Skip to content

To refer to a producer’s oeuvre is, at least to me, as ignorant as to refer to the oeuvre of a stockbroker.

—David Mamet

There are a lot of complaints these days about the declining quality of movie fare, and the worsening taste of the public is typically asked to shoulder a good part of the blame.

Other causes are cited as well. When the old studio system collapsed, the high-profile chiefs who once put their distinctive stamp on pictures were replaced by accountants and corporate executives with little flair, imagination, or passion. The exponential growth of video has made home viewing more popular than theatrical movie going, and the notions of film as community event, as theatrical experience, or as “something special” have all suffered terrible losses. More simply and immediately, there’s the preference for loud explosions and frenetic comic-book action over drama and character, escalating violence over tenderness, torrents of profanity over well-crafted dialogue.

But most of the blame falls on the overall coarsening of the audience. According to conventional wisdom, most movies are targeted to the teen and preteen market; the decreased literacy of the filmgoing public rules out most subtitled movies; and there is an overall dumbing down of American movies, with a notable increase in anti-intellectualism. Many commentators are quick to add that as the public grows increasingly apathetic, apolitical, jaded, and cynical, movies designed for their delectation would naturally follow suit.

The bottom line is always the same: the audience is ultimately to blame for what it winds up seeing.

Let’s concede that there’s some measure of truth in all these assertions—as there is in most assertions, if one bothers to look for it. But focusing on the last statement for a moment, might not the industry commentators have their cause and effect reversed? Couldn’t the movies, rather than their spectators, be spearheading as well as defining this decline? Don’t they share at least part of the responsibility for this overall dumbing-down? Given the uncritical promotion of the major studio releases, one might even posit that the press, in order to justify its own priorities, maintains a vested interest in viewing the audience as brain-dead. After all, if it showered most of its free publicity on more thoughtful and interesting movies it would run the risk of being branded elitist. How much easier it becomes to wallow in the slime if you and your editor or producer are persuaded that it’s the audience’s natural habitat—that the audience, not the press working in collaboration with the studios’ massive publicity departments, calls all the shots.

Furthermore, or so the producers tell us, the highly sophisticated forms of market research and testing that shape major releases at every stage in their development, from initial treatments to previews to final ads, scientifically prove that the studios are correct in their low estimation of the public taste. This is clearly the surest indication of what the audience actually wants, so how can the producers be faulted for catering to their preferences?

I believe that this line of reasoning is even more stupid, self-serving, and self-deluded than the movies it seeks to excuse, deriving from a set of interlocking rationalizations accepted by everyone from studio heads to reviewers. What’s more, it’s a logic that doesn’t even work: The majority of movies made according to these “scientific” principles bomb. In fact, it wouldn’t even be worth considering if it wasn’t one of the central explanations for why Hollywood makes so many bad movies.

For a bracing rejoinder to this set of assumptions, try the following on for size:

“It seems legitimate to wonder how men of the perspicacity which has distinguished some of the best work in market research could have failed to realize from the beginning that most of their research practices were not adaptable to the cultural field. The very attempt to adapt them to that field was virtually certain to cause damage.

“Obviously, the premise of a stable audience with reasonably permanent and objectively verifiable needs simply does not hold in the cultural field. Transplanted from economics, this premise becomes an obvious interference with the free play of human intelligence. To say that men would always need warm clothing in cold weather obviously was a statement of fact; but to say that men would always need soap operas in America was just as obviously plain insult. Yet the pollsters, straightfaced and singleminded, proceeded to ram their hypothesis down the public’s throat; the public, unaware of what was happening, had barely time to gag.

“What was this hypothesis and how did it affect the public’s mental health? Adapted from mercantile economics to a field where mercantilism does not apply, the theory assumed a shallow, slothful, and unchangeable crowd, forever doomed to frustration and th us forever dependent on the wish fulfillment of certain minimum needs—sex, glamour, adventure, wealth, power, and the rest of them. The whole range of subtleties which make up the pattern of civilized behavior was not only rejected as being beyond the grasp of the audience—it was dismissed as irrelevant to their real desires.”

Ernest Borneman’s commonsense dismissal of the basic assumptions of the American film industry in “The Public Opinion Myth” appeared in Harper’s Magazine in July 1947. But give or take a couple of sexist coinages that were standard for that period, I don’t see how it could be improved much today. Whether it can be absorbed and heeded is of course another matter.

Why it’s unlikely to be heeded is a matter of simple economics: Even by 1947, the industry founded on the public opinion myth was already too vast and too solidly in place to contemplate the prospect of dismantling it.

Now, I’m not really in a position to declare whether the audience is right or wrong about anything. Properly speaking, the audience is so many things, all of them overlapping and most of them scarcely known, that assigning it a label in advance effectively means ruling it out of discussion—which market research usually does.

I suspect that any studio publicist or production executive who read Borneman’s article today would argue that movie market research was still in its infancy in 1947, that the nature of both the audience and the film industry has radically changed since then, and that even if Borneman’s argument once had some validity it no longer applies to the contemporary realities of developing, making, testing, revising, publicizing, distributing, and exhibiting movies. Not only has the market research industry become more sophisticated, but the audience has undergone profound changes, temperamentally as well as demographically: It’s more demanding about some things (such as the quality of special effects) and less demanding about others (such as plots that make sense); it’s much younger; moreover, the video revolution has transformed everything having to do with movies. Today’s markets are defined differently because people attend movies as special events rather than as an everyday activity, the older segments of the audience tend to stay at home, and so on.

All these things are certainly true, but I can’t see how they alter the basic thrust of Borneman’s charge. Then and now, the operations of the media-industrial complex have been predicated on certain highly questionable assumptions about the audience, and charting box office grosses to “prove” those assumptions is merely indulging in a self-serving form of circular reasoning. The bottom line is always the same: the audience is ultimately to blame for what it winds up seeing. We are told that this is the downside of democracy—we can’t always expect to like what the mass public endorses—a sentiment that can only lead me to cite Borneman again:

“Does the whole process of audience testing … really qualify as a democratic process? Does it not resemble an election in which only one candidate has ever been introduced to the electorate? Have we ever been given a freely available standard of comparison between the pollsters’ “control card” and its best alternative? If the difference between any two alternatives is so negligible as to defeat judgment, have we, the public, truly returned a valid opinion? And, finally, have the pollsters ever provided us with the aesthetic training which would have enabled us to make a reasonable decision?”

Let’s translate these skeptical questions into a few practical applications pertaining to the Nineties. The December 17, 1993 Wall Street Journal carried a story headlined, “Film Flam Movie-Research Kingpin is Accused by Former Employees of Selling Manipulated Data.” The story reported that about two dozen former employees of National Research Group Inc., which handled most Hollywood test marketing, stated that their data were sometimes doctored to conform to what their paying clients asked for. These former employees ranged “from hourly workers to senior officials” and mostly included people who had left the company voluntarily. All the examples given in the story (e.g., L.A. Story, The Godfather Part III, Teen Wolf) involved boosting a movie’s score, but one could easily surmise that the reverse could have happened on occasion when the studio for one reason or other wanted a movie to fail—which actually happens more often than most moviegoers realize. Before dumping Peter Bogdanovich’s The Thing Called Love, for instance—a film about country-western music hopefuls in Nashville, all of them played by nonmusicians such as River Phoenix, Samantha Mathis, and Sandra Bullock—Paramount test marketed it by showing it to country-western music fans, a move that seems about as logical as previewing One Flew Over the Cuckoo’s Nest in a mental institution. This may not constitute “doctoring” test results in the usual sense, but it certainly sounds a lot like predetermining the outcome.

As a result of this Wall Street Journal story, National Research Company, Inc. lost most or all of its Hollywood clients, and I’ve been told that its successors have proceeded more cautiously. One might think that such a revelation would cast doubts among reviewers and other industry commentators on an already highly dubious practice, but no such reflection or soul-searching ever took place. The industry needs its self-fulfilling prophecies too badly to tolerate skeptics and most film reviewers are hardly independent of either studio interests or their alibis.

Writer-director James L. Brooks, who received most of his training in TV sitcoms, is a talented filmmaker who believes so religiously in test marketing that he seems fully willing to compromise his own work to the point of unintelligibility in order to conform to its supposedly exacting standards. This curious devotion to market research was demonstrated most cripplingly in the fate of his 1993 musical about contemporary Hollywood filmmaking, appropriately titled I’ll Do Anything. The movie was eventually released in 1994 as a nonmusical after a series of test screenings gradually persuaded Brooks to remove all of the film’s musical numbers. As a reductio ad absurdum of the perils of test marketing, I’ll Do Anything should be seen by everyone in its musical and nonmusical versions, but of course it won’t be, because seeing how much better it was before all the meddling took place exposes the patent absurdity of the process.

So I’m afraid you’ll have to take my word for it: Having had the opportunity to see I’ll Do Anything as a musical, I can report that it was immeasurably better in that form—eccentric and adventurous, to be sure, but also dramatically and emotionally coherent. The fact that the movie bombed at the box office as a nonmusical doesn’t of course mean that it would have scored commercially in its original form, but considering that it had an artistic logic and integrity, surely it had a chance of finding an audience if the studio had known how to market it. (Ironically, part of the movie’s plot is directly concerned with test marketing.) By the same token, it seems impossible to imagine how the release version could have found an audience under any circumstances because its emotional and dramatic raison d’être had been removed; following the biblical injunction, “And if thy right eye offend thee, pluck it out,” Brooks wound up eliminating so much of his original conception that what remained was meaningless.

If test marketing was used exclusively to determine how certain pictures could best be marketed and advertised, I would be inclined to consider it defensible on those grounds; clearly studios have to determine what segments of the audience a movie is most likely to appeal to, and to represent that movie in advertising according to their discoveries. In some cases, I might even defend the use of preview screenings to determine whether certain pictures could turn a profit and therefore whether they should be released. And I would agree that some directors, especially directors of comedy, can benefit from previewing their rough cuts to see when and how they get laughs before making their final cuts.

But using test marketing to impose last-minute changes on movies seems much harder to justify, especially because it assumes that an audience is qualified to make decisions of this kind. Of course we wouldn’t dream of policing the writing of novels or the composing of symphonies in this fashion. And if knowledge and expertise are necessary to arrive at intelligent decisions, it’s hard to defend the idea of spectators without these qualifications determining the shape and effect of theatrical features.

Test marketing assumes that an audience confronted with something new will arrive at a permanent verdict immediately after seeing it. But our experience of movies—apart from the most routine fare—seldom works that way. Our expectations play a considerable role in determining our first reactions, and once we get past them all sorts of delayed responses become possible; a day or a week or a month later, what initially made us querulous might win us over completely. The film industry factors out responses of this kind—responses that suggest we’re capable of learning and growing—thus denying our capacity to change before we can catch our breaths. Determining that a film’s success or failure has to register instantaneously, the studio then becomes locked into a treadmill of other assumptions that degrade the audience even further.