Introduction

“It is a curious fact of contemporary history that one of the Western countries in which radical geography has acquired its strongest position is the small and complacent kingdom of Denmark”, Steen Folke (1985, p. 13) reflected in an early overview of radical geography in Scandinavia. The meaning of “strongest position” can be debated, of course, but in comparison with other Nordic countries, there is no doubt that the radical-geographical movement that took form at Copenhagen University in the years around 1970 was particularly lively (Asheim, 1987). In Sweden, Gullberg and Lindström (1979, p. 4) bluntly assessed that “the radical critique and the Marxist alternatives are particularly rudimentary and undeveloped within the geographical disciplines.” With some exceptions, the same could be said about Marxist geography in other Nordic countries in the 1970s (Folke, 1985; Lehtinen & Simonsen, 2022).

Radical geography can take many forms. Among the radical geographers at Copenhagen University, however, the project quickly became unequivocally Marxist (e.g. Folke, 1972). Indeed, for the students and teachers who in late 1971 established Fagligt Forum as an alternative structure at the Department of Geography, the aim was to provide “teaching and research on a Marxist theoretical basis” (Buch-Hansen, 1972, p. 9). Similar turns to Marxism happened elsewhere, but for many Danish radical geographers the initial inspiration came from a somewhat unusual direction, namely from Elementare Theorie der ökonomischen Geographie by the German Democratic Republic (GDR) geographer Gerhard Schmidt-Renner (1966). This book became the basis for what we will discuss as the territorial-structure approach to human geography.

In this chapter we revisit the territorial-structure approach. In part, this is a contribution to the emerging contextual histories of radical geography (e.g. Barnes & Sheppard, 2019; Berg et al., 2022), which nuance and problematise the generalising and all too often Anglo-American-centric “textbook” accounts of this varied field (see also Ferretti, 2019). But we particularly examine the territorial-structure approach as an early attempt at theorising geography as a dialectical relationship between the social and the material, in this case from a Marxist position. We start by introducing the historical-geographical context for the territorial-structure approach.

Radical Geography at Copenhagen University

Danish radical geography, like many radical geographies elsewhere, emerged from the intersecting developments that have “1968” as their emblem. More than in Norway and Sweden, for example, the Danish “youth rebellion” was a “student rebellion” (Jørgensen, 2008), and the rise of radical geography at Copenhagen University was part and parcel of this. Until the establishment of Roskilde University in the mid-1970s, Danish geography was only institutionalised at the universities in Copenhagen and Aarhus. A radical-geographical environment emerged at Aarhus University, which during the “red decade” of the 1970s became a bastion of Marxism (Jørgensen & Jensen, 2008). For example, it was a group of geography students at Aarhus University who translated Schmidt-Renner’s book (Schmidt-Renner, 1977). But radical geography did not get the same foothold at Aarhus University. In significant part, this was because radical ideas found a particularly nourishing context at the Copenhagen Department of Geography (Larsen, 2022). As elsewhere, the geography students (and some young teachers) in Copenhagen rebelled against professorial hegemony and traditional understandings of research and education. “Break down the professorial regime – participation, now!” and “Research for the people, not for profit!” were slogans of the time. But the radicalising geographers at Copenhagen University also rebelled against what they saw as an antiquated approach to geography. This was less pronounced at Aarhus University, where many geographers in the 1960s had joined the “Quantitative Revolution” (Framke, 1982; Jensen-Butler, 1999). At Copenhagen University, on the other hand, the radicalising geographers saw their department as a quagmire of problematic specialisation, regional description and environmental determinism, all smothered in a heavy emphasis on natural science and a vocal distaste for “theory”. Somewhat like geography at Clark University, which also had allowed the “Quantitative Revolution” to pass by (Huber et al., 2019), geography at Copenhagen University was overripe for criticism.

The 1970 University Act (Styrelsesloven) did much to democratise Danish universities, notably by securing students influence in the governing boards of the universities and equalising the formal status of professors and non-professorial staff (Hansson, 2018). Through often bitter struggles the radicalising geographers also won important local skirmishes (but rarely the battles) at the Copenhagen department. But frustrated with internal departmental struggles and becoming increasingly politicised, a group of students and young teachers in late 1971 established Fagligt Forum as an alternative structure for radical geographical research and education (for elaborations of the following, see Folke, 1985; Hansen & Jensen, 1983; Larsen, 2022). This included educational activities, such as a rather gruelling introduction course in Marxist theory, as well as working groups for research (and action) on the European Economic Community (EEC), development (and Imperialism), urban issues and the production of alternative teaching material for the upper secondary school – the destination for many graduates. Later, in 1973, Fagligt Forum launched the journal Fagligt Forums Kulturgeografiske Hæfter (from 1979 simply Kulturgeografiske Hæfter).

The radical geographers never included more than a handful of the academic staff, but radical geography was dominant among the students at the Copenhagen department in the 1970s. This radical environment played an important role in establishing geography at the new Roskilde University in the mid-1970s, even if for some it was to deliberately avoid particular theoretical avenues – such as the territorial-structure approach – taken at Copenhagen University (Brandt, 1999). The radical geographers were also involved in launching the annual Nordic Symposium on Critical Human Geography (Lehtinen & Simonsen, 2022; Öhman, 1990) that inspired the still-existing Nordic Geographers Meeting (Clark, 2005). By the early 1980s, however, Folke (1985, p. 15) detected a “stagnation – some would even speak of crisis – in Danish radical geography.” Radical geography was indeed entering a crisis, but we will first (and foremost) focus on the Marxist theorisation of geography that took form during radical geography’s heyday at Copenhagen University.

The Need to Analyse Territorial Structures

Taking the cue from Schmidt-Renner’s Elementare Theorie der ökonomischen Geographie (Schmidt-Renner, 1966), which was reviewed in the first issue of Kulturgeografiske Hæfter (Nielsen, 1973), the concept of territorial structure was arguably the most distinct idea in Danish radical geography of the 1970s. The concept was systematically presented in an article by Buch-Hansen and Nielsen (1977a). Buch-Hansen was at the time a postgraduate researcher at the Copenhagen department, while Nielsen was a newly-minted lecturer. Antipode – misspelling both their names – subsequently published the article in translation (Buch-Hansen & Nielsen, 1977b).Footnote 1 Manoeuvring around some of the more obvious shortcomings of Stalinist Marxism-Leninism and introducing a view of geography that was still to emerge in the Federal Republic of Germany (FRG), Schmidt-Renner’s book received some positive comments but had little impact in FRG geography (Belina, 2014; Belina et al., 2022). Its impact was undoubtably greater in the radical-geographical environment in Denmark. For the group that translated Schmidt-Renner’s book (and noted some important problems in his GDR Marxism), for instance, the book was “an important foundation for the elaboration of a critical and materialist geography” (Pedersen et al., 1977b, p. 180).Footnote 2 The “travel” and “translation” of Schmidt-Renner’s theory into Danish radical geography is interesting and could be studied in its own right. Here, however, we will mainly focus on how the theory was articulated in Danish geography.

In positioning their articulation of the territorial-structure approach, Buch-Hansen and Nielsen argue that several attempts had been made to combine Marxist theories with geography in the Western world, but find that these attempts failed because they did not successfully integrate the two. On the one hand, many attempts at introducing Marxism in geography simply became a “repetition of what Marx might have written about that particular topic, or a more general repetition of the central factors of political economy” (Buch-Hansen & Nielsen, 1977b, p. 1). On the other hand, and closer to their concern, they argue that attempts at synthesising geography and Marxism had failed because they only dealt with the spatial expressions of classical Marxist problems. This was also the case for Soviet geography, which, a Marxist philosophy notwithstanding, was as non-synthetic as in the West and thus only amounted to very simple theories about the location of production. For Buch-Hansen and Nielsen, dealing with spatial expressions ultimately relied on a traditional “bourgeois microscale geographical approach” that was unable to “explain anything about inequality and poverty in the capitalist society”:

We have, as geographers, to get rid of our inherited bourgeois traditions and concern ourselves in our academic work with the task of producing an historical and dialectical materialist theory which develops the spatial aspects of development and underdevelopment to a higher degree than it has been until now by Marxists. (Buch-Hansen & Nielsen, 1977b, p. 2).

It was from these considerations that Buch-Hansen and Nielsen proposed the concept of territorial structure as a way to overcome the weaknesses of conventional economic geography as well as the embryonic attempts at Marxist geography. In this endeavour, they argued, a theory about the location of production is central because it enables a geographical analysis of the capitalist mode of production and its social effects.

Theoretically, the territorial-structure approach linked up with at least two overarching concerns in Fagligt Forum. First, human geography was seen as a social science by most of the radical geographers. In the words of a radical staff member, it was “unacceptable to claim that human geography is natural science” (Document, 1974). This was significant, because geography was (and is) at Copenhagen University located at the Faculty of Science, and natural science dominated in teaching, research and in the general outlook at the Department of Geography (Hansen & Jensen, 1983). For the territorial-structure approach, as for virtually all radical geography at the department during the 1970s, the social-scientific foundation was Marxism. Second, if closely linked, the radical geographers were highly critical of the specialisation and compartmentalisation of knowledge production. “Bourgeois science focuses on the detail and has neglected to understand the totality. The political aim is always well hidden”, Fagligt Forum (1972, p. 6) argued programmatically: “We find that Marxist science better explains the reality – its aim is exactly to uncover reality in its totality, not to disguise some parts of it.” It was primarily for this reason that the radical geographers resisted suggestions that they should be moved to the social sciences, as that would fragment human geography. The radical staff members also opposed the establishment of sub-disciplinary research groups (so-called laboratories) within the department. When they finally agreed to form a separate “laboratory”, it was characteristically called “General Human Geography” (“Almen kulturgeografi”). The overarching objective of this research group was to “clarify the relationship between mode of production, social formation and territorial structure” (Laboratorium, 1975, p. 13), and for Buch-Hansen and Nielsen (1977b, p. 3), it was “important to realize that the geographical structure can only be analyzed, described, explained and understood through the total social development.”

Aiming for an approach in which social relations and territorial structures were theorised dialectically, Buch-Hansen and Nielsen found encouragement in the newly-established journal Espace et sociétés. Interestingly, they did not refer to the work of one of the founders of the journal, Henri Lefebvre, who later became a key inspiration for Marxist geography as well as other critical-geographical perspectives (in this book, see Simonsen, 2022). But through a range of articles from the journal, they found support for their dialectical position: “a given social formation is reflected spatially” and “the spatial structure in itself is a factor in the development of society” (Buch-Hansen & Nielsen, 1977b, p. 3). However, they found the notion of l’espace somewhat unclear and instead preferred the concept of territorial structure derived from Schmidt-Renner.

As one would expect for a radical-geographical theory, the aim was overtly political. This was also the case for the wider radical-geographical movement at the Copenhagen department. Initially, however, the activities of students and a few staff members were mainly political in the sense that they were directed at radically changing how the university was organised and how teaching and research were performed. But from around the establishment of Fagligt Forum in late 1971, this radicalisation became more clearly aimed at changing society (Larsen, 2022). “We worked to establish an education that could serve the oppressed instead of our hitherto masters”, as the call to establish Fagligt Forum put it; but the radicalising geographers found that they had been drawn into university-political “pseudo-rebellions” (Document, 1971). Buch-Hansen and Nielsen (1977b, p. 4) provided an example of this radical re-orientation in their presentation of the territorial-structure approach: “it is in [the theory’s] ability to generalize the experience gained from class struggle and, thus, in its usefulness as a guide in this struggle, that our view of geography, like all other views and theories about the development of society, will be tested.”

Territorial Structure as Concept and Approach

The territorial-structure approach is a general Marxist theory about how and why different localities of production and consumption are connected, and how they are historically and geographically conditioned by the modes of production that determine the social and economic development of society. It is in this way an early attempt at a Marxist theorisation of geography. For its proponents, this could only be achieved by developing a human geography that takes its point of departure in the “laws of the development of society” (Buch-Hansen & Nielsen, 1977b, p. 3) and thereby analyse the relations between a society and its geography. This meant analysing the territorial structure. In fact, although they were critical of certain aspects, even the group that translated Schmidt-Renner’s book reaffirmed this position, arguing that “Without an understanding of [the laws of the development of society], one cannot analyse and explain the structuring in space of material production, neither abstractly nor concretely” (Pedersen et al., 1977b, p. 190).

Put simply, the territorial structure is an expression of the physical and functional spatial structure of localities characterised by production and consumption connected via infrastructure (Fig. 4.1). For Schmidt-Renner (1966), such localities are referred to as Standort, a term retained in the Danish translation with reference to Marx’ notion of locus standi (Pedersen et al., 1977a), but Buch-Hansen and Nielsen generally used “lokalitet” and “locality” (Buch-Hansen & Nielsen, 1977a, b). And as Brandt (1990) later noted, Standort was “one of the concepts that we never succeeded in translating to an understandable Danish.” Nevertheless, the central purpose of the approach is to analyse how the social and economic development of societies are both reflected in, and create, territorial structures, and how this enables an analysis of what was discussed as the “regional problem”; that is, in short, why productive activity, or economic development, occurs in one place and not in another, resulting in uneven geographical development (see, e.g., Nielsen, 1976b). For Schmidt-Renner, regional differences and inequalities within capitalism can be explained by the basic features of its mode of production. This was a significant argument for the Danish radical geographers, because it eschews explanations of regional differences based on nature, race or religion, arguments they accused traditional “bourgeois geographers” of advancing (see, e.g., Buch-Hansen et al., 1979). In an assessment of different Marxist theories about regional differences, Nielsen (1976b) finds Schmidt-Renner’s approach to be the most promising. And arguing from a historical-materialist point of view, which sees social and economic development as determined by the modes of production, the proponents argue that it is the modes of production that determine the localisation of production, and thereby determine the territorial structure. This is probably why some referred to the territorial-structure approach as the “mode-of-production perspective” (Pedersen et al., 1977b). In this way, a core aspect of the approach is to show how modes of production shape, and are themselves shaped by, territorial structures. In Buch-Hansen and Nielsen’s (1977b, p. 5) definition: “The territorial structure is – for the capitalist mode of reproduction – the totality of production localities (productive and unproductive), consumption localities and the localities of the external conditions with the infrastructure that physically and functionally ties it all together.”

Fig. 4.1
A flow diagram of territorial structure includes super structure, nature, population, the relations of production, productive forces, localisation process, factory, housing, school, etc. all are connected to each other.

Model of the territorial structure concept. (Adapted from Hansen, 1994, p. 111)

Full size image

Infrastructure has an important role in the territorial structure. But infrastructure should not be approached in isolation, something Buch-Hansen and Nielsen accuse “bourgeois geographers” of doing. Rather, infrastructure should be theorised and understood as part of the mode of production. It should be viewed in the totality of which it is part, and governed by the same “laws of development” as those that determine other parts of the material life of societies. Infrastructure, understood in this way, should be viewed as the physical and functional network that connects the localities of production with localities of consumption. The cultivation of fields, for example, or the production of raw materials or goods in any given locality, all demand certain infrastructural requirements like railroads, waterways and telecommunication networks in order to function and connect to localities where they can be consumed or used. Within this theorisation of geography, infrastructure takes a specific role in the capitalist mode of production, where it should be viewed as “the physical and functional manifestations of exchange” (Buch-Hansen & Nielsen, 1977b, p. 5). This is similar to arguments advanced by Marx about the circulation of capital, but references to Marx are surprisingly absent from Buch-Hansen and Nielsen’s (1977b) Antipode paper (elsewhere, however, they engaged more with the work of Marx and classic Marxist literature; see, e.g., Buch-Hansen, 1976).

The territorial-structure approach is based on five “elementary conditions” (Buch-Hansen & Nielsen, 1977b), or “Standort-factor groups” (Schmidt-Renner, 1966), which are theorised as the general determining factors of localisation and hence the form and function of the territorial structure. These are (1) the mode of production (composed of the productive forces and the relations of production), (2) nature (or the physical-geographical environment) and (3) the conditions (growth and density) of the population, which Buch-Hansen and Nielsen see as derived from “historical materialism”. In addition to this, they add (4) the social (political-ideological) superstructure and (5) the already existing territorial structure.

First, the mode of production is theorised as the main determining factor for the territorial structure, but to understand its geographical role it is important to distinguish between the productive forces (human labour power and the means of production) and the relations of production (the relations between labourers and the owners of the means of production), and how they develop in a dialectical relationship. The general idea is that since the development of the productive forces is always subject to different historical and geographical conditions, it manifests differently in different places and at different times. For example: “To transform nature into usable products, humans use tools and machines. The development of these has taken place as an uninterrupted process throughout history. Sometimes development is fastest in one part of the world, other times in another” (Buch-Hansen et al., 1979, p. 21). The development of the productive forces in this way comes to have an important influence on the processes that drive the localisation of production, often viewed in a long-term historical perspective. Furthermore, the development of the productive forces (particularly through industrial specialisation) is theorised as forcing a technical division of labour that, in turn, necessarily develops into a social division of labour. And since the division of labour manifests differently in different places and at different times, this also involved a societal division in the territorial structure, which leads to the conclusion that the class structure of capitalist society has, and creates, a distinct geography.

The relations of production, though importantly understood as developing in a dialectical relationship with the forces of production, is also theorised as playing an important role for the development of localities characterised by either production or consumption, and thereby for the form and development of the territorial structure. “The productive forces have developed throughout history,” Buch-Hansen et al. (1979, pp. 21, 23) argue, “but they do not develop by themselves and independently of society in general. On the contrary, the social structure of society is crucial” in terms of “the ownership of the means of production and the social distribution of the societal product.” An example highlighted in relation to this is the historical and gradual technical and social division of labour from industrial specialisation and the changes in the productive forces. This means that some locations are, or become, more profitable localisations for production than others, based among other things on social and geographical differences in the supply of labour power. Accordingly, it is maintained: “With the development of the productive forces, there has been an ever-increasing division of labour – technically, socially and geographically. Not only production and consumption, but also the individual parts of production have been geographically separated” (Buch-Hansen et al., 1979, p. 27). Nevertheless, while Buch-Hansen and Nielsen pay much attention to the relations of production and their connections to the territorial structure, it is ultimately the mode of production (or, more correctly, the purpose of production) that is the determining factor for the localisation of production. This assertion goes hand in hand with a broader critique of capitalism, underscoring that: “Under the capitalist mode of production it is profitability for the owner of the means of production that determines what will be produced” (Buch-Hansen & Nielsen, 1977b, p. 8). But it also extended to considerations of the general purposes of other modes of production. As Buch-Hansen and Nielsen (1977b, p. 8) put it, this can be “production to fulfil the needs of the producer (some precapitalist modes of production); production to accumulate capital (the capitalist mode of production); or production to fulfil social needs (the socialist mode of production).” They emphasise that although the relations of production play an important role in forming territorial structures, in any mode of production it is primarily the development of the productive forces that controls the development of the territorial structure. Still, the saying “each mode of production forms its own territorial structure” (Buch-Hansen & Nielsen, 1977b, p. 5) virtually became a theme-tune for Danish radical geographers in the 1970s (Brandt, 1990).

Second, nature constitutes an “elementary condition” for the localisation of production. Nature is here understood as the physical or natural-geographical environment, and as related to changes in the modes of production. Since the use of the natural environment changes over time, the localisation of production and hence the territorial structure also changes. This potentially entailed a profound subordination of “nature”, which was not accepted by all radical geographers (see below). Using the historical-geographical development of Sweden’s wood industry as an example, Nielsen (1976a) contends that it is changes in the mode of production rather than factors in the natural environment that cause transformations in the localisation of production and, thus, the territorial structure. From this perspective, “nature, the geographical milieu,” is not without significance, but “the mode of production defines what at any time is fit and useful nature” (Nielsen, 1976a, pp. 75–76). Since capitalist development is predicated on constant expansion, a global chase for resources has contributed to what Buch-Hansen and Nielsen (1977b, p. 9) call “a global development of the territorial-structure”, linking otherwise disparate places to each other through the mode of production. This is theorised within the territorial-structure approach. But it was a focus point that was mainly developed within the radical geographers’ research on imperialism and underdevelopment, which was Marxist but generally not as structuralist as the territorial-structure approach (e.g. Enevoldsen, 1978; Fagligt Forums Imperialismegruppe, 1974; Folke, 1973).

Third, the territorial-structure approach pays attention to population as a factor for the localisation of production. “The human being itself is the most important productive force”, Buch-Hansen and Nielsen (1977b, p. 9) argue; “where there is no population, there is no production.” The population is in this respect primarily understood in terms of growth and density. This related to the idea that the greater the density of the population, the greater the possibility for the social division of labour. Aspects such as the geographically uneven distribution of labour reserves and the differentiation of wage rates can in themselves influence the localisation of production, something that is highlighted in terms of the historical relationship between town and country. For instance, it is argued: “In general, developments in population follows developments in production, i.e., that the distribution of the population is linked to the distribution of workplaces” (Buch-Hansen et al., 1979, p. 97). Furthermore, based on considerations of rural-to-urban migration, geographical variations in profit and wage rates, changes in land rents and general shifts in production towards industrial specialisation and agglomeration, it is argued that the development of the capitalist mode of production not only deepens regional differences between centre and periphery, but also actively generates “economic and social differentiation within the urban area” (Buch-Hansen et al., 1979, p. 102). Generally, it is maintained that the distribution of population plays an important role in any mode of production, and thus also influences the form and function of the territorial structure. Importantly, however, for the radical geographers this has implications for class struggle. For example, Nielsen (1976b, p. 43) notes: “The real wages of workers is a result of their activity in the class struggle,” and argues that “Due to the development of the mode of production, the fighting conditions become most favourable in precisely the same centres where capital accumulation and monopolisation take place most strongly.”

Fourth, the social superstructure is theorised as an “elementary condition” that influences location. The focus here is on the political, juridical and ideological aspects that shape territorial structures, not least in terms of planning and regional-economic policies from state authorities and other institutions with territorial dimensions. It is emphasised that there are several ways in which the social superstructure can influence the localisation of production, not least depending on which type of authority is involved. Somewhat archetypal for Marxist scholars at the time, this involves the relative autonomy of the state; in the words of Harvey (1976, p. 89) this is about “how State power can be and is used in a society which remains basically capitalist while constantly shifting and changing its institutional forms.” Similarly, within the territorial-structure approach, the state’s relative autonomy is connected to theorisations of the state’s role in capitalism. Since the state has some independence from the mode of production, regional-economic policies by the state may not always serve capital accumulation and may diverge from the requirements of the capitalist mode of production. While this is considered in theory, Buch-Hansen and Nielsen are sceptical towards such ideas and argue that while there is something to this argument, state policies will in general reflect the requirements of the mode of production and serve as the political foundation for transformations in the territorial structure. Using the Danish state’s infrastructural policies in the 1960s as an example, they argue that such transformations can happen though investments in infrastructure in support of industrial agglomeration or in relation to state subsidies for localisation in peripheral areas, but also through direct location of state institutions themselves: “The state, being the political instrument of the dominant relations of production”, Buch-Hansen and Nielsen (1977b, p. 8) argue, can thus “either further or inhibit a development in the mode of production.” Occasionally, the social superstructure is discussed in blunter, and perhaps more politically potent, terms. For instance, it is argued that: “The state apparatus is part of the superstructure through which political power is exercised. Through the state apparatus, the possessing class (the one who owns the means of production) exercises its political power” (Buch-Hansen et al., 1979, p. 24). For the Danish radical geographers, perhaps not surprisingly, the clearest example of this is how “the private ownership of the means of production is enshrined in law and enforced by the means of the state power bodies,” arguing that in this way “a ruling class can use the state apparatus to strengthen the economic and social foundation on which its power is based” (Buch-Hansen et al., 1979, p. 24). This echoes the well-known Marxist dictum that “The executive of the modern State is but a committee for managing the common affairs of the whole bourgeoisie” (Marx & Engels, 1998, p. 37), but it also attests to the revolutionary spirit that characterised much of the academic left in Denmark in the 1970s. Other aspects of the social superstructure, such as institutionalised religion and culture, are theorised as having an influence on location, for example through specific prejudices and habits that can influence population mobility or through resistance towards adopting a wage labour system or entering a specific type of commodity production. This, Buch-Hansen and Nielsen (1977b) argue, is often seen in shifts from one historical period to another, such as when a pre-capitalist mode of production collides with a capitalist mode of production. In other works, emphasis is placed on the coexistence and possible combination of different modes of production. As an example, it is highlighted that even in Denmark, “which is a developed capitalist country, we still find conditions created under the feudal mode of production” (Buch-Hansen et al., 1979, p. 24).

Fifth, and finally, the existing territorial structure is seen as playing a role in the development of new territorial structures. “Every mode of production attempts to create its own territorial structure to match the given relations of production and the given development in productive forces”, Buch-Hansen and Nielsen (1977b, p. 9) argue; “it is within the territorial structure that a given mode of production will itself develop in consequence of the continuous changes in the productive forces and in the relations of production.” In this sense, an existing territorial structure may hasten the development of new territorial structures because of relative advantages in expenditures towards infrastructure or production activities. Similarly, an existing territorial structure may also hamper or restrict the development of new territorial structures. For example, it is pointed out that feudal towns often constitute a limiting physical structure for the development of a territorial structure to fit a different mode of production. There is, in other words, an inertia of territorial structures. However, it is generally maintained that “Based on the economic and political laws it contains, a new mode of production will seek to transform both the physical territorial structure and the distribution of the population so that they fit into the new economic, social and political framework (Buch-Hansen et al., 1979, p. 33).

In a somewhat self-critical conclusion, Buch-Hansen and Nielsen (1977b, p. 10) stress that it is “a complex dialectical coalition of the factors” that constitute a territorial structure. Nonetheless, they strongly emphasise that it is ultimately development in the productive forces that determines the development of the territorial structure. Other conditions can only either hamper or hasten this development. Significantly, and well in tune with its time, considerations about territorial structures and the prospects of Marxist geography also connected to ideas about the anticipated, if not inevitable, transition from a capitalist mode of production to a socialist mode of production. For instance: “In the socialist countries (the transitional societies), the formal property rights to the means of production are state- or collectively governed, which can make possible the real breaking down of the class society” (Buch-Hansen et al., 1979, p. 23). Such discussions naturally led to an interest in comparing differences in state formations and modes of production in the socialist countries or, as they preferred, transitional countries (such as the USSR, China, Yugoslavia and Albania), and their expected progress towards a communist society (see, e.g., Buch-Hansen et al., 1979). But even the antithesis to capitalism was within the approach theorised as dependent upon the development of the productive forces, which ultimately led the proponents to argue that: “Only a massive development of the productive forces makes it possible to replace the socialist mode of distribution – to enjoy according to one’s labour efforts – with that of communism: to work according to ability and enjoy according to need” (Buch-Hansen et al., 1979, p. 171).

Reception, Fate and Wider Influence

The territorial-structure approach was pursued in some studies (e.g. Andersen et al., 1977; Buch-Hansen, 1976; Jørgensen, 1978). But its substantial impact was arguably the upper-secondary school textbook Om geografi (On Geography), first published in 1975 and written by a collective of radical geographers as part of Fagligt Forum’s aim to produce alternative teaching material. The book was also published in translation in the FRG (Buch-Hansen et al., 1982). “Territorial structure was one of Om geografi’s code words”, Alex Bredsdorff (1988, p. 12) later noted: “If one did not grasp that, one had a problem.” More than 24,000 copies of the book had been sold by 1984 (Document, 1986), and it became an important tool for the many radicalised geography graduates, who became upper secondary school teachers during the 1970s. But the book lost its appeal in the early 1980s. Bredsdorff (1988, p. 12) suggests that this was because many sections were too abstract or unclear (“probably as a result of internal disagreements in the writing group”), because the book “consciously neglected nature/the natural conditions”, and because a slow movement away from “the – declared – genuine ‘Marxist’ standpoints” required change. These points also reflect on the territorial-structure approach.

The role of nature – and physical geography – was the most visible discord in Fagligt Forum. As Karsten Duus Jørgensen (1983) later noted, this was a “sore point” on which virtual “trench warfare” had been fought, as it concerned the identity of the field and was “messed up in a web of political conflicts in pretty well all directions.” Spearheaded by the proponents of the territorial-structure approach, the majority in Fagligt Forum sought to develop an unequivocally social-scientific Marxist geography. This ruled out an “ecological” approach bridging human and physical geography, which for Nielsen (1976a, p. 78) entailed a return to the environmental determinism of “bourgeois geography” and, in the final analysis, “ideological support for capitalism.” Opposing this position was a smaller group of self-styled “dialectical materialists”, who argued that the territorial-structure approach amounted to a kind of idealism: “there is a danger in singling out the territorial structure and make it an independent object of analysis” (Brandt et al., 1976, p. 94). For these radical geographers, who, for instance, published translations of Karl Wittfogel’s Die natürlichen Ursachen der Wirtschaftsgeschichte in Kulturgeografiske Hæfter (no. 1, 1973; no. 9, 1976), “Reality is a whole, dialectical materialism perceives events in their context, and the sciences thereby come to overlap if they are to have any explanatory power” (Brandt et al., 1976, p. 93). There was, in other words, a place for nature as well as physical geography for the dialectical materialists, who from a radical point of view – and spurred by mounting environmental concerns – sought to maintain the “geographical experiment” of “keeping nature and culture under the one conceptual umbrella” (Livingstone, 1992, p. 177; in this book, see also Holt-Jensen, 2022). The dialectical materialists were marginal in Fagligt Forum, however, but some found a “sanctuary” at the new Roskilde University (Brandt, 1999).

From radical geographers closer to the territorial-structure approach, more immanent criticisms were voiced during the 1980s and 1990s. Steen Folke (1985) suggests, for example, that what eventually made many radical geographers reject the theory was its interpretation of history and the overwhelming role accredited to the development of the productive forces, which resulted in a kind of historical determinism and neglected the role of the relations of production. Frank Hansen (1994) similarly argues that the territorial-structure approach was too structuralist. The massive focus on the material side of social change made the theory mechanistic and deprived it of serious considerations of the role of political conditions and movements in the shaping of territorial structures. Furthermore, and contrary to its intention of guiding class struggle, the limited focus on social problems made the theory “action-oriented only on a very general political level” (Hansen, 1994, p. 113).

Beyond such criticisms, the territorial-structure approach came up against more fundamental changes. As Hansen and Simonsen (1984, p. 44) put it in the early 1980s:

Critical geography in Denmark is synonymous with one form or another of a Marxist approach to the subject. Therefore, it cannot surprise that the current problems in critical geography – apart from specific geographical fixations – parallel the theoretical problems the neo-Marxist wave today faces within the social sciences.

Rather than developing the territorial-structure approach or similar theories, Danish Marxist geography was gradually – as also happened in other contexts (Best, 2009) – diversified into (or replaced by) wider critical geographies during the 1980s. For some, as Andrew Sayer (1995) suggests for radical political economy more generally, this involved a shift towards middle-range theory and empirical research, particularly in the form of more narrowly focused economic geography. But more turned to perspectives that had little or no place in Marxist approaches. As Ole Beier Sørensen (1990, p. 75) put it, for example, “we say farewell to the big chromy statements and we to some extent leave the ‘grand theories’ behind. Instead we enter the microsociology of the everyday.” This was reflected in the topics that entered Kulturgeografiske Hæfter during the 1980s, and while the territorial-structure anchored Om geografi had been the textbook of the 1970s, the next collaborative textbook – written by radical pioneers as well as representatives of the next generation – was a signpost of a realised shift to more diverse critical geographies (Christiansen et al., 1991; in this book, see also Simonsen, 2022).

Conclusions

The territorial-structure approach was the most systematic attempt at theorising geography in Danish radical geography. Not all rallied around this theory, as we have seen, and it soon lost momentum. In revisiting the territorial-structure approach, our aim is not to resurrect it. For us, the key significance of the territorial-structure approach is that – through consent as well as dissent – it helped to mobilise an important theoretical as well as political movement in the formation of contemporary Nordic geography. Beyond bringing attention to a mainly forgotten piece in the histographies of (radical) geography, we find that a critical scrutiny of the territorial-structure approach – and the context in which it emerged – provides interesting perspectives on the development of socio-spatial theory and the situatedness of knowledge production. Besides this, the territorial-structure approach and the context in which it emerged affords a glimpse into a time when geography was more feisty and politically engaged than is perhaps the case today, something we presently miss, although it is also all too easy to succumb to uncritical nostalgia when narrating past perspectives that you essentially sympathise with, but not always agree with.

The territorial-structure approach was a conscious attempt at theorising geography historically and dialectically using social theory. One could question whether this theory was the best way to do that, as some did and more came to do. But the territorial-structure approach marked a radical departure from the traditional naturalisation of geography, not least at the Copenhagen department. Geographical space was seen as a product of history and social relations, and in its circumscribed manner, the theory was in this way an early attempt at applying what David Harvey (1973) termed a relational concept of space. However, as Frank Hansen (1979) – himself a radical geographer at Copenhagen University – bluntly put it in a criticism of Buch-Hansen’s (1976) early articulation of the territorial-structure approach, “It is throughout a physical structure that is defined.” But also on the more developed articulation by Buch-Hansen and Nielsen (1977b), Hansen (1994) finds that the theory did not adequately reconcile the relationship between the spatial and the social: a society produces a territorial structure, but it is unclear which role the territorial structure plays in the reproduction of society. Somewhat ironically, there is a traditional ring to the territorial-structure approach. “It is possible to assert that beneath the Marxist terminology there lies a plain statement of the traditional geographical enterprise”, Eyles (1981, p. 1377) notes on early attempts at Marxist geography: “There are geographical variations not within regions but within modes of production” (for a Nordic articulation of such criticism, see Vartiainen, 1986). Similar criticisms were also voiced by radical geographers in Denmark. Thyge Enevoldsen (1978, p. 12), for example, argued that by making the totality of spatial structures the object for Marxist geography, proponents of the territorial-structure approach eliminate the critical element of Marxism and “run the danger of degenerating to a bourgeois analysis with (borrowed) Marxist terms” (see also, Büchert et al., 1980; Nielsen & Rørdam, 1980). But in stark contrast to traditional geography, most of the radical geographers were keen to assert human geography as a social science. The territorial-structure approach might have been deterministic, but it was most certainly not environmentally deterministic. Still, the massive emphasis on the material in the territorial-structure approach entailed that “the dialectics between the social and the physical space is lost” (Hansen, 1994, p. 113). If not always with a radical agenda, other radical and critical geographies have done a better job of articulating truly socio-spatial theories.

Not least with the territorial-structure approach in mind, Kirsten Simonsen (2004, p. 526) notes that the inspiration from German social theory “resulted in an independent (but maybe also insular) development” in Danish radical geography. Indeed, generated independently from emerging radical geographies in the United States and the United Kingdom in particular, and strongly linked to situated struggles (and contingencies) in and around the radical-geographical movement at the Copenhagen Department of Geography, the territorial-structure approach was – with its strengths and weaknesses – an “original” contribution to early radical geography. At the same time, however, the theory had little “impact” beyond its particular place and time. Some of this undoubtedly has to do with language barriers. Most of the literature surrounding the territorial-structure approach, and Danish radical geography more generally, was produced in Danish. Only on rare occasions was material published in English, such as Buch-Hansen and Nielsen’s (1977b) Antipode paper. When Buch-Hansen and Nielsen’s (1977b) paper does get mentioned by English-speaking colleagues, it is primarily only in passing (e.g., Peet, 1979, 1983; Smith, 1979).