British Journal of Industrial Relations 44:4 December 2006 0007–1080 pp. 601–604 © Blackwell Publishing Ltd/London School of Economics 2006. Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. Blackwell Publishing Ltd.Oxford, UKBJIRBritish Journal of Industrial Relations0007-1080Blackwell Publishing Ltd/London School of Economics 2006December 2006444601604Original ArticlesNew Actors in IRBritish Journal of Industrial Relations Edmund Heery is at Cardiff University. Carola Frege is at the London School of Economics. New Actors in Industrial Relations Edmund Heery and Carola Frege Famously, John Dunlop declared that an ‘industrial relations system’ com- prised three sets of actors: ‘(1) a hierarchy of managers and their representa- tives in supervision, (2) a hierarchy of workers (non-managerial) and any agents, and (3) specialized governmental agencies (and specialized private agencies created by the first two actors) concerned with workers, enterprises, and their relationships’ (Dunlop 1993: 47). This is a broad definition that potentially can encompass all sorts of actors that engage with and seek to represent interests within the employment relationship. In Dunlop’s own work and in the subsequent development of the field of industrial relations, however, the primary focus of analysis were the collective actors, trade unions and their management counterparts, and their interaction through the pro- cess of collective bargaining (Kaufman 2004: 254–5). State policy was of interest to the extent that it supported or hindered this process although the main emphasis, at least in the Anglophone world, was on the relatively autonomous behaviour of unions and employers and how this was shaped by product markets and technical change (Howell 2005: 7–12). This narrow definition of the field as the study of union–management relations continues to exert influence today — despite editorial statements suggesting otherwise, referees of articles submitted to BJIR that do not deal with trade unions still occasionally question whether they are relevant to the journal. It is increasingly subject to challenge (e.g. Edwards 2003; Osterman et al. 2001), however, and arguably is untenable given the decline of the labour movement and collective bargaining in countries where industrial relations has been institutionalized as a separate field of study. One response has been to tilt the subject towards the study of HRM and in the USA, in particular, management strategies of labour use have moved to the forefront of research (Frege 2005: 189–93). The emergence of the high performance paradigm, with its primary research interest in the link between HR practice and business performance, is the clearest indication of this shift (Godard and Delaney 2000). In certain respects, however, this also represents a narrowing of the field; a research focus on one dominant actor and a normative orientation that prioritizes business interests. A second response is to retain the concern with institutional actors and a plurality of interests, enshrined in Dunlop’s initial formulation, but to update