Politicians intent on demonstrating the decline of the quality of
American life for the purposes of re-election sometimes blame it on
the movies and television, all as if their own efforts in office
had nothing to do with it.
        The industry's leaders respond that their on-screen efforts
don't create American society, they merely reflect it.
        There's plenty of blame available for both sides to have a
generous helping, and the truth probably is more complicated than
either would prefer. Life and art influence each other.
        But the airy debate on exactly how much would be greatly
illuminated by some data on exactly what's showing on all those big
and small screens, and that's the project that absorbs Stanley
Rothman and his colleagues.
        They are studying leadership and social change in the
United States. Earlier volumes have covered the press ("Media
Elite") and television ("Prime Time"). The newest, written with
Stephen Powers and David J. Rothman, is "Hollywood's America:
Social and Political Themes in Motion Pictures."
        "I do data," Rothman says, and that's what distinguishes
this work from most film commentary, whether reviews of individual
movies or academic speculations from professors in departments of
film studies.
        "Hollywood's America" relies on two entirely different but
complementary approaches to movies. One is to look at the films
themselves, through a detailed content analysis of characters,
their motivations and the situations they encounter in the script.
        The authors took as their sample population the 10
top-grossing films of each year from 1946 through 1985, and
randomly picked a subset of that sample to study in detail.
        The second approach looks at the people who make the films,
with a survey and interviews of the writers, producers and
directors of the 50 most popular films made between 1965 and 1985.
        They're not even looking for bias, let alone conspiracy.
"The Hollywood elite," they write, "shares a set of political and
cultural assumptions that it views as natural (as all of us view
our own assumptions) and that it seeks, as do others, to put into
action."
        The totality of their work "suggests a more or less
coherent ideology ... generally left-leaning and highly critical of
traditional features of American society."
        On one obvious measure, whether they consider themselves
liberal or conservative, moviemakers are to the left of other elite
groups as well as the general public. Only 9 percent of military
leaders call themselves liberal, compared with about 25 percent of
the public and 66 percent of the moviemakers.
        The way these people see the world inevitably affects the
kinds of movies they make, but the relationship is complicated and
sometimes not clearly understood or articulated even by themselves.
Rothman and his co-authors look at the measurable effects rather
than theorizing about the relationship.
        They hired college students to watch the films in the
sample and to code each character with a speaking role according to
whether the role is neutral or not, and if not, whether the
portrayal of the character is positive, negative or mixed. For
non-neutral characters, more than a thousand of them, the coders
also record goals and methods, and additional information such as
whether the character is victimized in any way.
        In some ways, that's very simplistic and it certainly
doesn't capture the full subtlety of either human experience or its
on-screen simulacrum. But it's relatively easy to do, with a high
degree of agreement among the coders, who work independently and
aren't familiar with the purposes of the study, or any prejudices
the researchers may have about what they expect to find.
        And with such a large number of characters studied, over a
period spanning decades, it's very revealing of trends that may be
quite invisible to the very people who create them.
        Religion plays a much smaller role in the movies than it
does in real life, but small as it has always been, it has been
declining precipitously. In the first decade of the study, 19
percent of the characters were identifiable as Protestant,
Catholic, Jewish or otherwise religious. In the most recent period,
it was down to 4 percent.
        At the same time, the proportion of these characters who
were treated positively declined from 81 percent to 56 percent.
That may be more realistic, as few people qualify for angelic
status, but it also influences how people perceive reality.
        Military figures have seen their image drawn more darkly as
the decades have passed. The percentage of  military characters
employing violence, deceit or trickery increased, while positive
portrayals declined.
        Film critics often write that the movies of the '80s
reflected a sharp turn toward conservative and patriotic themes,
but it's hard to find that in Rothman's statistics. Even such
jingoistic effusions as "Rambo" draw the picture of the hero as
renegade, forced into independent action because the legitimate
channels to accomplish reasonable goals are ineffective or corrupt.
        Law enforcement fares badly as well. It's not surprising
that there's more crime in the movies than in real life, but the
change comes in who is responsible. The percentage of officers
committing crimes climbs from about 20 percent at the beginning of
the period to nearly half by the end. 
        Everyone knows that there are some bad officers in any
police force, but when the movies exaggerate how common they are
may mislead audience members, especially those whose real-life
interactions with the police are tense and hostile.
        Moviemakers view American life as more dismal than most of
their audience does. The authors of "Hollywood's America" have
shown just how much more dismal, and that's a good thing to know.