[Marian Griffith]
> One of the possible explanations given for that discrepancy (and
> it seems to be the one that the media jumped on) was that the
> differ- ence was a reflection of the same mechanism that keeps
> women from earning an equal wage for equal labour in the 'real'
> economy.
Assume this is true. It would, at minimum, have to be the same
mechanism acting on a completely different population - that is, the
people who are selling female avatars, and not women. And it also
has to act on an altogether different dependent variable. It is not
compelling, but it is conceivable.
> [example of somehow comparing forklifts and oranges snipped
> because I do not quite understand what it is supposed to mean.]
There are some similarities between a forklift and an orange (e.g.,
both could roll along.) There are many dissimilarities. In order to
use observations about one to draw conclusions about the other, we
would have to make some kind of argument that the dissimilarities
didn't matter for our purposes.
Castronova's situation differs from what it is supposed to model not
only in what it applies to (characters vs. women), but in which
dependent variable it applies to (price for character controlling
for level vs. wage paid to women for work controlling for education,
hours, children etc.)
Hence, if we are going to draw conclusions about the wage gap, we
will at least implicitly need to draw out the equivalence between
the two situations.
This equivalence is far from obvious.
> The important point to make here is that so far you are talking a-
> bout hypothetical items, while the research being discussed here
> is talking about factual items being sold that, from an technical
> point of view, should be equal, but are valued quite differently.
They only 'should' be equal by the value judgement that the price
for a character should not at all reflect anything but level and
base attribute bonuses, excluding appearance, labeling of the
character as female, and character persona.
Even as a value judgement, it's arbitrary - not far from saying that
one ought not to prefer strawberry ice cream over otherwise
equivalent chocolate ice cream.
Perhaps this seems like an unfair, wrong comparison. After all,
preferring 'male' to 'female' in some otherwise arbitrary choice is
important exactly because the options are 'male' and 'female' and
not 'chocolate' and 'strawberry.' Right? But exactly what I am
pointing out is that the options aren't 'male' and 'female' across
the two situations. Only if you already assume that the two
situations are equivalent (for whatever mysterious reason) will it
seem that the Ebay preference is discriminatory and
non-arbitrary. Without the equivalence, the Ebay preference is
exactly as trivial as chocolate vs. strawberry. The equivalence
hasn't been provided, so can't even be evaluated. It isn't good
enough to have the words 'male' and 'female' in there. If the choice
is truly arbitrary then it shouldn't matter what words the options
are labeled as, a person can choose whatever they like or have any
degree of preference without it being sexist.
But I don't actually think that the value judgement is here because
you think it is important that people should have no preference
about which gender character they play _per se_. That would be silly
and arbitrary, and anyway beside the point - the preference is
supposed to be diagnostic of a mechanism which is also behind the
wage gap. Changing only this preference would be shooting the
messenger.
Rather, I think the value judgement is retained because it's a
crucial link in the argument. "The prices should be equal (or it's
discrimination). But they are different, so it's discrimination (or
anyway, bad). Discrimination (at least with great plausibility) has
the same mechanisms wherever it is, all the more so because this
seems characterizable as gender discrimination in both cases (wage
gap & Ebay). Since there is no recourse to arguments relating to
childbirth etc., we have a model situation that allows us to study
the causes of discrimination without interference from those
arguments."
> Either there is a real difference in *actual* value, in which case
> the prices are fair, or there is not, in which case a discrimina-
> tory mechanism is in place. (discriminatory meaning that distinct-
> ion is being made on grounds that have no relevance to the perfor-
> mance or abilities).
Abilities meaning level and stats. But the 'performance' of a
halloween costume is not limited to some quantification of how good
the fabric or even craftsmanship are; not even how much other people
think it is a well done costume; but how well it fits you, and very
importantly how well you like to wear it, whether you think it's a
good costume.
>> That leaves 'female' characters in the same place, albeit sharing
>> a word with a real life group. But, once again, a toy duck is not
>> a duck just because it shares a word and both have some kind of
>> tail.
> Yes, but I think it is fair to say that the difference betweem
> male and female avatar is quite a bit less than between a toy and
> a liv- ing animal.
But the difference between a woman and a female avatar, which I was
pointing up, IS great. In fact, it is so great that they share
almost nothing even in how they are reacted to.
> From within the context of the game there is *no* difference. So
> the only difference between the two is in the eyes of the
> beholder, or player in this case.
Though you say emphatically that there is no difference in the first
sentence of the above, I know you mean 'no difference good to care
about.' I understand that you have declared any differences due to
appearance and labeling, or anything that falls out from these, as
(self-evidently?) discriminatory. But this is very arbitrary.
> and the fact that games offer one of the few si- tuations where
> the mechanism can be studied in an environment that is free of
> actual differences in ability or job history (in the sense that
> female avatars don't interupt their game careeer to get pregnant
> and care for children), is this quite significant from a viewpoint
> of sociological and socio-economical studies.
It is sad that the same reason it looks like such a potentially
valuable model (avatars aren't women, hence share nothing with women
that would invalidate an argument on the basis of factors which are
not present with avatars) is the same reason it is a bad model
(avatars aren't women, hence we have no remotely good reason to
suppose they share anything significant with women that would make
the comparison productive).