Fenced in by a borderline idea

Geopolitics has had a variable reputation within and beyond geography - as indeed has political geography, within which it is normally placed. In the late 19th and early 20th centuries, the latter was sometimes equated with the entirety of human geography (basically, it provided a gazetteer of the natural and human-made characteristics of the Earth within the template provided by state boundaries). But it later attained a separate standing through the works of Halford Mackinder, Isaiah Bowman and others on the geography of political power and state foreign policies.

Political geography and geopolitics - to some, especially Russian geographers, the two were synonymous - was largely discredited in the middle of the 20th century because of the association between the work of the German political geographer Karl Haushofer and the territorial claims and strategies of the Nazi regime. As late as the 1980s, it was not possible to create a political geography commission within the International Geographical Union because of Russian opposition: the compromise was a Commission on the World Political Map. Political geography was by then enjoying a revival in the Anglophone world, however, as was geopolitics - although in a very new form.

Klaus Dodds has been a leader in that revival: Global Geopolitics is his overview of its contributions to understanding how states are involved in the manipulation of spaces, places and territories - not just how these appear on maps but also how they are presented and how they change with events. Contemporary geopolitics is about visions of spaces and places, of "us here" and "them there" (within the context of what some present as an increasingly "borderless" globalising world). The impact of Edward Said is etched deeply into much of that literature.

From an admitted Eurocentric viewpoint, and using the events of September 11, 2001, as a foundation for much of the discussion, Dodds presents a clear overview of the field, ranging widely in subject matter and with many well-chosen exemplars of the interactions between geography, knowledge, power and political institutions. The lack of much on the geopolitical views of "them" means the book is, however, necessarily partial - unfortunately so, given Dodds's expertise on aspects of Latin American geopolitics. But there is little doubt that this will be a successful text that will probably have to be updated frequently as the world changes rapidly (a first version appeared - as Geopolitics in a Changing World - in 1999).

Territory is a key element in geopolitics: its role is crucial because sovereign states are territorially defined and use territoriality strategies as an important means of controlling people. David Delaney's short introduction to this concept is thus welcome and will be useful supplementary reading for many courses.

It starts well with two introductory chapters on territory and different approaches to its study in a range of disciplines. The other two chapters are less satisfactory: one is a 30-page precis/ exegesis of Bob Sack's seminal 1986 book on human territoriality, the other a detailed discussion of the deployment of territorial strategies in Israel-Palestine. The latter is excellent, but deals largely with one spatial scale only. Other exemplars, such as South African apartheid practices and forms of "turf politics", would have ensured a more rounded volume.

Mark Blacksell's introductory text on political geography as a whole illustrates the problems of such brief primers. Given the breadth of material to be covered, not surprisingly geopolitics gets 15 pages and national foreign policy 18 (without, like Dodds, any mention of the domino theory) - and territoriality gets just a two-page summary. The writing is clear but the coverage uneven - much more impressive on some topics, such as the politicisation of the oceans, than others, such as the rationale for the state as a territorial body - and there are some factual errors.

The book is presented as illustrating "what fun it can be to try and take a synoptic view of this vast and varied area of academic study that political geography has become". The attempt to be all-embracing means there is a lack of depth throughout, so the book is insufficient to take readers very far (the chapter on electoral geographies is particularly disappointing), and this may well mean it has little success as a textbook.

Ron Johnston is professor in the School of Geographical Sciences, Bristol University.