At the beginning of Citizen Kane, the dying Charles Foster Kane
whispers the word "Rosebud", and a reporter scurries about for a few
days and pieces together his entire biography from the two syllables. If
only life were as simple as the movies. In the late 1980s, a few years
before John Cassavetes' death, I had a series of "Rosebud" conversations
with him. The American independent film-maker told me things about his
life and work that he had never previously revealed.
Our
discussions covered a lot of territory, but one of the things I spent
most time asking him about was the fate of alternative versions of his
films. Because Cassavetes made most of his movies outside the studio
system and financed them himself (paid for from the salary he made
acting in other directors' films), he was free from the constraints that
limit Hollywood film-makers. He could take as long as he wanted to
shoot his projects, spend as much time as he needed to edit them and, if
he was so inclined, reshoot or re-edit them as much as he wanted.
The result was that at various points in their creation, most of his
works - including Faces, Husbands, A Woman Under the Influence and
The Killing of a Chinese Bookie - existed in wildly differing versions.
The film we spent the most time talking about was
Shadows.
Cassavetes' first feature, generally regarded as the beginning of the
American independent movement, had had a vexed history. The film-maker
in effect made it twice, filming an initial version in 1957, and
screening it in the autumn of 1958 at New York's Paris Theatre for
invited audiences. But, dissatisfied with the response, Cassavetes
reshot much of the movie in early 1959, replacing approximately half of
the footage in the original print with new material. In late 1959, the
so-called "second version" of Shadows premiered.
What made the
Shadows story particularly interesting was that a number of critics and
viewers who saw both versions were convinced that Cassavetes had made a
grievous mistake. Jonas Mekas's Movie Journal column, published in
the Village Voice on January 27 1960, can stand for all: "I have no
further doubt that whereas the second version of Shadows is just
another Hollywood film - however inspired, at moments - the first
version is the most frontier-breaking American feature film in at least a
decade. Rightly understood and properly presented, it could influence
and change the tone, subject matter, and style of the entire independent
American cinema."
At the end of his piece, Mekas expressed the
hope that Cassavetes would come to his senses, suppress the second
version and release the first version; but it was not to be. Cassavetes
withdrew the earlier print and refused to allow it to be screened again.
From that point on, the only version of Shadows anyone would ever see
was the re-edited version.
When I asked Cassavetes the
whereabouts of the earlier print, he said he doubted it still existed.
The likelihood of its survival was all the more remote when one took
into account the modesty of his film-making operations in the late
1950s. The film-maker told me that the first version of Shadows had
existed only as a single 16mm print. He had not had enough money to make
a duplicate or a backup, and the negative that that print was made from
had itself been cut up to make the second version of the film.
The
one small lead he offered me was that he said he vaguely remembered
donating the early print to a film school. Mekas subsequently told me of
a conversation he had with Cassavetes in which the film-maker was
slightly more specific and said that he had donated the print of the
first version to "a school in the Midwest".
Unfortunately for my
peace of mind, the damage had been done. I contacted every school in
the Midwest, starting with the alma mater of Cassavetes' wife, Gena
Rowlands, the University of Wisconsin. I also tracked down anyone I
could locate who had been associated with the schools' film programmes
when the gift would have been made. I had many wonderful conversations,
but came up emptyhanded.
Around the point Cassavetes died, in
1989, I expanded the search. I contacted staff members at every major
American film archive, museum and university film programme to see if
the print had somehow been squirrelled away in one of their
collections. After all, the title would have been the same for the first
and second versions. I began making announcements at film events I
organised or presided over. I asked friends to spread word of the quest.
There was no shortage of leads to pursue over the course of the
next decade. There were thrilling days when it seemed that the print
was within my grasp if only I could get in touch with a particular
person who knew someone who knew someone who knew someone. But that
final someone always eluded me. There were wild goose chases where I
flew into a strange city and met with a collector who I had been led to
believe had the print in his possession. Needless to say, each time the
film turned out to be the second version. There were also comical tricks
of fate. For example, while searching for the Shadows print in the
Library of Congress collection, I stumbled across an uncatalogued,
unrecognised, long print of Faces. Very interesting, very valuable;
but, sorry, wrong movie.
I can't say I didn't get discouraged.
Some time in the mid-1990s, I put the search on the back burner and
decided to take another tack. If I could not actually find the physical
print of the first version, I would imaginatively reconstruct it by
drawing on memories of the cast, crew and people who had seen it, as
well as by studying the second version, which included approximately 30
minutes of footage that had been in the earlier print. I reinterviewed
the cast and crew to pick their brains for memories about the first
version, then studied the composite second version shot by shot for
tell-tale clues about which footage had been filmed in 1957 and which in
1959.
Almost all of that research had to be done at actual
movie theatre screenings, since a video image doesn't reveal the kind of
detail I required. I pulled friends, dragging their feet and
complaining, into 35mm screenings in theatres, handed them clipboards,
and we sat together in the front row, whispering in each other's ears
and taking notes about how an actor's socks changed in two successive
shots. We noted the length of the shadows on the ground to tell what
time of day scenes were filmed; or the size of the leaves and the
openness of buds on bushes in the background of a park scene to decide
what month the scene was filmed. Everything eventually connected with
everything else. The title of a film on a marquee would allow us to date
an actress's hairdo, which would then allow us to date a scarf that an
actor wore in another scene and, three or four steps later, we could
conclusively induce that another scene, different from any of the
above ones, was filmed in March 1957. It took years, but I reconstructed
the entire first version out of such spider-web tangles of
relationships. As the jigsaw puzzle got filled in, tiny piece by tiny
piece, the big picture of what had been filmed in the two different
periods of shooting gradually emerged. I eventually published two books
with my conclusions.
Then, one day two years ago, one of the
friends I had told about the search called, saying he had run into a
woman who might have some information. When I finally tracked her down
and got in touch with her, it was your typical "good news/bad news"
situation.
The good news was that she confirmed that, yes, the
title sounded familiar. Her father had run a second-hand shop in
downtown Manhattan. One of the ways he replenished his stock was by
attending "lost and found" sales held by the New York City subway
system. There were so many forgotten umbrellas, mittens, eyeglasses,
hats, pens and other things left on the trains that the Transit
Authority annually auctioned them off. Though a nice watch might go for
$10-20, everything else generally went for a dollar or two per "lot" - a
box that might contain 50 or 100 umbrellas, mittens, or hats. One year a
long time ago (it was impossible to pin down the date), there was a
fibre-board film container in one of the boxes her father bought. When
he got home and opened up the carton, he saw the title Shadows
scratched on the outer leader of one of the reels, but since he had
never heard of the movie, she told me he simply put it aside and joked
that he was disappointed it was not "a porno film".
In this
case, the good news was also the bad news. The subway was the wrong
place to find the first version. Not only didn't it square with
Cassavetes' account of what he had done with it, it just didn't seem a
plausible scenario. If the only print in the universe had been left on a
subway car, why hadn't whoever lost it simply claimed it the next day? A
print found on the subway was much more likely to be one of the dozens
of prints of the second version being couriered to or from a college or
arthouse screening in the late 1960s or early 1970s.
Even worse
news was that all of this had taken place something like 30 or 40 years
earlier. In my very first conversation with her, the woman emphasised
that even assuming her memory of the title was correct, there was
virtually no chance the actual print still existed. The junk shop had
gone out of business long ago. The father had died years before. The
children had married, had their own families and moved away to other
cities. The store's contents had been sold off or thrown out years ago.
The woman put so little stock in the print's survival that she didn't
even really want to search for it when I asked her to. She emphasised
that she had no idea where to look.
It would take almost two
years of polite pestering on my part before she came up with anything;
but I have to admit that even as I went through the motions of talking
to her every few weeks to remind her to ask other family members if they
had any idea if the print might have survived or where it might be, I
privately wrote off this lead as one more dead end in a dead-end story.
That's why when the film was found in the attic of one of the
children's houses in Florida, shipped to me, and finally in my hands, I
didn't even bother to look very carefully at it for a while. I was
absolutely convinced that the odds were completely against it being the
right print of the film. Indifference suddenly turned to excitement, and
then terror a few hours later, when I manually unspooled four or five
feet of footage from the first reel and held it up to my desk light. All
I could make out was a figure walking down a street, but that was
enough. The known version of Shadows began with a crowd scene.
In
10 seconds, I went from being blasé to being afraid to touch the print.
I have a screening room in my house, but (although it took considerable
self-restraint) I didn't dare thread the film and project it. If this
actually was the long-lost first version of Shadows, the fruit of a
17-year search, it would be just my luck to have my projector scratch or
shred it. I carefully sealed up the carton and made an appointment the
next day at a professional film transfer house to have a high quality
video copy made, so that the original would never have to be run through
a projector again. There was an inevitable delay, of course. It would
be about a week until I could get an appointment - a week of suddenly
anxious sleep - fearful, with the completely irrational fear all
collectors know, that my house would burn down in the interim, before I
was able to get it to a lab and watch a movie that hadn't been seen in
45 years.
The print exceeded my expectations in every respect.
In terms of content, there are more than 30 minutes of entirely
different scenes that are not in the later version. The discovery gives
us a large chunk of new work by Cassavetes - a little like discovering
four or five lost Picasso paintings. Physically, although the celluloid
base was shrunken and brittle from 50 years of storage, the emulsion was
in unbelievably fine condition. Remember that, unlike movies we see in a
movie theatre, or the prints of the second version of Shadows that I
myself had seen, this print was not a duplicate or a blow-up, and it had
only been passed through a projector four or five times before
Cassavetes withdrew it. In film terms, it was pristine, as good as a
film can be. Not only was it custom-printed directly from the original
negative that had passed through Cassavetes' camera in 1957, it was
virtually brand new, unworn and unscratched.
One could ask if
the discovery proves Mekas right or wrong; but that doesn't really
matter. Each version of Shadows stands on its own as an independent
work of art. The value of the first version is that it gives us an
opportunity to go behind the scenes into the workshop of the artist. Art
historians x-ray Rembrandt's work to glimpse his original intentions.
Critics study the differences between the quarto and folio versions of
Shakespeare's plays. There is almost never an equivalent to these things
in film. That is the value of the first and second versions of
Shadows. They allow us to eavesdrop on Cassavetes' creative process, as
it were to stand behind him as he films and edits his first feature. We
watch him change his understanding of his film and his characters. His
revisions - the scenes he adds, deletes, loops new dialogue into, adds
music to, or moves to new positions as he re-films and re-edits Shadows
- offer an almost unprecedented glimpse into the inner workings of the
heart and mind of one of the most important artists of the past 50
years.
The odds against finding the print still astonish me -
not only because it was the only copy in the world and was found in such
an unexpected location, but because of the luck of the timing. The
junk-dealer's children are themselves now old people and the dusty
carton would almost certainly have been thrown out by the next
generation when they died. My friends used to joke that I was looking
for a needle in a haystack, but what I realised after I found the print
was that the situation was even more dire than that. The haystack was
not going to be there very much longer. If the print had been in an
archive or museum, it would have patiently sat there for the next 1,000
years waiting for someone to discover it; but as an unknown object in a
brown container gathering dust in the corner of an attic, it would not
have survived the clean-out of the next generation. Though I had had no
idea that the clock was ticking while I was engaged in my search, I
realised some time after I found the print that this was probably the
last chance to find it for all eternity.
· Ray
Carney is Professor of Film and American Studies and Director of Film
Studies at Boston University. He is the author of books about John
Cassavetes, Mike Leigh, Frank Capra, Carl Dreyer, and many other
film-makers, and manages a website devoted to Cassavetes:
www.cassavetes.com
Ray Carney @
'The Guardian'