As the principal decision making body of the EU, the Council is the arena in which the interests of national governments are represented, promoted and defended. At the same time, however, it ‘locks member states into per manent negotiation with each other’ (Lewis 2010: 159). The Council therefore operates at two levels: it represents national interests but it also makes collective decisions for the EU; and the distinction between the two levels has become increasingly blurred. National officials who participate in this system face a ‘continuous tension between the home affiliation and the pull of the collective forum’ (Hayes Renshaw and Wallace 1997: 279). In this way, the Council of the EU needs to be clearly distinguished from other institutions of inter state cooperation which are always more inter governmental in the way they function. The Council’s effectiveness as a driver of regional integration, is due largely to its ‘collective memory’ and its inclination to build incrementally on previous decisions. The argument here is that what is referred to variously, as ‘path dependence’ (Pierson 1996) or the ‘joint decision trap’ (Scharpf 1988: 256) means that govern ments are locked into a process from which they cannot easily escape and in which choices become increasingly constrained. Forward momentum is the outcome of a process of bargaining and compromise from which no one gets everything they want, but everyone gets something. As the membership of the EU has increased, problems of overload, and account ability, have created the potential for near paralysis in decision making. As far as accountability is concerned the Council is caught in a sharp trade off between effectiveness and transparency. The more the Council yields to calls for more ‘openness’ in its deliberations, the more reluctant are ministers to speak freely and reach effective compromises that may involve concessions. As far as overload is concerned, the Council has shown itself to be pragmatic: one consequence is the devolution of more decision making to the EU’s Committee of Permanent Representatives (COREPER) (see below), and another solution has been more informal negotiation ‘in the corridors’ enabling consensus voting in the Council to be maintained at virtually the same level as before the enlargements of 1995 and 2004 (Haege 2013: 481-504). This institutional pragmatism safeguards the forward progress of integration across a wide spectrum of sectors. Much of what has been said here about the Council as an institution applies afortiori to COREPER, to which we now turn for a more detailed discussion. Due to the increasing workload of the Council, which we have already alluded to, much responsibility for making decisions has been effectively devolved to COREPER. Strictly speaking, COREPER is not formally a sepa rate institution of the EU but its modus operandi and unique blending of national and communautaire policy preferences make it especially relevant to any discussion of institutions as drivers of regional integration. Although, and perhaps because, COREPER does not have a high public profile, its work can be achieved quietly and effectively away from the scrutiny of press and politicians. The membership of COREPER consists of the permanent representatives, and their deputies, of the 28 member states in Brussels. The formal role of COREPER is to prepare for the meet ings of the Council, but in fact their function is now greatly enhanced to discuss, and reach agreement on, as many council agenda items as pos sible in advance. Matters agreed by COREPER in advance are placed under ‘Points A’ on the agenda (for rubber stamping by the Council) while unresolved issues are placed under ‘Points B’ for further discussion by ministers. Proposals for adoption by the Council start their deliberative journey in one of about 300 permanent working groups (WGs) established under COREPER. These groups cover all areas of cooperation under the treaties and are staffed by officials seconded from relevant departments in national administrations, from the permanent representations in Brussels, and from the Commission. Even at the level of working groups efforts are made to reach decisions that will command consensus, by exploiting the blend of ‘European’ and ‘national’ personnel within the groups them selves. Within this institutional architecture, COREPER occupies a special place: authoritative enough to make important decisions but sufficiently subordinate to leave the tricky ‘political’ issues to its superiors in the Council. The positioning of COREPER at the intersection between communautaire and national interests enables it to mesh together in its delibera tions these sometimes divergent or conflicting interests. The French word engrenage (used normally to describe the engagement of cogwheels in an engine) provides an apt metaphor for COREPER’s unique role; and this exemplifies in a very conspicuous way how institutions can, given a propi tious external political environment, drive forward the integration process. COREPER has developed a culture of negotiation that places a high premium on consensus and compromise. Lewis (2002: 291-292) has described COREPER’s style as ‘tempered by a dense normative environ ment’, its conduct as based on ‘diffuse reciprocity’, ‘thick trust’, ‘mutual responsiveness’ and a ‘culture of compromise’. COREPER members see themselves as bound by a dual loyalty – to their own governments but also to finding a communautaire consensus. The Roman god Janus, who faces in two directions at the same time, is commonly accepted as an appropriate emblem for COREPER. A distinctive esprit de corps has developed based on arduous work patterns necessitating long hours of close interpersonalcontact. A sense of intense collegiality has developed out of long meetings, marathon negotiations and shared responsibility for consensual decisions jointly crafted. An emblematic feature of COREPER’s existence is the role played by social activities (lunches, dinners, picnics, and longer trips to the country holding the presidency involving cultural, sporting or culinary activities) in bonding together members of COREPER and their families. The secret ingredient of COREPER’s modus operandi is the extent to which interpersonal relations, and bonds of mutual trust, forged inside and outside the COREPER meetings, play a constructive role in facilitating successful decision making. Not surprisingly, it was feared that the entry of ten new member states in 2004 might upset the well established esprit de corps for which COREPER had become well known. In fact, research has shown that careful preparation by the new entrants, and their role as ‘active observers’ for a year before entry had the effect of socialising them into the ways in which COREPER does its business. One effect of a larger COREPER was that more business was done informally outside the official meetings: another effect was that, with stricter control over speaking time in meetings, new entrants often aligned themselves with the presidency and/or commission view thus blurring any risk of a cleavage between old and new member states developing (Lempp 2007: 31-52). For the most part, therefore, enlargement had surprisingly little effect on the existing collegiality of COREPER. The linkage between this highly developed insti tutional ethos and the maintenance of integration across multiple policy sectors and 28 member states, is instructive in itself but it also reflects how well the regional integration process in Europe has become embedded in the broader society.