Did Baz Luhrmann bow to studio pressure and change the sad ending of Australia to a happy one after test audiences gave the movie a huge thumbs down? This is the question that overshadowed the release of the director's new historical epic in Australia and the US last month. After completing his so-called Red Curtain triptych - Strictly Ballroom, Romeo + Juliet and Moulin Rouge - Luhrmann announced his plan to make a trilogy of epics, Australia being the first to see the light of day. However, even before the changes, the process of realising this project was turbulent. Just days before shooting was due to commence, Russell Crowe walked off the project, his role then filled by Hugh Jackman.
Australia is Luhrmann's treatise on his nation's past, dealing with such issues as British colonialism, white Australia's criminal heritage and the country's appalling treatment of its Aboriginal population. But of course, as is his signature, the story is recounted in a camp style that mixes the widescreen vistas of Lawrence of Arabia with the lighthearted adventure of Around the World in 80 Days.
As is the norm with nearly every big studio film - and a growing number of independents - a cut of the movie was shown to test audiences to gauge its appeal. A test audience is a group made up from members of the general public, randomly approached on the street and then asked several questions that will determine whether they are invited to attend an advance screening of a given movie. These enquiries can be about the respondent's socio-economic status or their cultural inclinations. Usually the formula works, but sometimes these questions can be misleading. For example, in 1995, prospective test-audience members were asked if they wanted to see a new movie starring Morgan Freeman and Brad Pitt.
At the time, Freeman was best known for his role in the gentle family drama Driving Miss Daisy and Pitt for Legends of the Fall. Fans of either film would have been unlikely to enjoy David Fincher's dark serial-killer thriller, Se7en. All the same, that is exactly what they ended up watching. The screening was a disaster. Researchers at test screenings find out what the audience thinks of a movie by handing out multiple-choice questionnaires once the credits have rolled. In addition to basic questions such as whether the viewer will recommend the film to their friends, some sections require a more in-depth response, such as what their favourite scene and least favourite scenes were. There is also often a section that invites any general comments about the film.
Test audiences for Australia were shown a version in which Jackman's character, Drover, dies. It has been said that the audience reaction was so negative that Luhrmann was forced to dump this conclusion to his $130 million (Dh480m) adventure. One test-screening attendee apparently described the film as "an action-filled tragedy" and urged Luhrmann to alter the ending. Another was more forthright, saying, in reference to Jackman's turn as Wolverine in the X-Men movies: "There is no reason to kill off Wolvie - come on ?"
Luhrmann has gone on record to state that the test-audience results didn't impact the final cut of the movie in the way that has been reported. He told the Los Angeles Times that he originally wrote six endings, shot three, and the one in the film is the one that felt right. He points out, quite rightly, that it is neither happy nor sad. The coverage given to the story highlights one undeniable fact - that a historical stigma is attached to directors who change the ending of movies as a result of focus-group reactions.
The first major audience research firm employed by Hollywood was Audience Research Inc (ARI), run by the pollster George Gallup. ARI tested rough cuts for films such as Casablanca, which, contrary to its later success, bombed with test audiences. However, the firm was usually successful in its predictions. Most of the time, these results didn't lead to any change with the cut of the film itself, but fed into how the studios marketed the films in question. Of the notable exceptions that did get changed, Orson Welles's The Magnificent Ambersons, remains one of the best known. The studio took the edit away from the director, cut 50 minutes, added a happy ending and, it is believed, destroyed the cut footage.
The most famous example of a film being changed after bad test-screenings is Ridley Scott's sci-fi classic Blade Runner. The original cut was edited without Scott's permission and included two drastic changes: an omnipresent voice-over narration that does all the thinking for the audience and, again, a happier ending - the polar opposite to the bleak vision laid down by Philip K Dick's novel Do Androids Dream of Electric Sheep, on which the film is based.
Another instance is that of Fatal Attraction. Originally, Glenn Close's character committed suicide and framed Michael Douglas for murder. The new ending saw Close get her comeuppance. While Adrian Lyne oversaw these changes himself, other directors are less compliant. For example, when test audiences reacted badly to the ending of Payback, Brian Helgeland walked, leaving the film's star, Mel Gibson, to take the reins and film the new conclusion.
However, some people claim that the test-audience system is useful and necessary, providing a means to fine tune a movie and iron out niggling problems. The producer Harvey Weinstein is a keen advocate of such changes and reshoots. Meanwhile, directors including John Madden (Shakespeare in Love) have talked of their positive effects. Spike Lee, a director who retains the right to approve the final cut of all his films, also runs test screenings for this purpose.
More and more, test screenings are being seen as just another part of the filmmaking process. Indeed a reshoot period is now often written into actors' contracts. That is not to say the stigma has completely lifted. Some powerful directors refuse to have test screenings. Steven Spielberg famously only shows his rough cut to friends and people he trusts. The internet has also contributed to studios becoming increasingly resistant to the idea of test screenings for fear of bad publicity before a film's release.
For now, the jury is out on the merits of test screening. The process is condemned as filmmaking by committee when it goes wrong and praised as a useful tool when it leads to improvements. However, what role it played in the final cut of Australia remains a whole other story.