1. The "summit process", or "summit system", which is the basic unit of analysis of this study, comprises the actual annual and special leaders meetings of the seven, the dedicated host country summits, sherpas meetings and sherpa network consultations that form the preparatory process for the summits, and the subordinate ministerial and official institutions and meetings linked to, or created by, the summit. This latter category includes the Group of Five and Group of Seven Finance Ministers meeting (given their origins in the same "Library Group" meetings that bred the summit) and Deputy Finance Ministers (G7DM), the Trade Minister's quadrilateral, the Foreign Ministers consultations during the United Nations General Assembly, and working groups on specialized subjects such as terrorism. Except as noted above, it excludes institutions with a lesser or greater number of countries (e.g. G3, G10, Berlin Four, Guadeloupe Four) and those which existed prior to the emergence of the summit from its Library Group origins. There are, of course, important questions about which part of the seven power system is most effective or appropriate for various purposes, and how the summit competes, co-operates and compares with other institutions such as the IMF, IBRD, and OECD.

2. A shock (defined as a sudden, severe, surprising and costly challenge to the prevailing international order and the stability of its great power protectors) is like an acute international crisis (defined by low time to respond, high surprise, and high threat to national values) in that it unanticipated by the affected governments, involves a sharp, abrupt shifts in the magnitude of actions in the environment, and poses immediate and major harm on them. A shock, however, unlike a crisis, does not require an immediate response (in the order of days or weeks). More importantly, It constitutes a threat not to a single or even several national actors but to virtually all countries in a group. As a result, shocks require objective transformations in national or international actions, interactions or transactions in the environment (involving commonly observable large magnitudes in short times? rather than merely the subjective interpretation of a few decision-makers in a single country (who cognitively create a crisis). A shock is thus an objective event, whose magnitude (severity) and content (originators, values threatened etc.) can be objectively compared to other shocks.

3. Interdependence is here treated as a single, static scope condition of the system, rather than as a fifth element that can vary dynamically to create or destroy a system and (above the floor that creates summits) generate varying levels of summit success. Such an analysis would consider how relative capabilities are transformed into varying vulnerabilities and sensitivities across concert members, and how some concert members (being less interdependent with the group) are able to insulate themselves in the absence of, and from the need for, a collective response.

4. Because the sherpa's agree that Rambouillet was the best summit of all time, it is easy to establish comparability with the Putnam and Bayne scale by fixing Rambouillet as A in the sherpa consensus scoring (as it is in Putnam and Bayne's). In the sherpa consensus scale, subsequent summits are scored by comparing them in the sherpa's judgments to Rambouillet and other summits, and by comparing them directly (in the judgment of Hunt and Owen) to the Putnam and Bayne grade.

5. Positions are defined as any prescriptive or descriptive statement linking a discrete subject or object with a verb (e.g. the MTN round should be concluded, growth is returning). Position areas are clusters of positions that are grouped together by the leaders in their documents, and separated from other such clusters, but their common or directly - related content, paragraphing, numbering, or assigned titles. The number and identity of position areas have been identified with an intercoder reliability of 83%. Additional, measurable, potential causes of summit success (independent variables) are: the summit experience of the leaders at the summit table; their knowledge of international economics through previous experience as finance ministers; their freedom from national electoral preoccupations in the immediate pre-summit or post-summit period. In addition to such individual-level variables, there exist such institutional level variables as the strength (density, frequency) of the supporting institutional/preparatory process, and the background of the sherpas, and such transnational level variables as macroeconomic performance (growth, inflation, unemployment, productivity). Additional, measurable, relevant indicators of summit performance (dependent variables) are the re-election of participating leaders and the creation by the summit of supporting institutions and processes.

---

Source: Prepared for the annual meeting of the International Studies Association and the British International Studies Association, London, March 29-April 1, 1989. Unpublished in print. Reproduced by permission of Professor John J. Kirton.