Addendum to the Bechdel-Wallace test

[jack]

Talking with Liv, I started suggesting addenda to the Bechdel-Wallace test.

* "is there a female (or other non-white-male-majority-culture) who you describe first in terms of what she does, rather than as 'this woman who...' or in terms of her sexual relationship to the main character?"

* "is there a prominent emotional relationship which is non-sexual?"

My questions would be:

1. Any more?
2. Is it sensible to suggest simplified heuristics for complex problems? (Upside: raise awareness. Downside: encourage simplistic evalutations and even worse, writing to pass the test rather than avoid the underlying problem)
3. Suggestions for names, possibly also webcomic-author related? (Should Shaenon Garrity have one? Should #1 be called after Ripley?

Comments

[rmc28]

I think a simple heuristic is useful as a monitoring tool, for trends rather than a simplistic good/bad judgment on a story (film/book/play/etc). Disney usually fails; Ghibli usually passes - guess who I'm spending more of my money on these days. But two of my recent favourite films fail (How To Train Your Dragon, Howl's Moving Castle) - they are still good films with strong characters of both sexes.

If the sexes were equally represented, the rate of Bechdel/Wallace passes would be around 50%, but this is very obviously not the case; the value of the test is in a quick way of collecting data to show the trend, not in judging individual films.

[naath]

I think if the world of cinema were equal then the rate of B/W test failure would be the same as the rate of the reverse-B/W test failure (sub in "men" for "women"). Some films pass both with flying colours; some films pass neither; and for-sure there are stories best told in a single-sex environment, but when was the last time you saw a film set in a nunnery (I have never seen a film with no men in it; although I think Mamma Mia! has men who are there only to be romantic interests for the women). I don't think it would be 50% but it would be a lot nearer 50% than it currently is.

[rmc28]

Yes, I think you are right in that the female-characters-interact test ought to pass at the same rate as the male-characters-interact test.

I think I may have mentioned Charles choosing "Tinkerbell: The Great Fairy Rescue" from the library. That has a completely reversed gender bias, but all the marketing is aimed at the pink-princess demographic, reinforcing the idea that girl-focused films are a special case, whereas male-focused films are the default. [I expected to hate the film, having internalised the message that fairy films are boring, and I found it surprisingly enjoyable.]

[jack]

Both rates in a gender-blind world would presumably be slightly lower than the current male-male rate (since, if some of the same roles are filled by women instead of men, some films that now pass the male-male test but only just wouldn't). But it would hopefully be higher than 50% -- the rate would be 50% if there's one lead, and several secondary characters of both sexes who only ever talk to the lead, which is more or less the case now, but surely some films have people talking to each other not with or about a single main character :)

[rmc28]

I meant that the rates ought to be equal to each other in the gender-blind world, not equal to what they are now, or 50%, for the reasons you give.

[jack]

Oh, sorry. yes.

[naath]

*there will probably always be stories that have only one character who is a a fully-realised character (rather than "the cook" or "the gadget maker"). In a world where character genders were randomly assigned with 50/50 split these stories would pass a lot more often than they do (although of course not always).

*there will also probably always be stories that are, for one reason or another, set in single-gender societies (or a close approximation). At present almost all such stories that get told are man-stories, but there are woman-stories on this model! We should TELL THOSE STORIES. A better world would have those stories be told.

[vyvyan.livejournal.com]

Actually I saw "The Calling" (http://www.imdb.com/title/tt1118660/) at the Cambridge Film Festival in 2009, which is set in a nunnery and has only women as the main characters. (It was really good, incidentally.) But I agree it's very rare!

[naath]

Cool! I new there must be SOME films out there like that... but I haven't seen any. So now I need to watch that one.

[ptc24]

The BWT has a certain amount of punch due to the short definition, and the (seeming?) nonfuzziness of the criteria. The addenda you add... I get the feeling that if you got a bunch of people to assign films using it, and calculated inter-rater reliability (at least using a metric such as Cohen's κ which factors out base-rate effects), then the reliability would be low (and liable to be variable depending on what sorts of people you recruit your annotators from). Alternatively you could write pages of guidelines and produce something which could be used to assign films with high reliability... but the process of guideline development would itself be suspect.

[jack]

Good point. In fact, in many ways I still like the original description in the comic more than more recent descriptions because it emphasises the three steps: that many films fail to have two female characters at all, and many of the rest fail to have them talk to each other at all.

I feel like you may be able to come up with some more similar succint three-point checklists but I admit I don't actually know.