It was humbling to have Geography and Geographers selected for discussion at a conference forty years after publication of the first edition and delightful that the other book discussed was by Peter Taylor: we collaborated closely on research for a decade around that time, and published a joint book—never yet revisited in a conference session (Taylor and Johnston 1979)!Footnote 1 It was also humbling and extremely pleasing to have our books described as ‘canonical’, ‘classic’, ‘foundational’, even ‘intimidating’.

So why was the book written? Its pre-history and subsequent history were rehearsed in the Prefaces to the various editions and my response in the Progress in Human Geography ‘Classics in Human Geography Revisited’ series (Livingstone et al. 2007). When I joined the University of Sheffield’s Department of Geography in 1974, I was asked to participate in a final-year undergraduate course on the history of geography that Malcolm Lewis inaugurated; he would prefer me to cover post-1945 changes. The 1945 start-date for what became the book was thus pre-set—but I would probably have chosen that date anyhow, as discussed in the book.

After some three years Alan Wilson asked for ideas for a book series he was editing. I suggested one based on the course, and he advised that a textbook would be better with another publisher. I sent an outline to John Davey at Edward Arnold who quickly accepted itFootnote 2; it was finished in late 1978. Further editions came at the publishers’ request.

Why did course and book take that particular form—maintained, with some modifications, in later editions (becoming problematic, as I discuss later)? In the mid-1960s I was working at Monash University, and Basil Johnson lent Peter Rimmer and me Frontiers in Geographical Teaching (Chorley and Haggett 1965) and Locational Analysis in Human Geography (Haggett 1965), which we devoured while on fieldwork in Ballarat and Bendigo. I had been dabbling with the ‘new geography’, having come across the work of Berry and others during my Manchester MA, and was convinced of the need to apply statistical methods to my (vague) hypotheses—Percy Crowe helped in those first very faltering steps (Johnston 2019). At Monash it was suggested that I do a PhD on Melbourne’s geography. I discovered the large American literature (reviewed in Urban Residential Patterns; Johnston 1971) and the statistical methods deployed to test models deriving from the Chicago School, aided by contacts with Duncan Timms and Frank Jones who were slightly ahead of me on the same track (Timms 1965, 1971; Jones 1969).

Peter Haggett’s book convinced me I was on the right course,Footnote 3 and Models in Geography (Chorley and Haggett 1967) showed that I was on the edge of a major movement to change the nature of geography, with which I was in total sympathy after the uninspiring regional courses that dominated so much of my Manchester degree. Their chapter ‘Models, paradigms and the new geography’ became my text. I assumed that the ‘new geography’ (locational analysis, or spatial science, or ….) would soon dominate the discipline, but I got a rude awakening in 1969. I was invited by Brian Berry to be a corresponding member of the IGU’s Commission on Quantitative Methods in Geography and attend its foundation meetings in Ann Arbor and London. The former preceded the Association of American Geographers annual meetings, which exposed me to what became known as radical geography (Barnes and Sheppard 2019); I realised that more than one inter-paradigm contest was in full flow. Hearing David Harvey when I was at the LSE in 1973 provided further evidence of the three-cornered fight (regional geography vs quantitative geography vs radical geography)—and Social Justice and the City (Harvey 1973; Johnston 1974Footnote 4) pressed the message home.

So when I started preparing my Sheffield lectures in early 1975, I was deeply imbued in Kuhn and paradigms and used them as the course framework. Indeed, by then I knew a fourth paradigm was looming over the horizon—what was then called humanistic geography—but it wasn’t given the emphasis accorded the other three until the book’s fourth edition (by when the course had been moved to the first year, taught by Paul White). I tried the structure out in a paper before writing the book (Johnston 1978).

Why was the book an apparent success? In part there was no real competitor—neither Freeman’s (1963) One Hundred Years of GeographyFootnote 5 nor James’ (1972) All Possible Worlds included anything on the paradigm wars that had already erupted in North America but not the UK.Footnote 6 By then, most British geography undergraduate degree syllabuses included a course (usually compulsory) on history and philosophy, and a text was desirable, especially one dealing with the contemporary scene. (Many teaching such courses—usually senior staff members (the professor!) —were troubled by the changes and needed an accessible guide.Footnote 7) And no early competitor appeared. The revised editions kept it up-to-date and any potential competitors at bay—or so it seemed.Footnote 8 Different books came along, notably Livingstone’s (1992) —excellent but with a different focus.

Geography and Geographers’ longevity became a problem unresolved by later editions—though James Sidaway’s contributions updated it on material increasingly passing me by. Its structure was determined by Kuhn’s paradigm model as embraced by Haggett and Chorley: the discipline was in considerable flux in the 1970s with different groups seeking hegemony. The book was about those battles and although I later modified use of the full paradigm model (realising that complete revolutions were never likely) I retained the concept as the structure within which the discipline was displayed—separate communities competing for established positions within the whole.Footnote 9

But then the battles virtually ended. In part (especially in the UK) many more university geography students, academics and researchers meant that there was room for different communities to offer their competing approaches, letting students decide which they preferred. Departmental wholes broke into separate parts—research groups that were, to a greater or lesser extent, internally coherent with little cross-group interaction (what Peter Taylor refers to as silos in his response). Conflict was replaced by tolerant co-existence; there was jockeying for position and resources, but little public debate about what (human) geography should and shouldn’t be (Johnston 2006; Johnston et al. 2014a, b—these are essays on intra-disciplinary politics; Geography and Geographers was more about its sociology).

So what were the later editions of the book to be about? As originally conceived it was not about what geographers study but rather the debates on how they should study—writings about human geography rather than writings that are human geography.Footnote 10 Later editions thus focused less on inter- and intra-paradigm debates and more on the increasing variety of work within each—not what the book’s structure was designed for, and is increasingly unsuited to, which is undoubtedly why the publisher has not suggested an eighth edition. The ‘real’ history that it tells—of the immediate post-1945 decades—is less relevant now; too long and detailed for an overview of a fairly distant recent past before courses turn to contemporary work.

A different book is needed, but is it feasible, especially as a single- or two-author enterprise rather than an edited volume, which will necessarily lack some coherence? Human geography is now so large and broad that mastering its many facets and introducing them to students in an accessible (undoubtedly relatively brief) form is a mammoth task.Footnote 11 It is now practised in a substantial number of discrete silos, some further subdivided into separate communities (small villages?) between which there is relatively little contact, as recently discussed in an essay on a leading quantitative geographer (Johnston 2020).

The commentaries published here, largely based on the later editions, reflect the difficulty of describing such a multi-polar discipline within its original framework. As admitted in my response to Jan Monk (Livingstone et al. 2007), I was tardy in giving some material, such as feminist geography, the coverage it deserved in earlier editions (and Sharon Cheong rightly suggests that there remain areas that the later editions don’t fully address); the same might be said now of material on race, an issue flagged by Katharine Hall.Footnote 12 The original, and sustained, goal was simply to present students with a narrative on the post-1945 debates within Anglo-American human geography. Books written about the discipline’s history from another geographical perspective, whether Brazil or Nigeria (Craggs and Neate 2019) or …, will tell a different story potentially challenging what some see as an Anglo-American search for hegemony—something that I have never conceived of promoting; and, as Fiona McConnell expresses it, different stories could be written from one of the (many?) margins of contemporary Anglo-American human geography. Regarding Mike Heffernan’s comments on Europe, after 1945 Anglo-American geographers paid little attention to what geographers were doing there unless it was to help them join their paradigm revolutionsFootnote 13; they still taught—to some extent—about Europe, but writings that are human geography were not Geography and Geographers’ focus.

Geography and Geographers is dying; this collection must surely be the last time it is given such attention. That it has endured through seven editions is most gratifying; that something else is now needed is made very clear by the commentaries. They have brought home the clear message that as the world changes so should the way we present it, and Fiona McConnell has explored possible future replacements. Human geography differs in substance—in every sense of that term; much bigger, much broader, much more sophisticated, eclectic and rigorous—from when moves to change it emerged in the 1960s/1970s when the book was conceived. It now needs, indeed deserves, something different from Geography and Geographers.