In his book The Territorial Peace, Douglas Gibler argues -- as I understand his argument based on his recent posts at The Monkey Cage -- that the 'democratic peace' is "a subset" of a more encompassing and fundamental phenomenon. He contends that involvement in territorial disputes (especially those involving core 'homeland' territories) leads to authoritarianism and 'centralization' (or pushes in that direction), whereas countries that are not involved in territorial disputes tend to be less authoritarian, more democratic. The reason is that militarized disputes over territory (defined, per the Correlates of War project, as anything from a brief display of force to full-scale war) produce large armies, which subsequently are often used for internal repression and more 'centralization' (measured by the number of 'veto points' in the state apparatus). An absence of territorial disputes has the opposite effect, thus leading to both democracy and peace. Questions can be raised about aspects of this argument (see e.g. the comment thread to this post and also further discussion below), but it seems intuitively somewhat plausible, or at least not completely implausible.
Gibler's work can be seen as part of a recent wave of scholarship which, in different ways and from different perspectives, addresses the effects and causes of an overall decline in armed conflict, especially traditional interstate war. Work on the territorial integrity norm (Zacher 2001 [pdf]) and the rarity of 'state death' after 1945 (Fazal 2007) attributes the reduction in interstate war to norms concerning the inviolability of state boundaries and the unacceptability of conquest. (Arguments about the obsolescence of great-power war, discussed elsewhere on this blog, also focus on norms and their development.)
Not everyone agrees, however, that settled territorial boundaries always lead to less conflict. Boaz Atzili in his book Good Fences, Bad Neighbors argues that (to quote from the abstract of an earlier article of his):
In regions in which most states are socio-politically strong, fixed territorial ownership is a blessing. It enhances peace, stability, and cooperation between states. In regions in which most states are socio-politically weak, however, fixed territorial ownership is largely a curse. It perpetuates and exacerbates states' weakness, and contributes to internal conflicts that often spill over across international borders.Atzili defines "the sociopolitical strength of the state...as the state's capacity to maintain a monopoly over the legitimate use of force,...rule effectively over its society (including extracting sufficient revenue and providing sufficient public goods), and...maintain a reasonable level of social cohesiveness and identification of its residents with the state as such" (Good Fences, p.33). Thus he treats "the ideational facet of the state" as "just as important, and sometimes more" important than the institutional dimension (p.4).
His basic argument, as the above quote indicates, is that the norm of fixed borders often perpetuates state weakness, which in turn facilitates internal conflict that can spill over boundaries and become a form of interstate conflict, not for the most part "Clausewitzian wars in which two regular armies meet each other in the battlefield" but "transnational conflicts" involving state and non-state actors (p.49). The border-fixity norm thus has different effects depending on the strength or weakness of states.
Good Fences, Bad Neighbors contains a number of case studies. Two of the four main cases -- Brandenburg-Prussia in the 17th and 18th centuries and Argentina in the 19th century -- predate the border-fixity norm, while the other two main cases -- Lebanon 1950-2006 and Congo (DRC) 1960-2006 -- are set in the fixed-borders world. Space and time preclude anything like a proper summary of the cases and of the various dimensions of the argument; however, a glance at the Congo discussion will give a flavor of the approach.
What is now the Dem. Rep. of Congo was "a very weak state at its independence" (p.141) and, with its existence effectively guaranteed by the norm against conquest, it did not face the same structural pressures and incentives to become a stronger state that polities in the 'flexible-borders world' of early modern Europe did. Mobutu's corrupt and kleptocratic rule had much to do with keeping Congo (then Zaire) weak, but Mobutu's successors Laurent Kabila and his son Joseph Kabila did not improve things greatly, because incentives for state-building remained largely absent. When Congo's weakness met the Rwandan genocide of 1994 and its aftermath, which sent large numbers of Hutu perpetrators across the Congo border to form "a state in exile" (p.184), the result was the long war that drew in a number of Congo's neighbors.
The argument about state weakness and border fixity, we are told at the outset, is "probabilistic rather than deterministic" (p.8). And one sees this illustrated, for example, by Tanzania, which, unlike Congo/Zaire, used "forceful policing and efficient sealing of the border by the Tanzanian military" to prevent Rwandan Hutu refugees in western Tanzania from staging attacks into Rwanda (p.184). In other words, Tanzania, existing in the same international normative environment as Congo/Zaire and facing the same structural incentives (i.e., no prospect of 'state death'), became a somewhat stronger state than Zaire. This does not invalidate Atzili's argument, since he acknowledges that outcomes may vary depending on leadership and political culture (p.9). But he maintains that leaders of weak states in a 'fixed-borders world' have a more difficult job of state-building than leaders in a 'flexible-borders world' had: "The task of building strong states in a world of fixed borders is daunting" (p.220).
On p.39 of his book Atzili discusses Gibler's article "Outside-In: The Effects of External Threat on State Centralization" (Journal of Conflict Resolution, 54:4, 2010). Atzili criticizes the article on several grounds (noting, for example, that he uses a "broad and holistic concept of state strength" in contrast to Gibler's focus on centralization), but also says that Gibler's findings support one of his (Atzili's) hypotheses, namely that "a world in which there is no normative prohibition on conquest and annexation (flexible-borders world) is likely to result, over time, in sociopolitically stronger states" (p.36).
Yet it seems to me that Gibler and Atzili approach the whole problem from somewhat different angles not only in terms of their methods (which is clearly the case) but in terms of the causal arrows (causal mechanisms, if you prefer that phrase) that each sees at work. For Gibler, the absence of territorial disputes -- as indicated by, among other things, settled boundaries -- leads to democracy and peace via less 'centralized' states. For Atzili, settled boundaries produce or enhance peace only under certain conditions, namely the presence of 'strong' states, where 'strength' is understood not as 'centralization' but more broadly, i.e. as a state's overall capacity and legitimacy.
In terms of policy recommendations, Atzili would not get rid of the border-fixity (territorial integrity) norm, since in large parts of the world its effects are positive, nor does he advocate returning to the era of territorial wars. He suggests what are, in effect, less drastic steps to put pressure on weak states to engage in state-building, such as the threat of ejection from international organizations for "states that cannot be considered states by any positive measure (such as Somalia and the DRC)" (p.220). He also suggests that "in some cases state building may need to take precedence over democratization" (p.220). I'm not sure what I think about this or indeed about Atzili's general argument: obviously I think it is interesting enough to blog about, but I have certainly not read the book with the care that would be required to reach a considered judgment. (Perhaps I will have some additional thoughts later.)
I'm going to leave it here, without a tidy conclusion. Comments are welcome, including those politely telling me that I'm confused and have got things all mixed up.