ntion is "supposed to fix blame on the minister heading a department for every failure of departmental policy or administration, whether it is the minister himself who was at fault, or a civil servant, or if the failure resulted from a failure of departmental organization. The minister must, in this orthodox version, submit to the judgement of Parliament and, if the failure is a serious one, should resign from office without waiting for a vote of censure."2 Recently, the convention has been diluted, and the refusal of ministers to resign in the event of various departmental failures in the last 20 years epitomizes this dissipation of the doctrine, which seems to vindicate the proposition that it is indeed a fiction. In the aftermath of the Maze break-out3, Mr James Prior refused to resign, saying that there was "no clear rule and no established convention [of ministerial responsibility]" But Mr Enoch Powell ostensibly acted as an advocate to the convention claiming that, "if the responsibility can be abjured by a Minister, a great deal of our proceedings in the House is a beating of the air because we are talking to people who, in the last resort, disclaim responsibility for the administration." It is a convincing argument from Powell but one which could have been made for reasons other than a mere affiliation for the existence of such a convention. Powell was sitting as an Ulster Unionist at the time and perhaps was more interested in inducing a policy reversal in relation to Northern Ireland itself.1 Nevertheless, he was admonishing the convention in the circumstances, having done so on previous occasions.2 However, evidence of an established convention itself is not a corollary to an utterance in the Commons. The crucial issue is that no minister resigned, the prison governor did. This was an example of what Andrew Barker calls a "practical view of administrative ministerial responsibility."3, a mut...