For actresses, James Bond films should a carry health warning: ‘Starring in a Bond film may cause your career to go down as well as up.’ There is a persistent view, most recently expressed in a paper by Claire Hines,* that being a ‘Bond girl’ is damaging to careers. Despite the usually high-profile and successful appearance in a Bond film, the actress rarely seems to achieve anything comparable subsequently, and is becomes confined to the television treadmill or pops up from time to time in low-budget sub-Bond thrillers. But is this really the case, or is the Bond girl curse an idea – a meme – that is replicated with little connection to the evidence?
We can test the validity of this successful meme, transmitted in books and cyberspace, with the help of the Internet Movie Database (IMDb). First, we must establish the relative merits of the Bond girls’ post-Bond films – and their pre-Bond films to better trace career trajectories – and compare the values against Bond films. Box-office receipts is a useful measure, but the data are unavailable for older and obscure films. The alternative is the IMDb rating, which gives for each film an average of website users’ scores out of ten. The average score for Dr No, for example, is 7.3.
For convenience, I have limited my dataset to up to five films before or after Bond, giving me a potential series of 11 films for each actress. Of course, each film usually has a number of Bond girls – Roald Dahl famously stuck to a three-girl formula – but for my purposes, the Bond girl in all cases is the main ally of Bond, the girl who is in Bond’s arms at the end of the film (yes, I know Olga Kurylenko was not in Bond’s arms at the end of Quantum of Solace, but she did survive the film, which comes to the same thing).
Let us examine the main trends. In chart 1, we see the mean score for each film (-5 being the fifth film before Bond, 0 the Bond film, and 5 the fifth post-Bond film). On average, Bond films have the highest scores, enjoying a mean of 6.84. Films before and after Bond films rarely achieve comparable scores and consequently their means are lower, largely scoring between 5 and 6 out of 10. Is there a curse of the Bond girl? The answer must be no, because the level of film success that Bond girls enjoyed after their Bond roles was generally no different from the level of success they had before Bond.
Chart 2 shows the difference between the overall mean for each actress and her Bond film rating in Bond film order. For many actresses, the difference is relatively large, highlighting the fact that the sort of success the actresses enjoyed in Bond films was rather exceptional. Lois Chiles, Michelle Yeoh, and Halle Berry, however, featured in non-Bond films which were on average as successful as Bond or in fact more successful. (To be fair on all actresses, some of their pre-Bond films were so obscure or otherwise little seen and were not rated on IMDb. This unfortunately lowered their averages. We should note too that celebrated stage careers, for instance that of Diana Rigg, are not reflected in the scoring.)
What can we conclude from this analysis? Looking again at chart 2, we see that in chronological terms the gap between non-Bond and Bond success becomes smaller with time, suggesting that the Bond producers were increasingly prone as the series progressed to select actresses who had been in films as well regarded as Bond films. Possibly the factor of beauty became less important through time, while acting ability became more important. We can now see the ‘Bond girl curse’ as a successful, but false, meme, which has been propagated well to become deeply embedded within popular culture. In reality, the pattern fits a phenomenon known as ‘regression to the mean’. In his book, Bad Science, Ben Goldacre, explains the phenomenon with the example of a cold. Colds tend to have a cyclical pattern. People feel fine most of the time, get colds, feel terrible, then improve. Some people do seek remedies, though, usually when they’re feeling at their worst, and credit their improvement to the remedy, rather than the fact that the cold would in any case disappear naturally. This is regression to the mean. The mean in terms of Bond girls is mediocre or poor film success. Bond films represent a blip – the once in a while cold – that punctuates that trajectory.
*Hines, C, 010 ‘For his eyes only? Men’s magazines and the curse of the Bond girl’, in James Bond in world and popular culture: The films are not enough (eds R G Weiner, B L Whitfield and J Becker), Cambridge Scholars Publishing, Newcastle, 167-175