The death of Peter Townsend, on 8 June 2009, has reverberated across multiple social science communities because the scope of his research and writing was unusually broad in an age of hyper-specialisation. Through an academic career spanning six decades, he made path-breaking contributions to the understanding and measurement of poverty and social deprivation, to the social meaning of disability, to the causes and consequences of health inequalities and to the sociology of ageing and later life. It is to the last of these fields that this appreciation is devoted. The first article that Peter published was ‘Old age and industry’, an account of the work of Alan Welford and his colleagues, which appeared in the Cambridge student magazine Varsity on 27 November 1948 (Townsend Reference Townsend1948). This early interest in ageing was sustained when he moved into employment in the research section of Political and Economic Planning (PEP), the forerunner of the Policy Studies Institute (1952–1954), as research officer at the Institute for Community Studies (1954–1957), and as Research Fellow and Lecturer at the London School of Economics (1957–1963). It was in this richly productive period as a sociologist and social policy analyst that he laid the foundations of his huge reputation in gerontology circles.

His 1954 PEP and 1955 Fabian Society pamphlets on pensions, the latter with Brian Abel-Smith, were important contributions to understanding the rise in the number of retirement pensioners and, especially, to making the case for a break with the minimalist approach to National Insurance pensions found in the Beveridge Report (Townsend Reference Townsend1954; Abel-Smith and Townsend Reference Abel-Smith and Townsend1955). The wellspring for subsequent UK debates about national superannuation and, much later, the state earnings-related pension scheme, can be found in this and other contemporary research. Also in 1955, Peter published an article in the Sociological Review that was the forerunner of the first of his four major books on old age. These four books constitute his main research on the sociology of later life, and each represents a building block of social gerontology.

The first of the four, The Family Life of Old People, concerned above all the embeddedness of older people in their families (Townsend Reference Townsend1957). It was mould-breaking in several respects. Its findings contradicted the false conventional and scientific wisdom of the day that older people were increasingly isolated, for he found very close familial ties despite residential segregation – what six years later Leopold Rosenmayr and Eva Köckeis (1963) were to label ‘intimacy at a distance’ – although he did also pioneer an approach to the study of loneliness that is referred to later. Secondly, the book presented poverty in old age not simply as a lack of resources but also as an inactive family life (p. 229), and thirdly it highlighted the critical role of family care (‘community care’) in maintaining older people in their own homes and preventing hospitalisation and residential care admission. The book was, of course, a product of its time, but while subsequent generations of social scientists have been highly critical of familism and particularly its gender bias, The Family Life of Old People is no paean to the family but rather a careful analysis of its complexity, including its negative aspects. Fourthly, in sociological terms, Peter used evidence to refute the claims of the functionalists that the extended family was a precarious anachronism in modern societies. Finally, in policy terms, this book was one of the first to make the case for ‘joined-up’ policies to support family care: health services, housing and social services. This is conventional wisdom now although, in terms of its realisation, still largely an aspiration.

In The Last Refuge published in 1962, Peter Townsend's sociological gaze shifted from the community to the residential institution and focused on the residents of old-age homes, a group that until then had been largely hidden from view. As before (and subsequently) there was a policy motivation as well as a sociological one: the fact that the National Assistance Act 1948 had imposed a duty on local authorities to provide residential accommodation for those ‘in need of care and attention’. After ten years, as Peter said, ‘we knew next to nothing about the system’ (p. 7). This painstakingly exemplary sociological investigation (in which Peter was aided by a small team that included Bob Pinker) came to a devastating conclusion:

Peter's study was one of the first to cast doubt on the goals of the post-war welfare state, regardless of how well intentioned were the architects and those with operational responsibility. It also posed sharp questions about the values of modern society which made uncomfortable reading in the context of early-1960s optimism. As a precursor to what was, much later, to become a major preoccupation with human rights, we find in The Last Refuge the full force of Peter's remarkable analytical and literary skills in his critique of the inhumane treatment of residents by poorly-prepared staff, too few in number, and over-bearing managers. Even more importantly, the book catalogues (with passion and barely concealed anger) the individual impact of total institutions in terms of increasing dependency and loss of human dignity:

In short, The Last Refuge was a massive indictment against institutional regimes. From a social policy perspective, the book should still be a key historical reference on the transition from the Poor Law to the welfare state, as well as one of the finest critiques of institutions ever written. Moreover, in it Peter made the case for replacing stigmatising institutional regimes with supported life in the community, which pre-dated long-stay hospital closures in this country by two decades. In the present academic vogue to quote only the most recent references, this important book is rarely appreciated in either social gerontology or social policy as an early example of the relationship between policy and practice and, specifically, the extent to which legislative change may be inhibited or facilitated by intermediaries and front-line staff or ‘street-level bureaucrats’, charged with implementation. As he understood, ‘social reforms are only as good as the individuals who put them into practice or as the means they are given to put them into practice’, and that this is as much about the political gap between rhetoric and reality as ‘the human problems which limit their fulfilment’ (pp. 228–9). In sum, The Last Refuge is both a close observation of the minutia of institutional life and, as is often the case in Peter's work, a plea for social justice, as with:

For the last words on this important book, I borrow those of an anonymous reviewer in the Times Literary Supplement:

The third main pillar of this remarkable corpus is Old People in Three Industrial Societies (Shanas et al. Reference Shanas, Townsend, Wedderburn, Friis, Milhoj and Stehouwer1968) – the first cross-national survey of old age. A landmark in comparative research as well as the sociology of later life, the study was based on representative samples and matched data covering Britain, Denmark and the United States of America. It is still an excellent starting point for anyone considering comparative research on ageing. The key focus of the book and, for me, a recurring point of reference, is the tension between ‘the integrative impulses of informal primary relationships and segregative impulses of formal industrial society’ (p. 426), or, echoing C. Wright Mills, between the personal and public issues concerning ageing under modernity. This book demonstrates empirically the paradox of strong integration in old age and the simultaneous general pressures towards group segregation (as the ‘old’, ‘pensioners’ or ‘retired’), as well as individual exclusion through poverty, isolation and disability. There are powerful policy messages which provide a bridge between sociological analysis and social policy, showing how exclusion could be overcome.

The fourth volume of Peter Townsend's major ageing corpus is The Aged in the Welfare State, which he co-authored with Dorothy Wedderburn (Townsend and Wedderburn Reference Townsend and Wedderburn1965). Although published before the three-nation study, it was a preliminary report of the findings from the British survey conducted in 1962–1963.Footnote ² It focused on the use of health and welfare services by older people and their income and assets, with Townsend reporting on the former and Wedderburn the latter. The chief interest of this book to current students of social gerontology is its historical reference point for contemporary controversies such as the failure of social assistance (then known as National Assistance and now the pension credit) to reach all older people in poverty and the ‘postcode lottery’ concerning access to local services. The survey also revealed evidence of the significance of family structure in determining institutionalism and ‘little evidence of health and welfare services being “mis-used” or “undermining” family responsibilities’ (p. 135). Of equal interest is the early demonstration of resilience in later life coupled with signs of the dawn of a more assertive era with regard to social services as the ‘traditional reserve and tight-lipped self-sufficiency’ of older people was becoming less rigid (p. 70).

Looking back on this collection of major foundation works in social gerontology, it is surprising to realise that the last of them was published more than 40 years ago. Other interests and priorities dominated Peter's work over these four decades, but there were two important further contributions to social gerontology. His paper in the first issue of this journal, ‘The structured dependency of the elderly: a creation of social policy in the twentieth century’, was a milestone in understanding the social construction of old age (Townsend Reference Townsend1981a). Although controversial at the time, it played a key part in the development of the political economy and, subsequently, critical gerontological perspectives. Students of social gerontology might like to approach this paper as a distillation of the weight of empirical evidence mastered in the four previous books. There will be a few who question whether students should be directed to a paper that is so obviously at odds with today's preoccupations with individualism and consumer culture, but evidently it remains a source of interest and inspiration to many – even in the last 12 months, it is the second most frequently read paper in this journal.Footnote ³ He regarded his 1986 paper, ‘Ageism and social policy’, as a reprise but it was in fact an even more radical statement about the damaging consequences of structural ageism (Townsend Reference Townsend, Phillipson and Walker1986). It is necessary to mention a final contribution to social gerontology that may be overlooked. Peter was one of the first to highlight the fact that the majority of disabled people are over pension age and, implicitly, to criticise the predominant disability stereotype of a wheelchair-user (Townsend Reference Townsend, Walker and Townsend1981b). Not for the first time, this set him against the official received wisdom, in this case that disability is ‘an inevitable feature of old age’. Of course this denial was designed to deflect the case for special benefits in recognition of disability in old age and the access of older people with disabilities to disability benefits.

It is impossible to do justice briefly to Peter Townsend's immensely impressive body of work. A massive understatement would be that he made a major contribution to social gerontology. Because the main body of his research was over 40 years ago, it is less cited than it should be: the tendency noted earlier to focus primarily on the most recently published sources (especially in doctoral theses) even when they obviously lack the rigour and insights of earlier work. Nonetheless it is reasonable in my view to claim that he was one of the main founders of this discipline. In the 1950s and 1960s, almost single handed, he took on the ‘acquiescent functionalism’ of theories such as disengagement and the excessive focus on individual adaptation to ageing. He made the case for a structural perspective that remains strong in the mainstream of social gerontology despite the recent cultural turn. There is no doubt also that his work inspired many of those currently researching the ageing field, including myself. Nor is his legacy solely of historical interest, because he developed various analytical strands that still need revisiting. The trade-offs between structure and agency are again very topical but, also in the sociology of later life, essential aspects of social structure, social groups and institutions have been downgraded in favour of an often excessive focus on individual-level factors such as adaptation.

Finally, an appreciation of Peter Townsend must mention his campaigning role, for he was to the core a public intellectual completely committed not only to understanding society but also to changing it. In this role, he is best known as co-founder of the Child Poverty Action Group and Disability Alliance and as long-term chair of both. Throughout his life, he was too an active campaigner on behalf of older people. His role in framing the case for a national superannuation scheme has been mentioned, but he also argued passionately for it in public. Similarly, during the 1970s and 1980s, he highlighted the high prevalence of poverty in old age and the need for increases in the Basic State Pension (BSP). During the late 1990s, when the election of the ‘New Labour’ government meant that restoring the uprating link between the BSP and an index of earnings was on the cards, and hopes for social progress were running high, Peter and I co-authored a Fabian Society tract published on the 40th anniversary of the publication of the first of the Society's pamphlets on pensions: it argued for a new form of basic pension with the entitlement based on citizenship rather than contributions (Townsend and Walker Reference Townsend and Walker1995). Despite encouraging words before the 1997 election and heavy campaigning with Barbara Castle, Jack Jones and others after it, means-testing remains a key feature of social security in old age in the United Kingdom and around one-quarter of older people live in poverty. Restoration of the earnings-related uprating is promised for 2012.

Although campaigns for social reforms most often lead to disappointment, Peter had more than his fair share of successes, including the State Earnings-Related Pension Scheme, the Attendance Allowance, a nationwide programme of building sheltered housing, and numerous measures to combat child poverty. Despite the setbacks, he was an indefatigable campaigner, driven by a passionate moral conviction for social justice and belief in the power of truth. All of those working on the field of social gerontology have good reasons to be grateful for his legacy that has also enhanced the wellbeing of millions of people.