Is the concept of “egalitarian societies” useful? For years, I’ve argued that it is. Over the past decade, I’ve conducted fieldwork in Ethiopia among Chabu, often characterized as egalitarian, and Hamar, who are socially stratified. In my comparative work, I’ve built databases from ethnography on leadership, social behavior, and political organization. But following the authors’ process-driven framework, we should abandon the concept. Their framework dissolves egalitarianism into modular processes, yet they retain a society-level definition.
The authors demonstrate that total equality is nonexistent, even among the most staunchly egalitarian societies. I press the inverse: total hierarchy is never achieved either. A process-based approach predicts egalitarian mechanisms will emerge in some domains and among subgroups within all societies, including those whose larger social organization is starkly stratified. If these mechanisms are as widespread and modular as the authors suggest, they should function in societies classified as hierarchical.
The Hamar are patriarchal pastoralists with marked wealth and reproductive inequalities, alongside age and gender hierarchies (Dubosson, Clack, & Brittain, Reference Dubosson, Clack and Brittain2018). Several of the authors’ mechanisms manifest. Hamar are famous for their ritual initiation (the ukuli bull-jumping ceremony), which involves intense social hierarchy. The maz age-grade structure subordinates younger men to their seniors. Yet, the maz (initiated men) form a bounded cohort with strong norms of solidarity and mutual obligation. Grazing land is not individually owned, and families routinely share livestock as a form of risk-pooling. Lydall and Strecker (Reference Lydall and Strecker1979) describe political life organized around competitive oratory and persuasion rather than coercion, and influential men maintain their standing through eloquence and consensus-brokering (prosocial reputation-building). Strecker (Reference Strecker, Widlok and Gossa2005) characterizes Hamar political life as “regulated anarchy” alongside its pronounced patriarchal structure. Despite stark gender inequality, some women serve as mediators of conflict between men and women (Garfield & Glowacki, Reference Garfield and Glowacki2022). The Maasai osotua risk-pooling partnerships and Borana consensus decision-making within gada age-grade cohorts (Legesse, Reference Legesse1973) further illustrate how the authors’ mechanisms operate within societies no anthropologist would classify as egalitarian. This creates tension within their thesis. Even within nominally hierarchical societies, one can find domains where the authors’ mechanisms generate relative equality – a prediction of a process-based, domain-specific framework and one of its key contributions.
Accepting that egalitarianism and inequality are found in all societies raises a levels-of-analysis question that the authors do not fully resolve. The six mechanisms operate across levels of society, including within families; residential, economic, and ritual settings; and age cohorts, all of which are nested within political or ethnolinguistic units (Garfield, Syme, & Hagen, Reference Garfield, Syme and Hagen2020). Multilevel social structures shape where the mechanisms function and for whom, suggesting the social domain and subgroup can be more analytically important than the societal level. The authors note that the processes are correlated with particular socio-ecological conditions but underemphasize social and contextual variation within societies. The authors’ review shows that some nominally egalitarian societies contain domains that exhibit levels of inequality comparable to pastoralist or agriculturalist contexts. Retaining the society-level label obscures this variation; the same holds for classifications of “non-egalitarian” or stratified societies.
The authors’ process-based framework is an excellent analytical improvement. But as presented, it risks pulling in two directions at once: dissolving “egalitarian societies” into modular processes while still treating “egalitarian” as a useful society-level classification.
Their definition raises unresolved issues. The authors define egalitarian societies as those where most individuals…secure their own resource access, status, and autonomy. But “most” is an underspecified quantitative threshold. Does 51% suffice? Does 95%? Societies near the threshold would oscillate in and out of being egalitarian depending on fluctuating conditions, undermining the stability of the society-level label. Their definition also remains static despite the authors’ acknowledgment that egalitarianism is dynamic. Over what period of time must “most individuals” secure these resources? Over a lifetime? Across generations? The definition doesn’t say.
The definition also lacks context about who “most” refers to. The authors note that many definitions of egalitarianism stipulate equality only “within any given age-sex grade” (Fried, Reference Fried1967). Does “most” apply across all individuals or within subgroups? Does the definition implicitly apply only within gender? If so, this also suggests egalitarianism is best understood as emerging within sections of societies, rather than defining them.
A key theoretical benefit of the authors’ process/mechanism-based framework is not classifying societies but identifying where, when, and for whom leveling processes are activated; how they function across social domains; and how they scale within and across subgroups. This suggests abandoning the concept of nominally egalitarian societies. The authors give three reasons for retaining the label despite their process-based critique: inertia, genuineness, and empirical validity of the processes. I agree that the second and third justify studying egalitarian mechanisms but not retaining a society-level classification. I am not persuaded by the first.
Paradigm shifts risk disjunctions between existing literatures and compounding the cross-disciplinary confusion that the authors diagnose. But inertia is not a scientific justification for maintaining a framework. The geocentric model of the solar system carried enormous social and scientific inertia, yet it was wrong. The early heliocentric models initially yielded worse predictive accuracy, and the transition was costly, yet it was useful scientific work to absorb those costs. If the society-level characterization of “egalitarian” is limiting our understanding of the social and cognitive processes that generate egalitarian and hierarchical dynamics, then we should be willing to abandon it.
The authors acknowledge that the six processes are not unique to hunter-gatherers but stop short of what follows: if the mechanisms operate across subsistence contexts and within nominally hierarchical societies, the society-level label does no useful classificatory work. None of this undermines the paper’s contribution as a theory of the mechanisms that generate and constrain inequality. I have aimed to build on the authors’ framework by pressing a logical extension: just as no society achieves total equality, no society lacks egalitarianism entirely. In abandoning “egalitarian societies” as a classificatory type, we stand to gain a more precise account of leveling dynamics, capable of locating them where they operate (e.g., age cohorts, ritual settings, and residential groups) as readily among the Hamar as among the Chabu.
You have
Access
Is the concept of “egalitarian societies” useful? For years, I’ve argued that it is. Over the past decade, I’ve conducted fieldwork in Ethiopia among Chabu, often characterized as egalitarian, and Hamar, who are socially stratified. In my comparative work, I’ve built databases from ethnography on leadership, social behavior, and political organization. But following the authors’ process-driven framework, we should abandon the concept. Their framework dissolves egalitarianism into modular processes, yet they retain a society-level definition.
The authors demonstrate that total equality is nonexistent, even among the most staunchly egalitarian societies. I press the inverse: total hierarchy is never achieved either. A process-based approach predicts egalitarian mechanisms will emerge in some domains and among subgroups within all societies, including those whose larger social organization is starkly stratified. If these mechanisms are as widespread and modular as the authors suggest, they should function in societies classified as hierarchical.
The Hamar are patriarchal pastoralists with marked wealth and reproductive inequalities, alongside age and gender hierarchies (Dubosson, Clack, & Brittain, Reference Dubosson, Clack and Brittain2018). Several of the authors’ mechanisms manifest. Hamar are famous for their ritual initiation (the ukuli bull-jumping ceremony), which involves intense social hierarchy. The maz age-grade structure subordinates younger men to their seniors. Yet, the maz (initiated men) form a bounded cohort with strong norms of solidarity and mutual obligation. Grazing land is not individually owned, and families routinely share livestock as a form of risk-pooling. Lydall and Strecker (Reference Lydall and Strecker1979) describe political life organized around competitive oratory and persuasion rather than coercion, and influential men maintain their standing through eloquence and consensus-brokering (prosocial reputation-building). Strecker (Reference Strecker, Widlok and Gossa2005) characterizes Hamar political life as “regulated anarchy” alongside its pronounced patriarchal structure. Despite stark gender inequality, some women serve as mediators of conflict between men and women (Garfield & Glowacki, Reference Garfield and Glowacki2022). The Maasai osotua risk-pooling partnerships and Borana consensus decision-making within gada age-grade cohorts (Legesse, Reference Legesse1973) further illustrate how the authors’ mechanisms operate within societies no anthropologist would classify as egalitarian. This creates tension within their thesis. Even within nominally hierarchical societies, one can find domains where the authors’ mechanisms generate relative equality – a prediction of a process-based, domain-specific framework and one of its key contributions.
Accepting that egalitarianism and inequality are found in all societies raises a levels-of-analysis question that the authors do not fully resolve. The six mechanisms operate across levels of society, including within families; residential, economic, and ritual settings; and age cohorts, all of which are nested within political or ethnolinguistic units (Garfield, Syme, & Hagen, Reference Garfield, Syme and Hagen2020). Multilevel social structures shape where the mechanisms function and for whom, suggesting the social domain and subgroup can be more analytically important than the societal level. The authors note that the processes are correlated with particular socio-ecological conditions but underemphasize social and contextual variation within societies. The authors’ review shows that some nominally egalitarian societies contain domains that exhibit levels of inequality comparable to pastoralist or agriculturalist contexts. Retaining the society-level label obscures this variation; the same holds for classifications of “non-egalitarian” or stratified societies.
The authors’ process-based framework is an excellent analytical improvement. But as presented, it risks pulling in two directions at once: dissolving “egalitarian societies” into modular processes while still treating “egalitarian” as a useful society-level classification.
Their definition raises unresolved issues. The authors define egalitarian societies as those where most individuals…secure their own resource access, status, and autonomy. But “most” is an underspecified quantitative threshold. Does 51% suffice? Does 95%? Societies near the threshold would oscillate in and out of being egalitarian depending on fluctuating conditions, undermining the stability of the society-level label. Their definition also remains static despite the authors’ acknowledgment that egalitarianism is dynamic. Over what period of time must “most individuals” secure these resources? Over a lifetime? Across generations? The definition doesn’t say.
The definition also lacks context about who “most” refers to. The authors note that many definitions of egalitarianism stipulate equality only “within any given age-sex grade” (Fried, Reference Fried1967). Does “most” apply across all individuals or within subgroups? Does the definition implicitly apply only within gender? If so, this also suggests egalitarianism is best understood as emerging within sections of societies, rather than defining them.
A key theoretical benefit of the authors’ process/mechanism-based framework is not classifying societies but identifying where, when, and for whom leveling processes are activated; how they function across social domains; and how they scale within and across subgroups. This suggests abandoning the concept of nominally egalitarian societies. The authors give three reasons for retaining the label despite their process-based critique: inertia, genuineness, and empirical validity of the processes. I agree that the second and third justify studying egalitarian mechanisms but not retaining a society-level classification. I am not persuaded by the first.
Paradigm shifts risk disjunctions between existing literatures and compounding the cross-disciplinary confusion that the authors diagnose. But inertia is not a scientific justification for maintaining a framework. The geocentric model of the solar system carried enormous social and scientific inertia, yet it was wrong. The early heliocentric models initially yielded worse predictive accuracy, and the transition was costly, yet it was useful scientific work to absorb those costs. If the society-level characterization of “egalitarian” is limiting our understanding of the social and cognitive processes that generate egalitarian and hierarchical dynamics, then we should be willing to abandon it.
The authors acknowledge that the six processes are not unique to hunter-gatherers but stop short of what follows: if the mechanisms operate across subsistence contexts and within nominally hierarchical societies, the society-level label does no useful classificatory work. None of this undermines the paper’s contribution as a theory of the mechanisms that generate and constrain inequality. I have aimed to build on the authors’ framework by pressing a logical extension: just as no society achieves total equality, no society lacks egalitarianism entirely. In abandoning “egalitarian societies” as a classificatory type, we stand to gain a more precise account of leveling dynamics, capable of locating them where they operate (e.g., age cohorts, ritual settings, and residential groups) as readily among the Hamar as among the Chabu.
Acknowledgments
None.
Financial support
The author declares no financial support.
Competing interests
The author declares no competing interests.