Unto Others: The Evolution and Psychology of Unselfish Behavior is a seminal 1999 book co-authored by philosopher Elliott Sober and evolutionary biologist David Sloan Wilson, which argues that altruism is a genuine and adaptive feature of both biological organisms and human psychology.[1] Published by Harvard University Press, the work integrates insights from evolutionary biology, philosophy, psychology, and anthropology to challenge prevailing individualistic explanations of behavior and revive the theory of group selection first proposed by Charles Darwin.[1] Sober and Wilson demonstrate through examples ranging from self-sacrificing parasites to cooperative insect societies that unselfish actions can evolve when they benefit groups, even at individual cost.[1] The book also examines the proximate psychological mechanisms underlying altruism, such as empathy and commitment, proposing that human unselfish behavior stems from evolved traits that promote group cohesion and survival.[1] Spanning 416 pages with extensive references, Unto Others has been widely influential, cited over 5,900 times in academic literature (as of 2024) for its rigorous defense of multilevel selection theory and its implications for understanding morality and sociality.[2]
Authors and Background
Elliott Sober
Elliott Sober, born in 1948, is an American philosopher renowned for his work in the philosophy of biology and philosophy of science. He earned his PhD in philosophy from Harvard University in 1974, under the supervision of Hilary Putnam. That same year, Sober joined the faculty of the University of Wisconsin-Madison, where he served as the Hans Reichenbach Professor and William F. Vilas Research Professor until his retirement in 2023.[3]Sober's early scholarship established him as a leading voice in evolutionary philosophy, with his 1984 book The Nature of Selection: Evolutionary Theory in Philosophical Focus offering a rigorous examination of core evolutionary principles. This work analyzed the conceptual foundations of natural selection, including trait-based versus gene-based perspectives, and provided essential groundwork for later discussions of multilevel selection by clarifying how selection operates at different biological levels.[4][5]In his broader contributions to the philosophy of altruism, Sober emphasized semantic analyses to define evolutionary terms precisely and causal analyses to distinguish between proximate causation—mechanisms operating within an individual's lifetime—and ultimate causation—the evolutionary reasons for those mechanisms. These distinctions help resolve ambiguities in how altruism, as a behavioral trait that benefits others at a cost to the actor, can be understood evolutionarily without conflating psychological motives with selective pressures.[6]Sober's role in Unto Others: The Evolution and Psychology of Unselfish Behavior (1998), co-authored with David Sloan Wilson, centered on supplying the philosophical framework for conceptualizing altruism and challenging strict individualistic selectionism in favor of multilevel alternatives. His analyses ensured that definitions of altruism were semantically clear and causally grounded, bridging evolutionary theory with psychological inquiries into unselfish behavior.[1]
David Sloan Wilson
David Sloan Wilson, born in 1949, is an American evolutionary biologist and SUNY Distinguished Professor Emeritus of Biological Sciences and Anthropology at Binghamton University, where he has taught since 1975.[7] He earned his B.A. from the University of Rochester in 1971 and his Ph.D. in Zoology from Michigan State University in 1975.[7] Wilson's career has focused on applying evolutionary principles to social behavior, emphasizing the role of group-level processes in natural selection.[8]In his early work, Wilson challenged the dominance of individual-level selection theories through innovative modeling approaches. His seminal 1975 paper, "A Theory of Group Selection," introduced the trait-group model, demonstrating mathematically how altruistic traits could evolve if individuals form temporary groups where between-group variance exceeds within-group variance, thereby favoring group productivity over individual fitness.[9] This framework, published in the Proceedings of the National Academy of Sciences, provided a rigorous alternative to kin selection and reciprocal altruism, sparking renewed debate on group selection's viability.[9] Subsequent papers in the late 1970s and 1980s built on this by incorporating empirical data to test model predictions.[10]Wilson's biological approach to altruism relies heavily on empirical observations from animal behavior, highlighting cases where group selection explains cooperative traits that reduce individual fitness but enhance group survival. For instance, in birds like robins, reproductive restraint—where individuals limit breeding to avoid overpopulation and resource depletion—illustrates potential group-level adaptations, as selfish breeders disrupt group stability.[10] In insects, eusocial species such as bees and ants exhibit extreme altruism, with sterile workers sacrificing reproduction to support the colony, a pattern that multilevel selection models attribute to strong between-group competition.[10] Laboratory experiments with flour beetles (Tribolium) further validate this, showing that selecting for group productivity over generations alters cannibalistic behaviors to favor cooperative outcomes.[10]In Unto Others: The Evolution and Psychology of Unselfish Behavior (1998), co-authored with philosopher Elliott Sober, Wilson provided the empirical and modeling backbone for arguments supporting group selection as a mechanism for altruism's evolution.[1] He contributed detailed analyses of trait-group and multilevel frameworks, drawing on animal behavior data to ground the book's biological claims in verifiable evidence.[1] Through this partnership, Wilson blended his biological expertise with Sober's philosophical rigor to advocate for a pluralistic view of selection processes.[1]
Publication and Overview
Publication Details
Unto Others: The Evolution and Psychology of Unselfish Behavior was initially published in hardcover by Harvard University Press in 1998 (ISBN 0-674-93046-0), with a paperback edition released on October 1, 1999 (ISBN 0-674-93047-9).[1][11] The book comprises 394 pages in the hardcover edition and 416 pages in the paperback, including an extensive bibliography and index.[1][12]The volume represents the culmination of a long-standing collaboration between philosopher Elliott Sober and evolutionary biologist David Sloan Wilson, who together sought to address key debates in evolutionary theory regarding altruism and selection processes.[13] This partnership developed in the context of post-1970s evolutionary biology, particularly following Richard Dawkins' influential 1976 work The Selfish Gene, which promoted a gene-centric view of evolution and cast doubt on the viability of group-level explanations for adaptive traits. Their joint efforts built on earlier individual and collaborative explorations of selection levels, aiming to rehabilitate group selection as a legitimate mechanism.No major revised editions of the book have appeared since its initial release, though it remains available in both hardcover and paperback formats through the publisher and major retailers.[1]The publication coincided with a renewed interest in multilevel selection theory within evolutionary biology, highlighted by David Sloan Wilson's organization of a seminal symposium on the topic in The American Naturalist in July 1997, which featured contributions from leading researchers and underscored the timeliness of Sober and Wilson's arguments.[14] This context positioned Unto Others as a key text in the ongoing revival of group selection discussions.[15]
Central Thesis and Structure
The central thesis of Unto Others posits that unselfish, or altruistic, behavior constitutes a genuine and prevalent aspect of both biological nature and humanpsychology, capable of explanation through natural selection processes that encompass group and multilevel selection, thereby challenging the prevailing individualistic interpretations of evolution that emphasize self-interest above all.[1] This argument integrates evolutionary biology with philosophical analysis to rehabilitate group selection as a viable mechanism, demonstrating how altruism can evolve without reducing it to disguised selfishness or kin-based reciprocity alone.[1]The book is structured into two main parts, preceded by an introduction and followed by a conclusion, providing a systematic progression from evolutionary foundations to psychological implications. The introduction, titled "Bentham's Corpse," establishes the foundational distinction between semantic altruism—defined as behaviors that benefit others at a cost to the actor—and psychological altruism, which pertains to the underlying proximate mechanisms driving such behaviors.[1] Part I, comprising Chapters 1 through 7, focuses on evolutionary altruism, covering its definitions, theoretical models, and critiques of individualistic paradigms. Part II, spanning Chapters 8 through 12, shifts to psychological altruism, examining ultimate motives and supporting evidence from human behavior. The conclusion synthesizes these threads, discussing broader implications for understanding cooperation in society and science.[1]A key framing device throughout is the explicit differentiation between semantic altruism, which describes observable actions that impose a fitness cost on the individual while benefiting others, and psychological altruism, which concerns ultimate motives oriented toward others' welfare rather than self-benefit.[1] The text draws on references from diverse fields including biology, philosophy, and psychology to substantiate its claims, reflecting an interdisciplinary synthesis.[1] Comprising 337 pages of main content, the book maintains an accessible yet rigorous style, interweaving formal theoretical models with illustrative examples from nature and human contexts to elucidate complex ideas without sacrificing depth.[1]
Evolutionary Arguments
Defining Altruism
In Unto Others, Sober and Wilson define altruism semantically in evolutionary biology as a behavioral trait that imposes a net fitnesscost on the actor while providing a net fitness benefit to the recipient, where fitness is measured in terms of expected reproductive success or offspring production.[16] This definition focuses on the observable consequences of the behavior rather than the underlying mechanisms or intentions, allowing for a neutral framework applicable across species. For instance, the warning calls emitted by birds, such as crows, exemplify this: the caller risks attracting predators and thus incurs a personal survival cost, while alerting others to danger increases their chances of survival.[17]The authors distinguish this semantic approach from causal definitions of altruism, such as those inspired by Ramsey's framework, which attempt to classify behaviors based on their evolutionary origins or the genes they purportedly serve.[18]Sober and Wilson critique causal definitions for their potential circularity, arguing that they risk rendering altruism impossible by definition—equating "selfish" behaviors with those that evolve via natural selection, regardless of their effects on others.[17] In contrast, the semantic view avoids assuming psychological states or specific causal pathways, enabling multilevel analyses that consider individual, group, or other levels without presupposing how altruism arises.[19] This emphasis on payoffs, rather than outcomes, is formalized through payoff matrices that capture the net costs and benefits independently of whether the trait spreads evolutionarily.[17]A key distinction in the book is between evolutionary altruism, defined by genetic fitness effects, and psychological altruism, which pertains to ultimate motives aimed at the welfare of others independent of self-interest.[16] Evolutionary altruism does not require conscious intent; it is a descriptive category for traits like those in social insects or mammals that may appear cooperative but are assessed solely by fitness impacts.[19] This separation prevents circular reasoning, as defining altruism in terms of motives would beg the question of whether such motives exist, whereas the evolutionary definition relies on empirical fitness measures. Parental care in mammals, for example, can qualify as altruistic in group contexts if it reduces the parent's direct fitness while enhancing the group's overall reproductive output, though it often aligns with kin selection in practice.[17]
Multilevel Selection and Group Selection
In Unto Others, Elliott Sober and David Sloan Wilson articulate multilevel selection theory as a framework in which natural selection operates concurrently at individual and group levels of organization. Under this theory, altruistic traits—those that reduce an individual's fitness while benefiting the group—can evolve when the variance in fitness among groups exceeds the variance within groups, thereby enabling between-group selection to outweigh within-group selection against altruists. This pluralistic approach avoids privileging one level of selection, allowing for trait-specific evaluations where group-level benefits may drive the spread of altruism in certain lineages.Sober and Wilson revive the concept of group selection, which had been largely dismissed in the 1960s and 1970s following critiques that portrayed it as theoretically incoherent and empirically negligible compared to individual selection. A seminal critique came from George C. Williams, who argued that adaptations are overwhelmingly individual-level and that group benefits are illusory or unstable due to free-rider problems within groups. In response, Sober and Wilson defend group selection's conceptual rigor, emphasizing the trait-group model in which groups form and dissolve dynamically based on phenotypic similarities rather than fixed genotypic units, thus facilitating multilevel dynamics without requiring implausibly rigid structures.The mathematical backbone of their argument adapts George Price's equation—a general description of evolutionary change—to partition selection across levels. The multilevel Price equation expresses the change in the average value of a trait \bar{z} as:\Delta \bar{z} = \frac{\mathrm{cov}(w_g, z_g)}{\bar{w}} + E\left[ \frac{(w_i - \bar{w_g}) z_i}{\bar{w}} \right],where z denotes the trait value, w is fitness, the subscript g refers to group-level averages, the subscript i to individuals, \bar{w} is mean fitness, \mathrm{cov} is covariance, and E[\cdot] is expectation. This formulation derives from the standard Price equation \Delta \bar{z} = \frac{\mathrm{cov}(w, z)}{\bar{w}} + E[\Delta z_i] (assuming no transmission bias for simplicity), by decomposing the overall covariance into between-group and within-group components: the first term captures selection among groups (favoring groups with higher average trait values and fitness), while the second term reflects selection within groups (weighted by deviations from group-mean fitness). For altruism to spread, the positive between-group covariance must exceed the negative within-group effect; for instance, in a simulation with groups of altruists (who pay a cost c but provide benefit b) and selfish individuals, altruism invades if intergroup competition amplifies group productivity differences beyond intragroup exploitation.Empirical support for multilevel selection includes eusocial insects, where traits like sterile worker castes in honeybees evolve despite severe individual reproductive costs, as colony-level success in resource acquisition and defense drives group proliferation. While haplodiploid sex determination facilitates kin-biased altruism, Sober and Wilson argue it is insufficient alone, as multilevel processes better account for the colony-as-superorganism dynamics observed across taxa. Similarly, in human hunter-gatherer societies, cooperative behaviors such as shared foraging and conflict resolution enhance small-band survival and expansion through intergroup competition and migration, with archaeological and ethnographic evidence indicating that groups with stronger internal cooperation outcompeted less cohesive ones over evolutionary timescales.
Psychological Dimensions
Ultimate Altruistic Motives
In Unto Others, Sober and Wilson distinguish between proximate and ultimate motives in the context of altruism, arguing that ultimate altruistic motives involve a genuine, non-instrumental concern for the welfare of others as an end in itself, rather than as a means to personal gain. Proximate motives, by contrast, refer to the immediate psychological mechanisms—such as empathy or emotional responses—that trigger behavior but do not necessarily reflect the deeper evolutionary goals. This framework posits that humans possess ultimate altruism when their actions stem from a desire to benefit others independently of reciprocity or self-interest, exemplified by empathy-driven helping where the actor's satisfaction derives solely from the recipient's improved state.[1]The authors provide an evolutionary rationale for the emergence of these ultimate altruistic motives through multilevel selection theory, which—as discussed in earlier sections—suggests that natural selection operates at both individual and group levels, favoring groups composed of altruistic members over those dominated by selfish individuals. In human evolutionary history, particularly in small ancestral groups, such motives enhanced group survival and cohesion, as altruists contributed to collective well-being even at personal cost, thereby increasing the group's competitive edge in resource-scarce environments. This process, rooted in group selection dynamics, explains why ultimate altruism could evolve as a stable psychological trait in our species.[1]Philosophically, Sober and Wilson critique psychological egoism—the view that all actions are ultimately selfish—by demonstrating its logical flaws and lack of empirical necessity, arguing instead for motivational pluralism where both egoistic and altruistic ultimate motives coexist within individuals. They contend that egoism oversimplifies human psychology by reducing all behaviors to self-directed desires, ignoring evidence that altruistic concerns can be intrinsic and non-derivative. This pluralistic defense aligns with evolutionary expectations, as a mix of motives allows flexibility in varying social contexts without undermining group benefits.[1]Illustrative examples of ultimate altruism include acts of wartime self-sacrifice, such as a soldier throwing themselves on a grenade to protect comrades, where the motive prioritizes others' lives over one's own without expectation of reward or reciprocity. Similarly, anonymous donations to strangers, like contributing resources to a community without personal recognition, reflect this intrinsic concern, as the donor's ultimate goal is the alleviation of others' suffering rather than any indirect self-benefit. These cases underscore the book's claim that such motives are not merely illusory but evolutionarily adaptive components of human psychology.[1]
Empirical Evidence from Psychology
One of the cornerstone empirical contributions to understanding altruistic motives comes from C. Daniel Batson's series of experiments in the 1980s testing the empathy-altruism hypothesis. In these studies, participants induced to experience empathic concern for a person in distress, such as a confederate receiving electric shocks, were more likely to help by taking on the aversive task themselves, even when easy escape options were available that would reduce their own discomfort without aiding the victim. This pattern held across variations designed to isolate empathic concern from potential egoistic drivers like mood enhancement or guilt avoidance, with empathic individuals helping at significantly higher rates than low-empathy controls. Sober and Wilson highlight these results as compelling evidence against psychological egoism, showing that helping can stem from other-oriented motives rather than self-interest.[1]Batson's experiments systematically critiqued alternative egoistic interpretations of empathy-driven helping, such as the idea that apparent altruism merely reduces the helper's personal distress. For instance, manipulations that equalized mood or distress levels across conditions still yielded helping driven by empathy, not self-relief. Similarly, when social approval was minimized through anonymity, empathic concern remained the strongest predictor of aid. These findings reject disguised egoism by demonstrating that the ultimate goal is the victim's welfare, not the actor's. The authors interpret this as aligning with ultimate altruistic motives, where behavior prioritizes others' needs independently of personal gain.[1]Further evidence emerges from analyses of cooperative behavior in prisoner's dilemma games, which reveal tendencies toward unselfish choices beyond reciprocity or reputation concerns. In one-shot iterations with anonymous players and no future interactions, substantial cooperation occurs even under high anonymity, suggesting intrinsic cooperative inclinations not reducible to calculated self-benefit. Sober and Wilson cite such data to argue that these outcomes reflect genuine altruistic dispositions, challenging purely egoistic models of social interaction.[1]Developmental psychology provides additional support through observations of spontaneous helping in young children, indicating early-emerging unselfish behaviors. Studies from the 1990s show toddlers as young as 2 years old proactively assisting adults in tasks like retrieving dropped objects, without prompts, rewards, or social pressure, at rates comparable to older children in naturalistic settings. These acts occur across diverse contexts and persist despite no immediate reciprocity, pointing to innate altruistic tendencies rather than learned egoism. The authors draw on this to underscore how such behaviors manifest prior to full cognitive understanding of social exchange.[1]Cross-cultural observations reinforce these patterns, documenting unselfish acts in varied societies where individual gain is absent, such as costly participation in communal rituals. In small-scale communities, individuals endure painful or resource-intensive ceremonies—like scarification or fasting—that signal commitment to the group without direct personal benefits, yet foster collective cooperation. Ethnographic accounts from diverse cultures, including hunter-gatherers and agrarian groups, show consistent displays of aid to non-kin strangers, with helping in experimental tasks mirroring Western findings. Sober and Wilson use these examples to illustrate the universality of altruistic motives across human societies.[1]
Reception and Influence
Critical Reviews
Upon its publication, Unto Others received widespread acclaim from scholars for revitalizing the debate on group selection in evolutionary biology and for bridging philosophical and biological perspectives on altruism.[20] In the New York Review of Books, Richard C. Lewontin commended the book for its "radical reexamination" of altruism's evolutionary origins through group-level processes, using examples like the myxomatosis outbreak in Australian rabbits to illustrate how groups can function as units of differential reproduction.[20] Similarly, a review in the Quarterly Review of Biology engendered lively debate by integrating multilevel selection theory with empirical evidence from across the animal kingdom. Kim Sterelny, in Biology and Philosophy, praised its clear, cogent, and occasionally entertaining defense of group selection as a coherent alternative to individualistic accounts, noting its accessible style for interdisciplinary audiences.The book's psychological arguments, which posit ultimate altruistic motives shaped by evolution, also drew positive attention for their novel approach to proximate causation.[1] Reviewers highlighted how Sober and Wilson's framework avoids reducing unselfish behavior solely to egoism, instead allowing for genuine altruism at both biological and psychological levels.[21]Criticisms emerged primarily from evolutionary biologists skeptical of group selection's necessity. Richard Dawkins, a proponent of gene-centered views, has argued that genic selection adequately explains altruism without invoking higher-level processes, though in the 2006 edition of The Selfish Gene he acknowledged aspects of multilevel selection. Some philosophers questioned the novelty of the book's core semantic-causal distinction, which separates definitional (semantic) from explanatory (causal) conceptions of altruism, suggesting it echoed earlier debates in evolutionary theory without substantial innovation.[22] Daniel C. Dennett, in his commentary, critiqued the authors' pluralism as "inconstant," arguing it inconsistently applied levels of selection and undermined the parsimony of gene-level explanations.The book sparked significant scholarly debates, particularly on the validity of multilevel selection. A 2002 symposium in Philosophy and Phenomenological Research featured a précis by Sober and Wilson alongside commentaries from Dennett, Brian Skyrms, and others, focusing on whether group selection truly offers new explanatory power or merely reframes individual selection. Later clarifications by Steven Hecht Orzack and Sober in works like their 2001 paper on adaptationism addressed methodological challenges in testing multilevel models, responding to critics by emphasizing statistical criteria for distinguishing selection levels.Overall, Unto Others has been well-regarded in academic circles, earning an average rating of 3.85 out of 5 on Goodreads based on over 100 user reviews.[23] As of November 2025, the book had garnered approximately 5,958 citations in scholarly literature, reflecting its enduring role in discussions of altruism and selection.[24]
Academic and Scientific Impact
Unto Others has significantly influenced evolutionary biology by reviving interest in multilevel selection theory, providing a rigorous framework that distinguishes it from earlier, often dismissed, group selection ideas. The book's detailed analysis demonstrated the conceptual coherence and empirical viability of multilevel selection, enabling researchers to partition variance in fitness across individual and group levels without conflating the two. This revival is evident in subsequent works, such as David Sloan Wilson's Darwin's Cathedral: Evolution, Religion, and the Nature of Society (2002), which directly builds on the multilevel framework to explore how group-level adaptations foster cooperation in human societies, including religious contexts.[25] The text has also been integrated into foundational literature, including Samir Okasha's Philosophy of Biology (2006), where it is cited as a key examination of kin versus group selection debates.In psychology, Unto Others bolstered research on the empathy-altruism hypothesis, particularly C. Daniel Batson's experimental paradigm, by offering an evolutionary rationale for genuine other-oriented motives underlying prosocial behavior. Batson himself engaged with the book in a 2000 commentary, acknowledging its service in bridging evolutionary and psychological explanations of altruism while critiquing certain aspects. The work influenced studies in positive psychology and moral development, exemplified by Felix Warneken's experiments in the 2000s demonstrating spontaneous helping in young children and chimpanzees, which align with the book's arguments for innate altruistic tendencies shaped by multilevel processes.[26]The book's broader legacy includes approximately 5,958 citations on Google Scholar as of November 2025, reflecting its role in shaping debates on cultural evolution and the application of group selection to religion, as seen in Wilson's extensions of multilevel theory to explain religious behaviors as group adaptations.[24] Critiques of the original multilevel models prompted refinements, such as the distinction between MLS1 (price-equation based partitioning without group properties) and MLS2 (trait-group models emphasizing group-level traits), which clarify when group selection operates independently.[27] These developments have informed applications beyond academia, including models of cooperation in public policy—such as analyses of reciprocity in administrative decisions—and studies in animal behavior that explore altruism in non-kin contexts, like primate food-sharing.[28]