Nussbaum places more emphasis on the idea that certain capabilities are critical for people to live good lives whether or not the people themselves recognize this. But the possibility of divergent developmental strategies reveals a different problem—that there are trade-offs between capabilities. Thus, if environmental differences can cue divergent developmental or behavioral strategies where each strategy leads to a different suite of preferences and capabilities, there may be no grounds for deciding which strategy leads to the better—or even just the happier—life.
The differences between these problems can be understood through the lens of dystopian—and utopian—fiction. Start by thinking about George Orwell’s 1984 and Aldous Huxley’s Brave New World (Huxley 1932; Orwell 1949). Orwell envisages a world of overwhelming oppression by a totalitarian government, one that prevents people from fulfilling their natural preferences, and seeks—with partial success—to eradicate the preferences themselves by brainwashing and torture. Huxley’s vision is different—in his world, the government is not oppressive but manipulative; it does not block or suppress people’s preferences—it shapes them. Huxley’s World State uses people’s natural responses to social agreement, bodily pleasure, and subtle (and not-so-subtle) environmental cues as tools to control their desires.
Evolutionary psychologists focus on the Orwellian threat of overt unhappiness resulting from people’s inability to fulfill their preferences, or from forcible attempts to suppress those preferences, but in doing so they overlook a cost that is much more difficult to assess but potentially very significant—the Huxleyan danger of a hidden or cryptic loss of happiness resulting from a limitation of people’s horizons that makes it impossible for them to form certain preferences in the first place. These environmental effects may be difficult to detect even for observers who are looking for them. They include instances such as girls who forgo mathematical training—and the intrinsic satisfactions, extrinsic rewards, and cognitive enrichment that it can bring—because stereotypes lead them to become convinced that they dislike math or are unable to learn it, men who similarly forgo the rewards of becoming primary caregivers for their children, and many others whose horizons are limited by stereotypes or socially enforced assumptions of which they are unaware. In some instances environmental factors may provoke developmental responses that affect the capacities and preferences that individuals come to possess at a neurological level. If Gowaty’s hypothesis is correct, for example, environmental factors indicative of limited female sexual autonomy may trigger developmental responses that cause girls and boys to grow up with exaggeratedly different cognitive capacities and preferences, foreclosing permanently on some possible life paths for members of both sexes.
The further implication of the role of response in development, however, goes beyond both the Orwellian and the Huxleyan worries to what might be called a Le Guinean conclusion. Ursula K. Le Guin’s novel The Dispossessed: An Ambiguous Utopia (Le Guin 1974; see also Davis and Stillman 2005) compares three fictional societies. Her focus is on the ambiguous utopia of the subtitle, a society based on a philosophy of cooperative anarchism distantly inspired by Jean-Jacques Rousseau. But this utopia is indeed ambiguous and flawed, and the alternative societies (representing the free market and collectivist options) are shown as having real (and differing) strengths. Le Guin’s societies differ not only in the satisfactions that they provide but also in the desires that they foster, and because some of these are incompatible, no society enables its members to “have it all.”
Le Guin’s lesson is the lesson of the response perspective. Because different developmental or behavioral strategies will elicit different constellations of desires and capacities, there is no single form of the good human life, sanctioned by human nature; instead there are many different ways that human lives can be differently more—or less—good. As an abstract point, this seems reasonable—perhaps obvious. But it poses a difficult problem for practical decision making. How are we to choose between such alternatives? Suppose a simple intervention in an individual’s environment could make the difference between two developmental pathways leading to lives with quite different preferences and satisfactions. Perhaps one pathway leads her to a gender-traditional life in which she develops a strong desire to have children with whom she has a close nurturing relationship, and this desire is fulfilled. Perhaps the alternative pathway leads to a life in which she has less interest in children but a strong desire for professional success and creative satisfaction as a musician, and this desire is fulfilled. Each life will also involve many other desires and many satisfactions and frustrations, large and small. If we had the power, early in a girl’s life, to tip her toward one or the other of these pathways, how could we tell if it is right to intervene? This problem is hardly resolved by pointing to population averages or “species-typical” outcomes as indications of “human nature.”
Similar considerations apply, once again, to discussions of freedom. Evolutionary psychologists focus on a narrow conception of freedom as the ability to act in accordance with one’s preferences, and they regard people’s preferences as fixed relatively firmly by evolved human nature. They thus overlook the possibility of “adaptive preferences” both in the economist’s sense (preferences adjusted to avoid intolerable conflict with people’s expectations) and in the more radical sense (preferences that vary as part of environmentally cued variation in evolved developmental and behavioral strategies).
But their simple conception of freedom, and of its relationship to happiness, has deeper problems than this. The facile presumption that when people—members of a population—can choose freely, they will maximize their happiness, for example, is a long-recognized and serious error. Where people are free to express and act on their prejudices, one of the things they do is to limit other peoples’ choices. People’s freedom is restricted when they are barred from some domains of activity by their own and others’ responses (such as fear of social exclusion or biased hiring practices) or by niche construction (such as social and physical environments that exclude some kinds of people). And people’s happiness is reduced where they make their “free” choices within environmental contexts that attach intolerable social costs to options they would otherwise have preferred. When girls are bullied for expressing an interest in stereotypically “male” science fiction or computer gaming, boys are bullied for showing any enjoyment of “female” play with dolls, or black students experience social rejection for showing an interest in “white” academic success, their choices cannot ensure their happiness.
The capability approach gives a broader conception of freedom not merely as freedom from constraint but as freedom to accomplish things—to act autonomously, to realize one’s potential. According to this richer conception of freedom, very many people who are now able to follow their preferences relatively unimpeded are nonetheless far from free. And impending changes to our global environment will surely further drastically diminish many people’s freedom—taken in this broader sense—unless we are able to respond to these changes with appropriate social reform. Yet as in the case of happiness, a puzzle remains about how to compare lives with different constellations of preferences and capabilities, lives that realize different and incompatible sets of potentials.
The problem of how to weigh the trade-offs between alternative possible developmental trajectories for an individual and the lives these might lead to is already a difficult one. But it is complicated further when we consider change at the social level—interventions capable of affecting the lives of many different individuals in different ways at the same time. Thinking about capabilities as a key measure of a good human life can help to clarify what is going on in discussions about phenotypic trade-offs in this broader context.
Consider some characteristics that are good for people to have—phenotypic traits that we see as contributing, directly or indirectly, to people’s flourishing: good health, intelligence, compassion, curiosity, linguistic facility, physical coordination, self-confidence, musical ability, social sensitivity, mathematical aptitude, a sense of humor, and so on. Thinking about “happiness” and “freedom” is too abstract—it is important to consider the specific characteristics in which these abstractions are realized in order to take account of the different ways in which they contribute to good human lives. According to the capability approach, what is of value is not these phenotypic characteristics themselves but the contribution that they make in particular environments to people’s capabilities and resulting experiences—e.g., to achieving mastery in particular areas of human endeavor such as mathematics, language, sport, music, or the arts; to acquiring rich knowledge of the natural world or of cultural traditions; to exercising social leadership, making good judgments, showing compassion, sharing aesthetic experiences, creating works of art or novel scientific hypotheses, or building close social relationships. But the phenotypic characteristics are crucial nonetheless, for they provide the basis for these capabilities.
How can we compare the value of these various traits and the capabilities they support? As we noted at the outset of this book, there are problems both about how to weigh different phenotypic goods against each other and about how to weigh individual outcomes against distribution across the population. Consider again a situation in which we have the chance to make an intervention, but this time imagine an intervention capable of changing the phenotypes of a whole population, increasing or decreasing individuals’ performance on measures of some traits. Some such interventions can change the range of variation itself, modifying the peak performance—the highest performance achieved by any members of the population. We have three options:
• Option A increases peak performance on some measures but decreases other measures for the same individuals.
• Option B increases peak performance on some measures but decreases peak performance on other measures affecting other individuals.
• Option C decreases peak performance on one measure but increases performance on the same measure for many individuals at lower levels, increasing equality on that measure across the population.
There is no way to compare these outcomes to each other or to the status quo in the abstract; if the question is which option is the best choice, different answers will be appropriate for different measured traits depending on the kinds of value that they supply. In many cases it seems likely that even quite rich information about the effects of a particular intervention would leave its cost–benefit ratio hard to determine simply because the goods involved are so disparate. This situation makes such assessments particularly susceptible to bias resulting from the tacit assignment of relative values or weights to particular goods. The examples favored by evolutionary psychologists suggest that they are inclined to overvalue goods that they deem “natural” and to undervalue those they deem “unnatural.” This bias is widely shared, of course, as is a general tendency to assign high value to goods that are accorded high status in our culture. Both of these biases can contribute to assessments of the costs and benefits of change that echo existing gender stereotypes.
Option C on the list raises a special problem: how to weigh the value of equality against the value of maximizing peak performance on the variables that are deemed desirable. This is an important type of case for evolutionary psychologists since one of their main claims about the prospects for social change is that equality between the sexes, on many measures, is achievable only at a cost that might well be unacceptable. Recall Steven Pinker’s argument in favor of providing the special environments needed for rare talents to be realized and against the “leveling down” that a simple-minded egalitarianism might lead to. Pinker notes that we all benefit from the realization of rare talents, and so we all suffer if those talents go unfulfilled as a result of the pursuit of equality. But this is too quick, for we might all also benefit from the full realization of the unexceptional capacities of large numbers of people as well as from equality itself. The question is, how much? No doubt there are some traits for which what really matters most is what happens at the peak—for which a few extra-ordinary performers make a large difference for everybody, or at least for many. Pinker is thinking of cases like that. But a lot of work would need to be done to establish which traits really work like this. Even in the cases that appear to be of this sort, the assessment seems quite speculative. Do we really know that a single Mozart or Einstein is worth more than thousands or millions of more modest achievers in the arts or the sciences? And on the other hand, just how should we assess the loss to the most talented, if they are unable to develop their talents? These questions are not merely hypothetical. Finland, for example, does exceptionally well at educating a wide range of children. Finns sometimes complain, however, that their educational system does not fully develop the talents of the most gifted.
The problem of how to compare different distributions of goods is complicated by the possibility that equality or justice may have value of its own, as many philosophers have supposed. (It may also have very broad beneficial social effects, as some social scientists argue [Wilkinson and Pickett 2010], but this is a different point.) The “utopian” proposals for social change that evolutionary psychologists call into question often do take social equality as a serious goal. As I note in chapter 1, this is a goal that evolutionary psychologists often dismiss out of hand on the assumption that in many cases equality can be achieved only by means of a destructive “leveling down.” This assumption concerns issues of feasibility, and some reasons to reconsider it are offered in the next chapter. For now, it is sufficient to note that the very important and separate question of what the value of equality itself might be cannot even be raised within the individualist frameworks of classical cost–benefit analysis or evolutionary psychology.
A related point concerns the kind of equality in question. Social reformers often take equality as one of their highest goals, but as Amartya Sen has asked (Sen 1992), equality of what? Evolutionary thinkers discuss equality in two main contexts: in arguing for the impossibility of achieving “perfect gender equality” in behavior and in considering the harmful effects of “leveling down” in pursuit of equality. In both cases their focus is on (exact) equality in phenotypic outcomes. This is the kind of equality Kurt Vonnegut describes in “Harrison Bergeron”:
The year was 2081, and everybody was finally equal. They weren’t only equal before God and the law. They were equal every which way. Nobody was smarter than anybody else. Nobody was better looking than anybody else. Nobody was stronger or quicker than anybody else. All this equality was due to the 211th, 212th, and 213th Amendments to the Constitution, and to the unceasing vigilance of agents of the United States Handicapper General. (Vonnegut 1961, 5)
But though Vonnegut’s story is read by almost everyone (including Pinker) as a short, crude analog of Orwell’s 1984 or perhaps of Ayn Rand’s Anthem, its point is something quite different. It is a satire not of egalitarianism but of the fear of egalitarianism, and one of its main targets is the very idea that equality “every which way”—simple equality in people’s qualities and abilities—is what egalitarians aim to bring about (Hattenhauer 1998). The question of just what form and degree of equality is a good social goal is a difficult one. Sen offers a promising approach in Inequality Reexamined (Sen 1992) based on the capability view of what makes a good life. He argues that the inequalities that are harmful are inequalities in people’s abilities to achieve the ends that they value. This view is explicitly intended to make sense of how we can strive to reduce harmful inequality while recognizing that people differ from one another in many important ways so that both their preferences and their responses to particular features of their environment are diverse.
The costs of change that we have considered so far have taken the form of side effects: unintended and undesirable effects of actions undertaken to achieve desired aims. Costs of another sort, what I have dubbed investment costs, appear to fit into the story in quite a different way: as expenditures of resources or effort deliberately undertaken in order to bring about the social change in question or to maintain it once achieved: “the added energy for education and reinforcement,” as E. O. Wilson put it (1978, 148).
From the perspective of ordinary cost–benefit analysis, this conception of cost makes eminent sense; indeed, it is the only kind of cost that classical cost–benefit analysis—which weighs expenditures against payoffs—normally counts. Yet in the present context there is something peculiar about this notion of “cost.” Social change undoubtedly requires that resources and human effort be spent to achieve the transition, and perhaps to maintain the new state of affairs, and this expenditure is usually counted as part of the cost of such change. But tremendous quantities of resources and energy—in the form of education, advertising, cultural production, legislation and law enforcement, and individual social activity ranging from gossip and peer pressure to threats and violence—also go into reinforcing the patterns that evolutionary psychologists term “natural.” Should these be counted as costs associated with the status quo?
Many optimists about human possibilities think the answer is yes. These are investments of resources and effort that people are choosing to make—expenditures by individuals, by businesses, and by public institutions. We could make other choices instead without changing the amount of the investment. If all the resources—human and material—that we now spend on communications and activities that reinforce gender and racial stereotypes, encourage selfish consumerism, and aggrandize violent aggression were spent instead on alternatives that encourage generosity, compassion, and nonstereotypical thinking (optimists suppose), we might achieve substantial changes in the human condition with no extra investment cost at all.
This hypothetical possibility raises obvious questions about freedom of choice and about happiness in the form of people’s ability to live according to their preferences. The current allocation of resources, including human effort, is often presented as being the result of free choices in a free market. Advertisers use stereotypes to sell their products (it is said) because customers respond to them; movies and video games offer endless representations of men in violent conflict and women as sexual objects because that is what viewers want. This perspective sees the current use of resources not as a cost or expenditure but as an expression of various aspects of human nature. From this perspective, any sort of requirement that those resources be diverted to sending other messages looks like a combination of censorship and propaganda—it would obstruct people’s freedom to choose what messages they send, what cultural expressions they attend to, and what activities they pursue.
Philosophers have long debated whether existing social institutions express human nature or restrict it. Thomas Hobbes saw social institutions as the natural outgrowth of human nature as well as the only possible means of checking its excesses; Rousseau saw many social institutions as the chains that prevent human nature from expressing itself fully. The niche construction perspective makes it clear that this question needs rethinking: institutions (and many other human activities as well) have a double face—they express our nature but they also shape its expression. The choices that people make (as legislators, marketers, teachers, artists, consumers, parents, or community members) limit the choices available or even conceivable to others and to themselves. It cannot simply be claimed that the expenditures that people freely choose are not costs but an expression of their nature, whereas expenditures imposed upon them must be counted as costs. People make different choices under different circumstances, and their very choices constantly impose changes on others’ circumstances (and their own) that in turn shape the further choices of those affected. Indeed, we are evolved to struggle for control over ourselves, our environments, and each other. No choices are “free” in the sense of being independent of the choices others have made. The distinction between “freely chosen” and “imposed” expenditures of effort and resources cannot ground a distinction between expenditures that do not count as costs and those that do.
Is there another way to distinguish costs from other human expenditures of effort and resources? One intuitive idea is that some such expenditures are themselves natural to us and so do not count as costs. But this is again to accept an internalist distinction between phenotypes that are “natural” and those that are “forced” by environmental circumstances. A more promising way forward might be not to distinguish between natural and unnatural phenotypes or natural and unnatural environments but to tackle both together in the form of constructed niches and the traits that shape and respond to them. Perhaps some social systems—some combinations of social and material environments and associated phenotypes—are special in that they are self-sustaining, whereas others require ongoing intervention to keep them going. This idea is suggested by Wilson’s mention of the cost of “reinforcement” of new behavior patterns. It is made more explicit by discussions by Pinker and Buss of attempts to create new social systems that failed because the combinations were not self-sustaining. The fate of the utopian communities of the late nineteenth century and communes of the late twentieth century and of the more radical features of the kibbutzim, according to these evolutionary psychologists, reveals how human nature destabilizes social systems that are unnatural for us. No extra push is needed to keep people doing what is natural and re-creating the environments that in turn support them in doing so, but the unnatural requires ongoing investment to block reversion to the natural state (and is likely to fail, even so). Natural social systems, according to this perspective, are thus those that are self-sustaining; the use of resources by which they sustain themselves should not be counted as a cost but as part of their functioning.
For this view to make sense, the notion of “self-sustaining” systems—as distinct from those that are sustained by paying a cost—must be made clear. One obvious sense in which a system can fail to be self-sustaining is if its maintenance depends on inputs from an external source. This is the sense in which “natural” ecosystems and some traditional forms of agriculture are self-sustaining while modern agriculture that depends on inputs of fuel and inorganic fertilizers is not. Some local social systems certainly fail to be self-sustaining in this sense in that they are maintained only by reinforcement from without: prisons and colonial social systems are obvious examples. At the level of whole societies, however, this way of thinking seems to break down; the means by which fragile features are shored up (such as government funding or law enforcement) are themselves internal to the society. At this level of organization we can see more clearly that what matters is the robustness of the system as a whole: its capacity to reproduce and reinforce its own structural features, including the mechanisms by which its most fragile elements are maintained, and so both to survive external changes and internal disruptions and to renew its capacity to keep so doing.
The social systems that evolutionary psychologists regard as “natural” may well be robust in this sense, but it does not follow from this that other—and quite different—robust social systems are not possible. Whether such alternatives exist is a key question for any attempt to gauge the feasibility of social change and is addressed in the next chapter. An equally important question concerning the value implications of robustness has already been answered. There is no question that some very robust social systems are not good for human flourishing—a look at history makes this clear. Robustness is not itself a good and is no guarantee of goodness—if a social system makes people miserable or diminishes their lives, the fact that it is robust is no compensation; quite the contrary.
What I am suggesting is that the response perspective demands a shift not just in what counts as a cost but in what counts as an act or environmental condition whose costs must be reckoned. We cannot set aside some acts and conditions as “natural” and therefore to be omitted from our accounting; they must all be counted. The question is not whether they are natural but how robust are the social systems that give rise to them and what effects do they have on human lives. The measure of investment cost disappears, therefore, to be replaced with a measure of the negative effects the proposed new social system will have on human flourishing and a measure of how difficult it will be to achieve and maintain. This last is not a value measure at all but something equally important—a measure of feasibility.