Forder (2014) has recently put into question the standard narrative, as put forth in Friedman’s (1977) classic contribution, of how the Phillips curve was employed as a menu for policy. While Friedman had it that the Phillips curve was employed as a piece of analysis supporting inflationary policies as a measure to maintain full employment, Forder argues that barely any trace of this interpretation of Phillips (1958) can be found in the academic literature of the period. One of the arguments employed by Forder in his case against Friedman is that the very idea of the trade-off was either ignored or overtly rejected in the academic literature that followed the publication of Phillips’ (1958) article. In this paper, we take a different perspective, more focussed on the policy debate, coming to a different (but not opposite) conclusion. Building on a hypothesis that has been put forth but not developed by Wulwick (1989), and following analogous threads provided by Brown (2000), Lipsey (2000), Holt (2000), Sleeman (2011), and Bollard 2016 (among the others), we show that Phillips (1958) came out in a context in which the trade-off that was circulating as a menu for policies was the opposite of the one popularized by Friedman as typical of the short-run Phillips curve era. The reactivation of monetary policy in 1951, the publication of the White Paper of 1956, and Thorneycroft’s ‘September measures’ of 1957 triggered a debate in which the idea that stable prices could only be obtained at the cost of rising unemployment was tentatively (re)introduced. More specifically, we show that this kind of ‘reverse’ trade-off emerged as one of the positions among the evidences submitted to the Radcliffe Committee on monetary policy in 1957 and 1958. Many of these evidences came from LSE colleagues of Phillips, who argued in favour of abandoning the idea of targeting full employment and inflation containment at the same time. In their opinion, policy should have had a single target, that they loosely defined as the level of unemployment that was compatible with stable prices or lower inflation. While Harcourt (2000) claimed that it was “an intellectual and political disaster” that “the Phillips trade-off came to be identified with Keynesianism”, our conclusion is that it was indeed very curious that this may have happened.