The execution of air warfare is now an extremely complex process because of the sophistication of the avionics and weapons systems used (Bacon et al., 1990; Clancy, 1996; Nordeen, 1999; Rendall, 1998; Spick, 1995, 2000a, 2000b; Thornborough, 1995; Walker, 1989). The complexity can be judged by accounts of Desert Storm (Clancy and Homer, 1999) and by more staid documents like the RAF doctrine outlined in AP3000 (Ministry of Defence, 1999). The complexity of the on-board systems has increased in response to increasing threats from enemy forces both in the air and more recently on the ground. There have been repeated statements concerning the need for more effective cockpits which meet the needs of a increasingly reduced crew complement (Penney, 2000b). However, there is increasing recognition of the need for a balance of power between the technology and the human (Dudfield, 2000; see Penney, 2000b). The critical issue discussed in this paper is how that balance is to be managed and whether it can be dynamically changed. This balance critically depends on the automation policy applied to on-board systems and this paper explores the use of automation with regard to the specific area of self-protection.
