Abstract
I restate my argument as a critique of lexical conservatism and the so-called history of the present as a history of words. This leads to a discussion of what we mean by genealogy and agency in contemporary theoretical work. I then briefly respond to some of the key points raised by the commentators.
Listening to my interlocutors at the 2014 Tampa Association of American Geographers (AAG) session made me more aware than previously of what motivated me to write the piece for the lecture in the first place. So, I do not propose replying to the many specific points raised in the commentaries one by one, the authors clearly express their own well-argued and defensible positions on the general argument that I made. I would rather address some of the more contentious specific issues that crop up in the commentaries. I hope that in so doing I can make the intentions of my argument as clear as possible without in any way diminishing or trashing the alternative and well-articulated views of others. Before I raise the various specific issues I want to revisit and restate my major points in somewhat different language from that in the original piece reflecting a few general observations I made at the AAG session.
My view about the role of the history of ideas in human geography reflects a constructivist realist position that I have long held but that I fear is being sidelined by either discursive constructionism (the main target of my critique) or newly invigorated realist empiricism (Agnew, 2001). To cut to the chase, I am skeptical of the relevance of tracing the history of words as putative concepts over long arcs of time in terms of present-day applications of those words as all we need to understand contemporary social and political practices. Such words can have continuing constitutive powers, I would not dispute that this could be so as Matt Farish perhaps suggests I do, but, and this was the reason for citing Ian Hacking, I do not think that these words always do what we need them to do or are sufficient in themselves in a world that is real and open and hence cannot always be understood in yesterday’s terms. I am thus objecting to two features of contemporary thinking in human geography that I think will be ultimately futile for the field as a whole: the notion that it is past meanings of words that should be a guide to understanding their uses in the present and that single words themselves when isolated for study and given canonical meanings can entirely capture the reality to which they putatively refer and with which we are primarily concerned.
The first feature involves a lexical conservatism that undermines the creativity of making our own theories of the world in the interest of reproducing those from the past. I have encountered this in my own writing on sovereignty to the extent that my attempt at stretching the word to encompass a number of ‘sovereignty regimes’ has been dubbed ‘lexical promiscuity’ by those completely ensconced in the Hobbesian and Lockean traditions for which sovereignty is held to be tied entirely and eternally to a territorially bounded state. Negative reactions to my efforts to pluralize the meanings of region, geopolitics, and territory from singular historic precursors with respect to present-day relevance (Agnew, 1994, 1999, 2004) have usually followed similar tacks.
It seems to me that the logic of writing ‘the history of the present’ through tracing word meanings across historically selective theoretical texts indicates two problems with the history of ideas in its various manifestations as I have discussed the approach in relation to the three geographical concepts. One problem is a peculiar rendition of ‘genealogy’ as a method (when mentioned explicitly in connection with the lineage of the word territory and when implicit in a particular tracing of a word such as region), when I would have thought that a writer such as Foucault with whom the term is often closely associated and Nietzsche who seemingly coined it as something to criticize would probably wish to have nothing to do with it as such (Stevens, 2003). In popular usage, the word perhaps does license the search for origins and definitive meanings, but Nietzsche was profoundly critical of such considerations and Foucault was, if nothing else, a critic of the transcendental subject and its objectifications. I like to think that Foucault might approve of my lexical promiscuity more than the sectarian textualism that has long been de rigueur in disciplinary tracings of key concepts. The second problem is that lexical conservatism can lead to discursive determinism. What I mean is that the possibility of authorial agency is denied when the roles of agency and autonomy in any discourse are elided. As Mark Bevir has plausibly argued in his discussion of agency and autonomy in Foucault, but as with genealogy the point is more general:
A rejection of autonomy need not entail a rejection of agency: we can say the subject always sets off against a social background that influences him, and still insist he then can reason and act in creative, novel ways so as to modify this background. (1999: 68)
The second feature of much contemporary thinking, not about the concern with words with either old or newer meanings but about what exists beyond them, arises because of the priority given to epistemology over ontology in contemporary human geography (and social and political theory). What I mean by this is that we prefer writing and talking about ‘experience’ or what is written and said (how we know) rather than discussing the layering of reality in which multiple agents draw on discursive and material resources to effect actions (what exists) be they directly political (e.g. territory) or in a research process (e.g. geopolitics and region) (Connolly, 2014). In terms of my constructivist realism, Stuart Elden’s recourse to epistemological excavation of the word territory as the preferred technique for addressing the so-called territorial trap, for example, simply misses much of the problem as I have defined it. As Simon Reid-Henry (2010) has carefully noted, the entire point of describing territory in the dominant meaning it has acquired as being a ‘trap’ combines an epistemological critique of the reductive nature of thinking in the field of international relations theory that has made an eternal ontological form out of a historically contingent idea of how ‘best’ to organize a polity with an ontological critique of the anchoring of states in the closed world of interstate relations in which the working of power at other scales and across networks is essentially ignored so as to define a field of study and better ‘model’ the so-called interstate relations. In my view, ignoring the latter in a world context in which ‘territory’ takes on an increasingly different meaning from that of the 17th-century theorists is to give up intellectual agency in encountering our present world to a determining discourse inherited from the past. Theories of geopolitics, territory, or whatever are not the same as the world they tentatively try to explain.
This does not mean that we should entirely abandon attempting to understand the history of the concepts that we have inherited from the past because the ontology of the world we live in changes and past theories and concepts only offer partial matches anyway. What it does require, however, is that even if we are still enamored of a given word, such as territory, we need to focus our efforts on reshaping the concept and reorienting the discussion to get a better match between concept, related ones (space and place) and reality over time (e.g. Agnew, 2009 on sovereignty; Banai et al., 2014 on territory; Jerram, 2013 on space). Discovering past tracings can only be a small and potentially misleading part of this research program. In fact, more valuable is theoretical innovation based on juxtaposing concepts and developing cross-fertilized versions of them that can inform metaphorical applications and extend or stretch terms in new directions. Many years ago, for example, David Livingstone and Richard Harrison (1981) proposed a ‘metaphor-myth-model’ to develop a geographical epistemology alert to the complexities of the social and political worlds with which human geography is involved. We should not just trace the history of the concept of ‘frontier’, one of their examples, but also actively try to expand and apply it in new ways and in novel situations.
Beyond these major points of clarification and reiteration, I would like to respond briefly to some of the specific issues raised by the commentaries. My critique of geopolitics is not intended to be directed at all to ‘critical geopolitics’, even though I do think that into the 1990s this new conception of geopolitics did tend to privilege looking backward to deconstruct classical geopolitics rather then expend sufficient energy coming to grips directly with the geopolitical present. Be that as it may, my intent was certainly not to rubbish what in my view has been arguably the most important innovation in relation to ‘geopolitics’ in the past 60 years. I am not clear how I could be construed as doing so. The focus on practical reason in studies of foreign policies by critical geopolitics hardly privileges theoretical ideas inherited from the past. It engages with them only when they have putative practical renderings and consequences. I would also disagree that the examples I present are relatively ancient ones and thus that the argument does not apply to the field today. My claim, which I think does have empirical support, is that the argument I make is of both long-standing and current relevance. I do agree that much of what goes for the history of ideas, if you agree with my conception of it as Wickberg (2001) defined it, is Eurocentric and largely confined to sources in a limited range of languages. I do not believe, however, at least in this limited area that there is excessive English-language bias. Many of the leading sources come from Latin, French, German, and other European languages. Nevertheless, the world–regional and gender biases are obviously serious. I would also dispute the claim that my purpose is to police the field in any manner, certainly analogous to Hartshorne in his heyday. Far from it, my intent has been to argue for and justify a critical pluralism in relation to concepts. Finally, I would acknowledge that my argument could be used to say that there has been a displacement of effort in contemporary human geography toward an obsession with the provenance of ideas and writing in an excessively ‘theoreticist’ mode. This again is an empirical question. I think that it does have some justification, even though I agree that not everyone needs to be doing solely what I recommend and that some should be engaged in intellectually excavating the past.
Even so, to usefully dredge the past both to understand ideas and their influence in past times and their uses today I think needs paying much more attention to the politics and geography of knowledge rather than to textual exegesis of selected theorists. In this regard, I would suggest engaging with the approach to intellectual history pioneered by Reinhart Koselleck (1985) geared specifically to examining the historical process of how old words are taken and given new meanings in new historical–geographical contexts. This draws attention to the active struggles between schools of thought, academic generations and political factions over concepts and the meanings to be ascribed to them (e.g. Agnew et al., 1996). In particular, concepts arise from and diffuse across places. There is therefore a geography to how ideas take root, are spoken and read about and spread over space and time (Agnew 2007; Livingstone 2007). Looked at in this way, the history of concepts makes the concepts available to contextual cultural and political analysis rather than reducing them to the simple tracing of words over time as they crop up in the writings of canonical authorities.