Even if scholars and policymakers are able to observe a consistent and coherent usage of the architecture metaphor (and we think this is somewhat unlikely), the problems with using this terminology will continue. Architects design buildings in which families, workers and other social groups exist. They provide blueprints for the construction of walls that keep people in. They use formal structure to define the environment. Even landscape architects use the structure of formal plantings and inanimate objects to organise a garden; they choose what goes in and what stays out. This approach presents problems for our subject: Asia’s future security order.

The architecture debate encourages a focus on processes and structures rather than relationships and outcomes. Some participants in the discussion favour an ASEAN-centred universe in which the ‘architecture’ is built on and around the existing multilateral processes championed by many of the 10 South-East Asian countries (see, for example, Goh 2007–08). Others suggest that the foundations of that architecture are provided by Washington’s set of bilateral alliances (see Baker 1991/92). Those with more pluralistic and flexible tastes suggest that we need to weld these together so that we can mix the hope of multilateral progress with the insurance of alliance (see, for example, Tow and Acharya 2007). While this combination reflects the inclination towards hedging strategies to cope with the uncertainty of Asia’s strategic future, it is a potentially unmixable set of ingredients.

The inclusion of US-led alliances in Asia as part of the regional architecture is problematic on theoretical and practical grounds. In theoretical terms, it suggests that these alliances contribute to regional order because of their nature as organisations (compared, for example, with what the ARF can offer as a different type of intergovernmental grouping). Alliances, however, contribute to regional order because of the promises (including promises of security assistance) that exist between their members. These promises condition the expectations of states inside and outside the alliance relationship. As Coral Bell (1991:46) argued some years ago in a short study of Australia’s alliance relationship with the United States, ‘International politics works, unfortunately, on nothing more substantial than a system of expectations.’ As shapers of expectations, alliances make their major contributions to order. They do not do so as organisational alternatives or supplements to multilateral diplomatic forums.

The practical difficulty in including alliances becomes clear when we remember that a regional architecture can be a set of walls that defines the conditions of existence for those who live within them. As such, it very quickly becomes a basis for containment and control. If one of the leading requirements for Asia’s future order is China’s effective participation in strategic relationships with other great powers in the region, an overarching architectural vision that incorporates the US alliance system (alongside everything else) could fall well short of the mark. While promising to provide an overarching framework for the management of Asia’s strategic challenges, such an architectural vision will do more to divide the region and raise suspicions. It would resemble a Western attempt to reassert control over Asia’s political evolution.

Each one of the architectural options assumes that order is a product of the way we organise ourselves. Here, order tends to be viewed as a function of external structure and not of strategic interaction, although somehow the latter is supposed to be inspired by the former. As an alternative to this organisational view of order, a relationship view focuses directly on the quality of the strategic interaction between the powers, rather than on the external structures within which their relationships are meant to develop. This means that the true building blocks of an Asian regional community are not a set of overlapping groups (from ANZUS and the Five Power Defence Arrangements through to ASEAN, the ARF and the EAS) and that the job of regional community building is not to work out which of these survive and how they might relate to one another. This is the nearly impossible task that Rudd and his colleagues have set for the APC on the basis that Asia’s existing attempts at regional organisations don’t yet add up to what we need (Rudd 2008a). In response to some of the criticisms within the region that have been levelled at the APC proposal, Rudd and his colleagues have also demonstrated the continuing appeal of the additive approach by pouring praise on the achievements of ASEAN, which they have retrospectively claimed the community can build on (see Smith 2008).

This mistakes the pots (which are simply containers) for the plants (where the true life exists). The real building blocks are the relationships between the powers highlighted in their strategic interaction. Here, the most important institutions of regional politics are not the formal organisations that hold regular summits, but the rules and patterns of behaviour that operate between the major actors on a daily basis. Such an informal approach would suggest that the basis of Asia’s strategic future, including China’s role, is not a regional architecture—which seeks to organise and perhaps even control the actors—but a set of regional bargains that nourish and support their most important strategic relationships. ^[2] One of these bargains is an effective but informal compact between the United States and China that they will recognise each other’s leading role in regional affairs.

It is entirely possible that formal constructs could develop from these bargains: for example, a set of bargains between Asia’s great powers might then be reflected in an Asian security mechanism of some sort involving China, the United States, Japan and India. There is, however, no reason to suggest that the relationship will work in reverse. One might establish such a forum only to find that the powers that have agreed to attend do not regard one of the others as a legitimate participant. Similarly, in domestic politics, the formal trappings of statehood rely on an informal bargain that is the basis of a social contract. (Bull [1977:59] saw pre-modern African societies that were stateless as a ‘spectacle of “ordered anarchy”’.) Similarly, we need to insure ourselves against believing that the cooperative relations between Europe’s former great powers since World War II have been due to the formal institutional relations between them and not to their political determination to limit the possibilities for dangerous competition between them, which allowed these regional organisations to develop. Likewise, for ASEAN to exist in the first place, its original founders needed an informal but robust bargain between them that represented their mutual interests in unmolested sovereign independence. That original and informal bargain is infinitely more powerful and important than ASEAN’s more recent and more formal charter, which has attracted a mix of admirers and sceptics (see Desker 2008b).

A similar logic can be applied to the argument that the best answer for Asia’s evolving power equation is a concert in which power is as much shared as contested. There is much to be said for this approach (see, for example, White 2007), at the heart of which is an informal agreement between the great powers to collaborate in the management of the relations between them. Indeed, Bull (1977:74) nominated the ‘managerial system of the great powers’—which could take the form of a concert or a condominium—as one of the institutions of international order, by which he did not mean formal parts of an architecture within which the great powers were housed as essential pieces of the furniture. Instead, the institutional aspect relates to the patterns of behaviour among the great powers. The institutions here are more about practice and relationships than about formal structure. Bull (1977:74) explains that ‘[b]y an institution we do not necessarily imply an organisation or administrative machinery’ (which is so commonplace in the contemporary discussion of Asia’s ‘architecture’), ‘but rather a set of habits and practices shaped towards the realisation of common goals’.

Instead of thinking about which gatherings, arrangements, forums and processes can combine to produce Asia’s future order, we are better to focus on what habits and practices the great powers need to adopt—and how smaller and medium powers in the rest of the region (such as Australia) can encourage that adoption. At the heart of the European concert of the early nineteenth century was not an ornate edifice of formal diplomacy (the trappings of Vienna) but a common realisation by the great powers that what Napoleon had been allowed to do to the regional balance could not be allowed to repeat itself. ^[3] Similarly, while Washington’s diplomatic recognition of Beijing marked part of China’s formal entry into the family of nations in the 1970s, the real work was done in the earlier bargain engineered by Henry Kissinger (a disciple of Metternich) and Zhou En-Lai, which gave informal but very powerful recognition to the triangularity of Asia’s strategic relationships (see Isaacson 2005:333–54).

The arrival of a genuine concert in Asia is no easy matter (Acharya 1999). For a number of the reasons outlined above, to stand any sort of chance it would probably need to begin in an informal fashion—as a series of mutually supportive great-power bargains—before the architects gained the opportunity to formalise (and often to de-energise) the cooperation. In today’s climate, if one or more of the great powers made a serious proposal for a concert, it would most likely be treated as yet another architectural option. We would immediately shift to a discussion of when (and more importantly where) the first such annual gathering should be held, what its agenda should be and who should be invited (and left off the list).

Despite sporadic and very painful progress (and sometimes no progress at all), the Six-Party Talks that Beijing has hosted to address North Korea’s nuclear weapons program offer a very subtle, almost undetectable, hint of great-power concert. Three of the major players—China, the United States and Japan—are on board, and signs of convergence between the positions of Beijing and Washington (if not shared by Tokyo) have perhaps been the main benefit of this process. Again, the value of the Six-Party Talks lies not in another architectural option for Asia’s future order, but in the quality of the relationships between the great powers. An architect might propose that these talks are the building block or the scaffolding for an eventual East Asian security mechanism. The notion of a ‘6–1+3’ approach (North Korea out and India, Australia and New Zealand in) presents, however, an architectural illusion that needs to be resisted (see, for example, Kessler 2006). Instead, the magic, such that it exists, lies in the very small and pale embryo of concerted behaviour—a pattern that might just be utilised further afield. Policymakers should aim to protect the relationship and not the organisation.