Paul Spencer's death on 21 July 2015 at the age of 83 ended a distinguished and long career as a foremost and globally celebrated Africanist and social anthropologist. Although his ethnographic focus was on East Africa and on pastoralism and age organization in the region, he increasingly applied his extensive and accumulated knowledge to a range of other problems not confined to this part of the world. His was a classic case of a scholar first developing a formidable empirical grounding and expertise and then using it to address increasingly theoretical issues, including those to do with social and economic development. His work can indeed be regarded as making up a triangle of interlinked ethnographic, theoretical and applied approaches and contributions of great sophistication.

His many volumes on Maa-speaking peoples began with his work on the Samburu, which provided a kind of comparative template for much of his subsequent analyses. As well as drawing attention to similarities, he documented detailed differences between the ways in which Samburu and Matapato Maasai articulated their age organizations to create contrasting modes of authority, especially in relation to cattle-keeping and ownership. The status and role of elders and of the so-called young warriors, or moran, were always central issues in his work. Although he did not quite use the terminology, he can be regarded as having provided a model of the various transformations that age organization can take in its adaptation to varying ecologies, wider conditions of change and the need sometimes to shift from a primarily pastoralist to semi-farming and even hunting-gathering mode of subsistence. Unlike many such models, however, his was built on remarkably solid evidence of a kind rarely seen in modern anthropology. While the various forms of Maa social organization existing today have altered greatly since Spencer's ethnographic research, he has left a legacy that will invite analytical reflection and comparison for generations to come. For, while age organization of the Maa kind is exclusive to East Africa, its implications for understanding the wider potentialities of human growth and self-perpetuation are immense and can be regarded as central to questions of socio-cultural evolution.

Spencer's work shows that age organization, with its often critical distinction between age and generation sets, provides opportunities for the emergence of polities ranging from firm to more flexible gerontocracy and even to greater egalitarianism. Moreover, it indicates that pastoralism is also subject to alteration, sometimes adversely. One of Spencer's most compelling analyses was to demonstrate that the Maasai ‘commitment to growth’ of their cattle (and, by extension, of children through polygyny) conflicted with the expectations of modern economic growth. Thus, under conditions of drought, as pastureland became more restricted, overexploited and scarce, they did not sell cattle to match available feeding resources, preferring instead to try to keep their herds large regardless. This resulted in some herders having to become wage earners and others monopolizing diminishing herds. This relationship between demographic, environmental and economic change was investigated in a number of papers and was always based on his prior ethnographic study and publications.

It has to be emphasized that, right from the beginning, Spencer was interested in diachronic social processes. His data show that, within the so-called cyclical process of age organization, mutation could occur and was part of its flexibility. Similarly, pastoralism could not be labelled as a static, single mode of subsistence but was related to other modes as people moved through them. Phillip Gulliver's pioneering work had shown how the Maasai called Arusha had become agriculturalists and Spencer extended this flexibility to people such as the Rendille, who shifted between cattle and camel herding, and the Dorobo, who also depended heavily on hunting and gathering.

Curiously, one or two earlier commentators of his work characterized his ethnographies as ‘functionalist’, apparently in the sense that they were allegedly non-historical and probably because Spencer wrote in the ethnographic present. Yet this could not be further from the truth, for it was precisely his focus on social process that indicated that he was interested in depicting movement and the potential for new directions of movement, evidence for which emerged ever more obviously and clearly as his writing developed.

Spencer's interests were always related to each other. Thus, his interest in dance was by no means an anomalous turning away from his interest in age and pastoralism. In seeing dance as not just an expression of collective social forces but also an expression of individual dancers’ inner states, he analysed how Samburu acquired through dance a cosmic confidence that helped them overcome personal uncertainties experienced at a particular stage of the lifecycle. Indeed, in much of his work Spencer innovatively wove the subjective and sometimes emotional aspects of the peoples he studied into their forms of social organization. He was entitled to do so, having got to know people intimately through long-term fieldwork, learning their languages and cutting himself off from ‘outside’ contacts at a time when it was possible and desirable to do so.

Away from fieldwork, however, his work was intentionally directed towards the understanding and solution of practical as well as theoretical problems affecting more than just the peoples he studied ethnographically. As well as contributing to debates on the causes and effects of pasture deprivation, he wrote about marriage, gender and the status of women in the context of human growth and development – in some respects he was ahead of his time on these issues.

In fact, for nine years he had carried out research at the Tavistock Institute in the 1960s on city council policy making in Britain, in particular in Coventry. He joined the Institute in 1962 after submitting his Oxford doctorate on the Samburu, based on twenty-seven months of fieldwork there between 1957 and 1960. His first degree was in engineering at Cambridge and he used and developed a skill in statistical analysis at the Tavistock Institute – a skill that he continued to apply to his teaching and research in anthropology all his life. He again stood out in this respect compared with most other anthropologists, whose use of figures was usually little more than elementary. When he joined the School of Oriental and African Studies (SOAS) in 1971, he immediately impressed with his numeracy skills, which he used in a course on methods while at the same time showing how the traditional qualitative approach remained the sine qua non of the discipline of social anthropology. His teaching and research continued many of the interests of his fellow East Africanist and scholar of age organization, pastoralism and social change, Phillip Gulliver, whose own contribution to the department and scholarship had been immense.

In addition to his long association with SOAS, Spencer made a major contribution to African studies through his role as Honorary Director of the International African Institute from 1996 to 2004. During his directorship, Spencer was particularly active in revising and modernizing the statutes of the Institute and establishing its present status as a charity.

Spencer's last publication appeared in 2014, testimony to a lifetime's commitment both to his discipline and to the peoples he worked among. Like their indelibly recorded societies forever changed, he is irreplaceable.