ROCKRULZ........STAY RAW...........

ITS A SMALL PRODUCTION HOUSE. WE WORK AS AN INDEPENDENT FILM MAKERS."WE DONT THINK FILM MAKING IS A COSTLY AFFAIR FOR US FILM IS NOT AN INDUSTRY RATHER AN ACTIVISIM IT IS NOT ALL ABOUT GLOSS".
“Quit your complaining. It’s not the world’s fault that you wanted to be an artist. It’s not the world’s job to enjoy the films you make, and it’s certainly not the world’s obligation to pay for your dreams. Nobody wants to hear it. Steal a camera if you have to, but stop whining and get back to work". IT SEEMS THAT JUST ABOUT EVERY ONE HAS AN INTEREST IN FILM MOVIES. BUT EVERY ONE KNOWS, MOVIE-MAKING IS COLOSSALLY EXPENSIVE, AND TOLLYWOOD IS SOMETHING OF A CLOSED COMMUNITY. SO MORE AND MORE FILM MAKERS ARE GOING INDEPENDENT AND PRODUCING MOVIES ON THEIR OWN.

Sunday 18 March 2018

Tuesday 6 September 2011

TRENCHANT

 Next venture of rockrulz and a dream project by souradeepta and ria chowdhury......the movie is an attempt to bring forth the gradual urbanisation dat is occuring....in d villages...and how the element of 'true india' and its rustic flavour is losing out from our lives.....having d steel city as its backdrop, the movie encircles around the life of people who have moved on....and are developing...the stark difference between the posche city and urban life is exhibited....and just lyk every other rockrulz project, we attempt to suprise the audience positively....the movie directed by souradeepta chowdhury and enscripted by ria basu has its promo released today......we hope to carve a niche in the hearts of all yet another time...........so...await its arrival....

Monday 30 May 2011

                                           
                                                                         DVD COVER

STAIRWAY TO HEAVEN FILM POSTERS AND DVD'S

                                                             
                                                        STAIRWAY TO HEAVEN DVD 

Saturday 28 May 2011

Struggle on Two Fronts:A Conversation with Jean-Luc Godard



Jean-Luc Godard; Jacques Bontemps; Jean-Louis Comolli; Michel Delahaye; Jean Narboni;
Cahiers du Cinema; D. C. D.
Film Quarterly, Vol. 21, No. 2. (Winter, 1968 - Winter, 1969), pp. 20-35.
Struggle on Two Fronts:
<A Conversation with Jean-Luc Godard>
"What we demand is the unity of politics and art, the
unity of content and form, the unity of revolutionary
?>oliticalcontent and the highest possible perfection
of artistic form. Works of art which lack artistic
quality have no force, however progressive they are
politically. Therefore, we oppose both works of art
with a wrong political viewpoint and the tendency
toward the "posters and slogan style" which is correct
in political viewpoint but lacking in artistic
power. On questions of literature and art we must
carry on a struggle on two fronts."-"Talks at the
Yenan Forum on Literature and Art" (May 1942),
Quotations from Chairman blao Tse-tung, Peking,
Foreign Language Press, 1966, p. 302.
Because of the kind and the degree of its commitinent,
people are wondering whether La Chinoise
doesn't risk losing adherents to all the political
"lines," and whether it doesn't, then, in the final
analysis, just bring it all back down to fi lm.
I f that were the case, it would have missed its
mark and be reactionary. What you say reminds me
of what Phillipe Sollers told me about it. Though
he, unlike the people you speak of, bases his view
of it on the idea that it doesn't as a matter of fact
"bring it all down to film." To give support to his
view, he points to the conversation between Anne
Wiazemsky and Francis Jeanson on the train. According
to Sollers, the scene is reactionary. It's reactionary
because it pits the "real" talk of a real person
-the talk has to be "real," he says, because the character's
name, like the real man's, is "Jeansonnagainst
the "fictional" speech of a pseudorevolutionary,
and because the scene seems to justify the
former.
Do you think it does?
I think it justifies Anne Wiazemsky's position. But
spectators side with whichever they choose.
A taped intemiew by Jacques Bontemps, Jean-
Louis Comolli, Michel Delahaye, and Jean Narboni,
Cahiers du Cineina #194 (October 1967) pp. 13-26,
66-70; reprinted by permission. Slightly abridged -omissions available from translator.
Why did you ask Francis Jeanson to be in the
movie?
Because I knew him. So did Anne Wiazemsky.
She'd studied philosophy with him. That meant
they'd be able to talk. Anyway, Jeanson's the kind
of man who really likes talking to people. He'd even
talk to a wall. He has the kind of humanity Pasolini
defined when he said, in the movie Fieschi
made about him for television, he didn't like talking
to dogs in the familiar terms ~ou'res upposed to use.
In any event, I needed him, Francis Jeanson, not
someone else, for a TECHNICAL reason: the man
Anne talked to would have to be a man who understood
her, who'd be able to fit his speech to hers; it
would be just that much harder when Anne's text,
if you can call it a "text," wasn't her own: I whispered
it to her. I'd tried to find phrases that didn't
sound too much like slogans. But they'd still need to
be linked. So I had to have a man with Jeanson's
skill. As it was, and although he was replying to
really disjointed remarks, he always found the right
answers; it looks like a coherent conversation, now.
I was really relying on the allusion to Algeria. It
places him well. It outraged Sollers. Others just say
Jeanson's an ass, and leave it at that. It's a mistake,
if only because he agreed to play a role. Others refuse-
Sollers is one; I asked him to be in my next
movie; so is Barthes; I'd asked him to appear in
Alphaville. They were afraid they'd look like fools.
That isn't the issue. Francis has the sense to know
that an image isn't anything but an image. All I ask
people to do is listen. Start by listening. I was afraid
I'd hear people say what they said when they saw
Brice Parain in Vivre sa Vie, that "they wished that
old shit would shut up," or even that I'd meant to
mage him look a fool. Because of the allusion to
Algeria, they can't. When I interview someone, independently
of the personal reasons I have for preferring
one man to another, the position I take is imposed
by technique. Because he'd taught Anne philosophy,
I thought at first that I'd film a lesson in
philosophy-a mind giving birth to an idea,
prompted by Spinoza or Husserl. But it became in
the end what you see in the movie now: the idea
being that Anne would reveal to him plans of action
he'd try to dissuade her from, but that she'd go
GODARD
ahead with it anyway. To know whether that all
exists only in fiction is another question; it's hard to
say; when you see your own photo, do you say you're
a fiction? To have an interesting debate on this whole
thing ~ou'd have to have Cervoni, say, for the one
side and somebody from the Cahiers Marxistes-
Leninistes for the other. Or Regis Bergeron and
Rent Andrieu. They'd cover each other with shit
for a start; but they might, still, come up with something
in the end; but only if they'd agreed to start
with film before they finally get into it.
The reaction from the Marxist-Leninists wasn't
the one you'd expected.
No, it wasn't. They didn't know what to think at
the Chinese Embassy. They were really ut out.
Their big omp plaint was that Leaud isn't a17 bloody
when he unwraps the bandages. They obviously
haven't understood. That doesn't mean, of course,
that they're wrong; but, if they're right, they're right
at the first remove and not the second, or vice-versa.
They were afraid, too, the Soviets might take advantage
of Henri (a character who for a good many is
far more convincing than I ever thought he'd be) to
justify their own position. They weren't too far off
the mark: AndrC Gorz (Henri reads some passages
from his book Socialisme diflicile in the first shot)
was telling me it was "the first time he'd really liked
one of my movies; it was clear, coherent; the concrete
triumphs over the abstract, et cetera." I guess
I didn't make it clear enough that the characters
aren't members of a real Marxist-Leninist cell. They
ought to have been Red Guards. I'd have avoided
certain ambiguities. The real activists-the kids who
publish the Cahiers Marxistes-Leninistes; they impress
you with their real, deep commitment-maybe
wouldn't have been as annoyed by it as they were.
Because they shouldn't have been. It's a superficial
reaction, I think, not too far different, when you get
down to it, from the kind of reaction it got from the
collaborators on Le Figaro: "It's ridiculous! They
say they want to make a revolution. Look where
they're going to make it-in a plush bourgeois flat."
Though this is said in the movie itself, quite clearly.
Can you explain this sort of misunderstanding?
People still don't know how to hear and see a
movie. That's what we need to be working on now.
For one thing, the people who have training in
politics hardly ever are trained in film too, and viceversa.
My training in politics came out of my work
in film; I think it's almost the first time that ever
happened. Even if you think of a man like Louis
Daquin, you realize all he's doing is coming to film
with an education he's gotten elsewhere; a poor one
2 1
at that. As a result, the movies he makes are just
fair; they aren't the good ones he might have made.
All right, what can I say for my movie from this
point of view? I can say I think it quite clear that it
views the two girls with sympathy-with something
like tenderness even; that it's they who form the
support for a certain political line; and, finally, that
you have to start with these bvo girls if you're going
to understand its conclusion. It's anyway Chou
En-lai's. They haven't made a Great Leap Forward.
The Cultural Revolution is only the first step in
another Long March ten thousand times longer than
the first. If you now apply this conclusion to the
personal cases, the character played by Anne Wiazemsky,
prepared as she is for it, is bound to go
farther. So is the character played by Juliet Berto.
Lkaud really goes a long way: he finds the right kind
of theater. Henri makes a choice; he decides for the
status quo; he sides with the French Communist
Party; he's at a standstill, somewhere inside himself
-the fixed-frame shot, the absence of cutting inside
the shot characterizes this. As I view it, then,
he's cut himself off from all the real problems-but,
I repeat, only if in judging a movie you start with a
filmic analysis-it can be a "scientifically" or a "poetically"
filmic analysis, but it's got to be a filmic
analysis-and not the fictional or the political plot.
Krilov is the only one who really fails. This is all
quite clear. Anyhow, it's the Third World that
teaches the others the real lesson. The only character
in the movie who's really balanced is the young
black, I think. I wrote his speech too; it's coherent,
though it too is in fact made up of fragments: a
paragraph from the preface of Althusser's Pour
Marx, quotations from Mao, clippings from Garde
Rouge. Of course, though it's coherent, there's still
something to it that's slightly unsettling; Pierre Daix
has pointed it out: the questions they ask him have
less to do with the situation they find themselves in
than with much more general problems. Still, this
young militant agreed to be filmed, to use his real
name, and to make the slightly peculiar speech I'd
written for him. But we're talking now like men of
the same world-we might say the same cell. The
one really interesting point of view here would be
the view from the outside-the way it would look to
the Cuban movie-makers, for example. There's a
real gap between film and politics. The men who
know all about politics know nothing about film,
and vice-versa. So, I say it over and over again, the
one movie that really ought to have been made in
France this year-on this point, Sollers and I are in
complete agreement-is a movie on the strikes at
22
Rhodiaceta. They are typical-much more instructive
than the strikes at Saint-Nazaire, say, because,
viewed in relation to a much more "classical" kind
of strike (I'm not taking into account the hardships
they involved), they are, ~roperlysp eaking, modern
in the way the strikers' cultural and financial griefs
interact. The thing is, once again, the men who
know film can't speak the language of strikes and
the men who know strikes are better at talking Oury
than Resnais or Barnett. Union militants have realized
that men aren't equal if they don't earn the
same pay; they've got to realize now that we aren't
equal if we don't speak the same language.
Two or three years ago, you told us you thought
it extremely hard to make political movies: there'd
have to be as many points of view as there were
characters, and an "extragalactic" viewpoint as
well, to include them all. How do you feel about it
now?
I don't think so, now. I've changed. I think you're
right to favor the correct view at the expense of the
wrong views. The "elegant" Left would say that's
another one of the Little Red Book's truismsthough
I don't think they are truisms. If you're not
carrying out a correct policy, you're carrying out a
wrong policy. When I told you that, I was thinking
that you were obliged to be objective-the way the
press is "objective": you pay everyone equal attention-
or, as they put it, "democratic." But in the
sketch I've made for Vangelo 70 it's put quite
plainly that, on the one hand, there is what you call
"democracy," on the other, revolution; that's it;
that's all.
How do you feel now about the movie in which
you first got into politics, Le Petit Soldat?
It's okay for what it was. I mean, it's the only
movie a man born a bourgeois and just beginning
to make movies could have made if he wanted to
get into politics. The proof is that Cavalier used the
exact same theme when he made his movie on
Algeria. There just aren't that many. It's close to the
theme of some pre-war novels, Aurelian or Reveuse
Bourgeoisie-film lagged so far behind life. It's too
bad nobody else made his own movie about itthe
underground Jeanson organized, or the French
Communist Party. They'd have been hard to make,
of course. But, once again, if I didn't know what I
needed to be saying in my movie, the ones who did
didn't know how to say it in movies. My movie's
all right in so far as it's film; it's wrong for everything
else; which means it's just average.
Let's go back to the line that concludes La Chinoise.
It's put in the simple, preterite past and pronounced
in a "distant" tone of voice. Mightn't it
GODARD
risk, as a result, making us think everything that
precedes it a phantasy, a day-dream?
It's a simple, not a complicated past. The tone
isn't "distant": it's the tone of voice Bresson's heroines
always have. As for it being a "phantasy," it's
precisely because she's realized so much that Veronique
will be able to make it something more than
a day-dream. Besides, the tone in which she says the
line is soft; it's calm, like the Chinese. I was really
impressed at the Chinese Embassy by how softly
they speak. It's the tone of a final report. She realizes
she hasn't made a Great Leap Forward. Just
one timid ste in advance-though she has, in fact,
already seen i'ots of action; she's gone so far as to
kill the man who "never wrote Quiet Flows the
Don!"
A movie on the strikes at Rhodiaceta would have
led to a quite different kind of realisation . . .
Yes, it would. But if it were made by a moviemaker,
it wouldn't be the movie that should have
been made. And if it were made by the workers
themselves-who, from the technical point of view,
could very well make it, if somebody gave them a
camera and a guy to help then1 out a bit-it still
wouldn't give as accurate a icture of them, from
the cultural oint of view, as tffe one they give when
they're on %e picket-lines. That's where the gap
lies.
The movie-maker has to learn how to be their
relief.
Yes, he has to learn how to take his place in the
line. Learn how to pass the word along, a new way,
to others.
In La Chinoise, film assumes so many, such diverse
forms that they might cancel each other out.
The thing is, I used to have lots of ideas about
film. Now I don't, none at all. By the time I made
my second movie, I no longer had any ideas what
film was. The more movies you make, the more you
realize that all you have to work with-or against, it
comes down to the same thing-is the preconceived
ideas. That's why I think it's a crime that it isn't a
man like Moullet whom they hire to make movies
like Les Adventuriers or Deux Billets pour Mexico.
The way it's a crime that Rivette's being forcedhe
now after all the others who've been exploited
by the Gestapo of economic and aesthetic structures
erected by the Holy Production-Distribution-Exhi-
Mtion Alliance-to reduce a statement five hoi~rs
long to the sacrosanct hour and a half.
Do you think you've made any discoveries in
film?
One: what you must do to be able to make a
smooth transition from one shot to the next, given
GODARD 1
4
Anne Wiazemsky
and Jean- Luc
Godard during
shooting of
LA CHINOISE.
two different kinds of motion-or what's even harder,
a shot in motion and a motionless shot. Hardly
anyone ever does it, because they hardly ever think
of doing it. So, you can join any one shot and any
other: a shot of a bicycle to a shot of a car, say, or a
shot of an alligator to a shot of an apple . . . People
do do it, I guess, but pretty haphazardly. If you edit
not in terms of ideas, the way Rossellini edits the
beginning of India-that poses quite different problems-
but in terms of form . . . when you edit on the
basis of what's in the image and on that basis only
. . . not in terms of what it signifies but what signifies
it, then you've got to start with the instant the
person or thing in motion is hidden or else runs into
another and cut to the next shot there. If you don't,
you get a slight jerk. If you want a slight jerk, fine.
If you don't, there's no other way to avoid it. The
women who do my cutting can do it all by themselves,
now. I hit on it in A Bout de Souffle and I've
been using it systematically ever since.
You said you don't have any ideas about film now.
But it's still vey much there in La Chinoise. It's men
thematic . . .
It asks questions about film because film is beginning
to ask itself questions. I don't see anyway how
I could have kept it from coming into the movie
less than it does-though it tends in effect, paradoxically,
to narcissism. In this sense, the camera that
filmed itself in a mirror would make the ultimate
movie.
As in your sketch for Loin du Vietnam?
NO, not entirely. There wasn't any other way to
do it, there. It had to be pushed to just that extreme.
Because we are all narcissists, at least when it comes
to Vietnam; so we might just as well admit it.
Your characters think the Soviet communists have
"betrayed" Marxism. Do you think so too?
I've made a movie I call La Chinoise, in which I
adopt, against the point of view of the French Communist
Party, the point of view of the writings of
Mao Tse-tung or the Cahiers Mamistes-Leninistes.
I repeat, it is film that's imposed the direction I take,
which explains why the Cahiers Mamistes-Leninktes
can accuse it of being "leftist" and why m u -
maniti Nouvelle can even attack it for being a
"fascist provocation." But, even if there is some
truth in these opinions, it's still not quite that simple;
for, insofar as it's a question of film, the question's
been poorly framed.
How do you explain the impact the rmeotsionisC
Henri's siatement has had on a good many?
I hadn't foreseen it, but it makes sense to me now.
At one point, four gang up against one. That's all.
If you'd film Guy Mollet one against four, it's Guy
Mollet, that stupid ass, who as the underdog is going
to get all the sympathy.
Henri's the only one of the five who explains
himself completely.
No, you're wrong. People think he's the only one
who explains himself '%ompletely." The others don't
need to, to the extent that things are just that much
clearer for them. You have to take into account, too,
that peo le are apt to favor the guy whose views
they prefer; that, in any case, they're incapable of
being good listeners; and that they don't, in addition,
ever attempt to make a final accounting of
what they've heard the characters say.
Renoir has already asked what immediate dect
film might have. He's remarked that the war broke
out just after he'd made La Grande Illusion-a movie
24
in behalf of peace.
Exactly. Film hasn't the slightest effect. They
thought, once, that L'ArrivPe du Train en Care
would scare people out of their seats. It did-the
first time, but never again. That's why I've never
been able to understand censorship, not even its
ontological grounds. It seems to be based on a notion
that image and sound have an immediate effect
on the way people behave.
Though you can't really trace the influence an
image exerts. . .
Correct. But, then again, no more and no less than
the effects any of the rest might have-in other
words, no more than you can the effects of the whole
thing. Because everything exerts some influence. I f
you leave out that part of film that people call "television,"
we could say that film "has the influence" of
scientific research, theater, or chamber music.
Does this diminish your confidence in film?
No, not at all. But you've got to realize that the
millions of people who've seen Gone with the Wind
have been no more influenced by it than the many
fewer who've seen Potemkin. There've been some
attempts to blame film for juvenile delinquency. But
the people who've tried it don't seem to have noticed
that in precisely the same period that juvenile delinquency
was on the rise in the USA, movie-attendance
was dropping off sharply. The sociologists
haven't even begun to study the question.
The first shots you'ue ever made of the rural scene
come in La Chinoise: the two shots of the countryside
that renmrks on the farm-problem accompany
o f l . . .
Yes. L'HumanitP called them picture-postcards. I
don't know. All I can say is, as soon as we saw a
meadow, a cow, and some chickens, we stopped the
car and shot some footage. Then we turned around
and drove home. I don't see anything wrong in that.
I had to have these shots, because Yvonne had come
up from the country, and because one of my characters
had a couple of things to say about rural problems.
The character Juliet Berto plays is new for your
film.
I wanted something besides Parisians. I wanted
someone who'd come up from the country, so I
could illustrate another of the vices of our society:
centralization. Someone, too, who in contrast to the
others has nothing, who's dispossessed. Someone
sincere, who has a feeling there's something their
little group can do. She has access through them to
the culture that's been refused her. She used to
think it dropped from the skies. Then she started
GODARD
reading the papers. Now she's selling them. It's a
first step.
In the traveling shot along the balcony during the
theoretical presentations, the division of space by
the three windows diuides the " class" into three
groups: "professor," and Yuonne, the
maid, who's shining shoes or washing dishes the
whole time.
I had to show that even for those who'd like to
live without them, social classes still exist. It's just
at that moment you hear someone asking, "Will class
struggle always exist?"
The first two categories-"professor" and "students7'-
can still relate, interact. But the third is
cflectively kept to the side.
But it's only physically, not mentally, that she's
"forbidden" a part in the discussion. Or else it's
"tactically": because at the end of the movie she's
no longer forbidden to take part in it all. For one
thing, she's voted. There's no doubt she discovers
that it's she who, in the final analysis, has come much
closer to the others than they have to her personal
reality-which they should have explored, but they
haven't; they've put if off. So, of all the characters
it's the little farm-girl who covers the most ground.
Then comes LCaud, then Anne, then Henri.
The movie is made up of a series of short sequences
that seem to be quite independent of one
another.
It's the kind of movie that's made in the cutting.
I shot self-contained sequences, in no particular
order;I put them in order afterwards.
Does that mean it might have been diferent?
No, it doesn't. There was an order, a continuity
that I had to find. I think it's the one that's in the
movie. We shot it . . . in the order that we shot in!
Though as a rule I shoot the sequences in order, in
some kind of continuity; I mean, with some clear
idea of the movie's chronology and its logic-even
if I've found myself having to change the order of
whole sequences. This is the first time the order in
which I shot a movie presupposed nothing. It happened,
of course, that I'd know right when I shot
them that two different shots would go togethertwo
shots in the same discussion, for example; but
not always . . . For the most
pendent. The linking came Part, they were indeater.
So they aren't
independent now; they're at least complementary
if not also coherent.
That was the point of view on which you relied?
Was it some notion of a purely logical kind of coherence?
Or was it emotional? Or was it simply a
uiwl coherence?
GODARD
Logical. Always. But logic can be conveyed in a
thousand ways. Let's take an example. One of the
texts in the presentation is a speech of Bukharin's.
Right after it's read there comes a title: "Bukharin
made this speech." Next, you see a photo of Bukharin's
accuser. Of course, I could have used a
photo of Bukharin himself. But I didn't need to:
you'd just "seen" him in the person who reads the
speech. So, I had to show his adversary: Vichynskiand,
eventually, Stalin. okay: photo of Stalin. And
because it's a young man who speaks in the name of
Bukharin, the Stalin in the photo is young. That
takes us then to the time when the young Stalin was
already at odds with Lenin. But by that time Lenin
was married. And one of Stalin's greatest enemies
was Lenin's wife. So, right after the photo of the
young Stalin: photo of Ulianova. That's quite logical.
What has to come next? Well, it's revisionism
that toppled Stalin. So, next, you see Juliet reading
an ad in France-Soir: Soviet Russia is busy publicizing
Tsarist monuments. Right after you see the men
who in their youth killed the Tsar. It's a little like
a theorem that presented itself as a puzzle. You
have to see which pieces fit. You've got to use induction,
feel your way, deduce. But, in the final
analysis, there's only one possible way to fit them
together, even if you have to try several things to
find it.
So what you do when you edit is work that most
movie-makers do in their rhooting-scripts.
In a sense, yes. But it's work that just isn't interesting
if you do it on paper. Because if it's paper
work you like, I don't see why you make n~ovies.
On this point, I'm in agreement with Franju: as
soon as I've imagined a movie, I consider it made:
I can more or less tell it; so why should I go ahead
and shoot it? Oh, to do right by the public, I guess:
Franju says it's "so the public has something to
chew on." He says something like this: "When I'm
done with my eight hundred pages, I really don't
see what else I've got to do. So they want me to
shoot it. Okay. I shoot it. But it's all so depressing,
I have to get drunk first." There's just one way to
avoid that: don't write scripts.
So it's us if you shoot in the dark, but in complete
freedom too?
No, that isn't it. It's only in shooting that you
find out what you've got to shoot. It's the same
thing in painting: you put one color next to another.
Because you make film with a camera, you
can just as easily get rid of the paper. Unless you
decide to do what McLaren does-and he's one of
the greatest men working in film-and write your
2 5
movies right on the stock.
So when you shoot it's as if you cbllect a lot of
stuff you have to sort later. . .
No, it isn't. It's not just "a lot of stuff." If it's a
"collection," it's a collection that always has a
particular end in view, a definite aim. And it isn't
just "any" movie: it's always a particular movie.
You "collect" only the stuff that can meet your
needs. It's almost the reverse for my next movie:
the structure's all there; it's entirely organized. All
I had for La Chinoise were the details, lots of details
I had to find how to fit together. I've got the
structure for Week-end,but not the details. It's sort
of frightening: what if I don't find the right ones?
What if I can't keep my promise-because, after all,
for the money they give me, I promise to make
them a movie. No, that's all wrong. You shouldn't
think about work in terms of a debt or a duty-in
the bad sense of the word; you should think about
it in terms of some normal activity: leisure, life, and
breathing evenly; the tempo has to be right.
One of your characters says that Michel Foucault
has confused words and things. Do you share his
opinion?
Oh God, the Reverend Doctor Foucault! The
first thing I did was read the first chapter in his
latest book, the analysis of Velasquez' las Menim.
I skipped through the rest of it; I picked up a little
here and there-you know I can't read. Some time
later I was at Nanterre, looking for locations. In
talking to students and professors there, I began to
ap reciate the real inroads the book had been
maKing in the academic establishment. So I went
back to it again, with this in mind. It began to look
really debatable. The current vogue for the "humanities"
in the daily press seems very suspicious.
I heard that Gorse had been thinking about making
Foucault head of the Radio-Television. I have
to admit I preferred Joanovici.
In this connection, how do you view the use of
linguistics in the study of film?
As a matter of fact, I was just talking about it
with Pasolini, at Venice. I had to talk to him because,
as I've told you, I can't read, or at least not
the stuff men like him have been writing about film.
I just don't see the point. If it interests him, I mean
Pasolini, to talk about "prose film" and "poetic
film," okay. But if it's somebody else, well . . . If
I read the text on film and death Cahiers published
in French, I read it because he's a poet and it talks
about death; so, it's got to be beautiful. It's beautiful
like Foucault's text on Velasquez. But I don't
see the necessity. Something else might be just as
26
true. If I'm not so fond of Foucault, it's because he's
always saying, "During this period, people thought
'A,B,C'; but, after such and such a precise date,
it was thought, rather, that '1,2,3'." Fine but can
you really be so sure? That's precisely why we're
trying to make movies so that future Foucaults
won't be able to make such assertions with quite
such assurance. Sartre can't escape this reproach,
either.
And what did Pasolini say?
That I was a stupid ass. Bertolucci agreed, in
the sense that I'm too much of a moralist. But . . .
Well, I'm still not convinced. It means you're going
to wind up in the kind of "filmology" they used to
teach at the Sorbonne, or even something much
worse. Because, when you get right down to it, Sam
Spiegel's in complete accord with all this stuff
about "prose film" and "poetic film." Though he'd
say that "he's going to make 'prose film': 'poetic
film' bores the public shitless." It's the same old
thing all over again: people borrow and then distort
some interesting ideas; Hitler revisiting Niebsche
. . . I view linguistics the way Leclerc might
-or, even worse, Poujade. But I still have to agree
with Moullet. At Pesaro he talked commonsense . . .
But it's precisely a man like Levi-Strauss who
refuses to make random use of linguistic terminology.
He uses it only with the greatest caution.
I agree. But when I see him use Wyler as an
example when he talks about film, it makes me unhappy.
I tell myself that if he, as an ethnologist,
prefers the Wyler tribe, I much prefer the Murnau
tribe. Here's another example: Jean-Louis Baudry
has published an article in Les Lettres Francaises.
As I was reading it, I kept saying, "This is really
good writing! Here's a guy who ought to write
something on Persona. He'd do a really good job."
This thing is, the article I was reading was supposed
to be an article on Persona. Metz, too; he's a peculiar
case. He's the easiest to like of them all: because he
actually goes to movies; he really likes movies. But
I can't understand what he wants to do. He begins
with film, all right. But then he goes off on a tangent.
He comes back to film from time to time;
he'll poke around in it for a bit. But then he's off
again on another track. What bothers me is that
he seems not to have noticed; it's unconscious. If
it were a question of research in which film were
only a tool, I'd see no objection. But if it's film
that's supposed to be the object of the research,
then I don't understand. It's not that there's contradiction
in what he's doing; it's more like some
real antagonism.
GODARD
But Metz just isn't interested in what interests us.
All right. But there's still some common ground
it's all got to be based on. The way it looks to me,
they leave this common ground much too often. I
can understand, in some general sense, the intuitions
Pasolini begins with; but I don't see the need
for the logical development that follows. If he
thinks a shot in a movie of Olrni's is "prosaic" and
a shot in a movie of Bertolucci's "poetic," all right.
But, objectively, he could say just the opposite.
Their tactics resemble Cournot's when he rejects
one whole kind of film because, in his view, it just
"isn't film"; so, he's forced to reject Ford; but only
because he can't tell Ford from Delannoy! That's
not in the least enlightening. This all brings to
mind Barthes' recent book, the book on fashion. It's
impossible to read, for one simple reason: Barthes
reads things he ought to be seeing and feeling instead:
it's something you wear, so it's got to be something
you live. I don't think he's really interested in
fashion: it isn't fashion as such that attracts him;
it's some kind of dead language that he can decode.
You had the same kind of thing at Pesaro. Barthes
scolded Moullet the way a father scolds his kids.
But we're the sons of a filmic language; there's
nothing in the Nazism of linguistics we have any
use for. Notice: we always come back to how hard
it is for us all to be talking about "the same thing."
The people who publish Tel Quel seem capable of
making some really basic discoveries in science and
literature. But as soon as it's film, something seems
to elude them. Men who know film really well talk
about it in quite different terms-whether it's you
on Cahiers or Rivette and I when we're talking
about the movies that have just come out or the
people on Positif when they're talking about Jerry
Lewis or Cournot when he says of Lelouch that
"it isn't a question of 'feeling,' but it isn't a question
of 'thinking', either." This reminds me again of
the talk I had with Sollers. He re
talking "in examples." "He said I \roached me for
ept saying "it's
the same thing as" or it's like." But I don't talk
"in examples." I talk in shots, like a movie-maker.
So I just had no way to get him to understand me.
I'd have had to make a movie we could have talked
about afterwards. What it signifies on the screen
for him is maybe what "signifies it" for me. There's
got to be something right there that we've got to
clear up; it's probably pretty simple, too. It's somewhat
similar with painting: if Elie Faure moves us,
it's because he talks about a painting as if he were
talking about a novel. Somebody should finally get
around to translating the twenty volumes of EisenGODARD
27
stein that nobody's read: he'll have dealt with it
all in very different terms. He began with technique,
too, the very simplest problems, so he could get on
to the hardest. He goes from the travelling to NB
theater so that he can get back to explaining the
Odessa Steps. The place to look for an ideology is
in a technique. The way Regis Debray finds the
revolution in Latin America in the guerrilla. The
only thing is, the ideology of film has so decayed,
it's so rotten that it's harder here to make a revolution
than anywhere else. Film is one of the things
that exists in purely practical terms. You'll find that
here, too, the economic forces at work have laid
down an ideology of their own that has, little by
little, eliminated all the rest. The others are beginning
to re-emerge, right now; some of the best
are among them. In this connection, a lot of the
stuff Noel Burch has written is very interesting.
What he has to say about raccords is stictly practical.
You have a feeling they're the view of a man
who's done it himself, who's thought about what is
involved in doing it-a man who has come to certain
conclusions on the basis of his physical handling
of film. Well, all you'd need to get it all down
in a orderly list is some serious, well-organized
team-effort. The best work a new nation could do
to get started is something along those lines. All
they've got to do is buy some good movies, start
a film library, and study movies. They can make
them later. They can learn while they're waiting.
Before getting yourself involved with what the linguists
call a "scientific" analysis of film, you'd do
better to list the scientific facts of film. Nobody's
done it. Though it still could be done: the projections
at the Grand Caf6 weren't that long ago;
Niepce's first plates are still at Chalon. But if you
wait too long, you won't be able to do it. Movies
disintegrate. Even books fall apart. Movies fall apart
a lot faster. In two hundred years you won't be able
to find a single one of our movies. There'll be a few
bits and pieces-of bad movies as well as the good:
the laws to protect the good movies still won't have
been made. So, the art we're working in is really
short-lived. When I started to make movies, I
thought film something that lasts forever. Now I
think it something really short-lived.
So the incompatibility in the language of the
writers and movie-makers is just as severe as it is
for movie-makers and the strikers at Rhodiacetathough
the writers have already had a good deal
to say about film.
Well, if they have, it's often only because movies
sometimes refer to literary forms or simply just
cite literary texts.
Do you think it's your w e of collages that leads
Aragon to write about you?
Maybe it's the digressions that have attracted
him: the fact that there's someone who uses them
as digressions, besides as a structural device. In any
event, Aragon is a poet, which means that anything
he has to say is beautiful. If you don't talk about
films in poetic terms, then you've got to be talking
about it in scientific terms. We haven't reached
that point yet. Notice this one simple fact: you go
to a theater to see a movie; you never ask why;
though there is simply no reason why movies should
be shown in theaters. This in itself is revealing.
Of course, they way things are, you've got to have
theaters. But they shouldn't be more than something
like a deconsecrated church or a track field:
you should hold onto them; people will go a theater
to see an occasional movie; there'll be a day when
they'll want to see a movie on a big screen; or like
the way an athlete will go out to train by himself
in the middle of the week; he wants to be far from
the frenzy, the racket, the drugs of the weekend
meets. Ordinarily, you should be able to see movies
at home, on a television set or a wall. It's feasible,
but nobody's doing anything about it. For a long
time, now, the factories ought to have had screening-
rooms; someone should have investigated what
increasing the size of TV screens involves, practically.
Nobody has. They're all scared.
Wiazemsky and Lkaud: LA CHINOISE.
28
Do you think there is a connection between the
ways film is distributed and exhibited-theaters,
chains, and so on-and its aesthetics?
If these conditions were to change, everything
else would change, too. A movie is subject today to
an unbelievable number of really arbitrary rules.
A movie is supposed to last an hour and a half.
A movie is supposed to tell a story. All right. A
movie tells a story. We all agree. The only thing is,
we don't agree on what a "story" is, what it's "supposed"
to be. You see, today, that the silents had
immeasurably more freedom than the talkies-or,
at any rate, what they turned the talkies into. Take
as unimaginative a director as Pabst: he gives you
a feeling that he's playing a grand. A movie-maker
today who has no more than Pabst's talent, if he
analyzes his own case correctly, has to feel that he's
playing not much more than a toy. It's all a
state of mind. For example, when someone's building
a theater, he never takes the trouble to ask the
advice of a cameraman or a director. And nobody's
ever going to ask advice of a viewer. So, as a result,
the three most interested parties never have a
chance to make their desires known. It's true they
I~uild houses this way too. But the guys who design
theaters are always the worst they can find. And
they're never the ones who go to see movies.
What could we do, at short range, to change it?
The best we can do is attack the technical problems,
everything that results from the economic
forces at work in film: production, processing, proiection
. . . The young men who are just getting
their start in film don't have to know everything
about it. They can get along very well without
knowing anything about Lumicre or Eisenstein.
They'll run into them sooner or later then~selves.
The way it isn't until he was thirty that Picasso
got onto African art. .And if he hadn't just then,
he'd have painted Les Demoiselles d'duignon a
few years later. He'd have done something else in
the meantime. The young men have all the luck:
they can always start over. People have been doing
a lot they can benefit from, even if it's been fairly
haphazard, disorganized. They need to make a
long list, get everything on it, the little things as
well as the most important: everything involved in
film that lust won't do. Everything: from theaterseats-
the worst are in the art-houses-to editingtables.
I bought an editing-table recently. It didn't
take me long to discover that nobody had asked
the right questions. They're manufactured by men
who've never done any editing. I'm holding onto
it. I'm hoping I'll get the money to have it rebuilt,
so that it will work right.
GODARD
In what sense has it been badly conceived?
The way they're manufactured is the result of a
particular aesthetics. They've been conceived as
little projectors. That's fine for men who think
editing a few pencilled notes: the director shows
up Monday morning; he tells his cutter where to
make cuts and splices; she takes the footage off
the editor and does the work she's been told to do
at another table. Or, if it's someone like Grangier
or Decoin she works for-they just can't be bothered,
she'll do the whole thing herself. But in any case,
the real editing gets done somewhere else, not at
the editing-table itself. But, there are movie-makers
-Eisenstein's the first, Resnais is the second, I'm
the third-who do their editing, each in his own
way, of course, right at the editing-table, with the
image and against the sound. The problems you
have with handling the film are completely different.
I keep winding the film back and forth.
I make splices without ever taking the reels off.
And if the table hasn't been manufactured with
work of this sort in mind, it's not easy to do it.
Again, it comes down to a simple economic gimmick
that all by itself bears out a whole ideology.
If that's how they manufacture editing-tables, it's
because three-fourths of the people editing film
edit this way. Nobody's ever told the manufacturers
to do it differently. I use editing as an example,
but the same kinds of thing turn up everywhere
else. If you're trying to make revolutionary movies
on a reactionary editing-table, you're going to run
into trouble. That's what I told Pasolini: his linguistics
is a shiny, new, reactionary editing-table.
Besides, the more movies I make, the more I realize
just how precarious a thing a movie is: how hard
it is just to get it made, and then how hard it is to
get it shown-in other words, just how distorted
the whole thing is. If problems like these were ever
solved-though I don't think they'll ever be solved
in the West-then we just might discover some new
ways of working-ways to make film that's really
new. Things as new as the discoveries made in the
very first years of film. Everything we're using now
was invented in the first ten or twenty years of the
silents. Technique was moving right in step with
production and distribution, then. Right now, we've
lost sight of the ways they're connected. Everything
goes its own way-if you think it's going anywhere
at all. The only thing I'd want to write for Cahiers
now-it would take time to do it; I'm always running
into something else to say on the subjectwould
be something about the ways to get film off
to a complete new start. I'd discuss it in terms of
the problems a young African would have to face.
GODARD 29
I'd tell him, "All right, your nation has just won
its freedom. Now that you're free to have a film
of your own, you and your comrades have been
asked to get it started. Okay. Get Jacquin and
Tenoudji out of your theaters."-You know, even
in Guinea, the most revolutionary of the new nations,
the theaters all still belong to Comacico. And,
though the Algerians have nationalized their filmindustry,
they've handed it right back to the distributors,
which means that in no time at all it'll be
back in private hands again; it'll be just the way it
was before. "You've decided to have a film of your
own, to make film of your own. This means that
you're not going to import any more trash like La
Marquise des Anges. Book Rouch's movies, or
movies made by some young African he's trainedanything
that interests you. If you work for De
Laurentiis, don't go to him. Make him build studios
for you here. In other words, since you have it all
still to do, turn it to your advantage. Make a thorough
investigation of everything that's involved in
the production and distribution of movies. Build
or rebuild your theaters-or what might replace
them in the eyes and the hearts of your militant
countryn~en." Things like that. It's impossible to
list the mistakes that have to be corrected. You'd
have a list as long as the lists in Rabelais or in Melville.
But you'd have to try, if you really wanted
to redefine film. To get back to Algeria: They should
use the money they've made in co-production deals
to build processing plants of their own, not in
financing (aside from a couple of things like Le
Verzt des Aures) Jacquin's movies: I know it's hard
to believe, but half the money in Le Soleil Noir is
the Algerians'. They haven't even got their own
processing plants: they send their newsreels to
France or Italy, on Air France or Alitalia; they
don't trust Air Algeria.
It's sometimes only too obvious that movie-makers
in the new nations imitate the very worst in
our film when they're making their own first shorts.
Of course it's also an individual, a mental problem.
But if you want to get off to a start, you've got
to base yourself on a non-mental thing-on technique.
The new mentality can develop out of it.
Obviously, things are hard, everywhere. The director
of the Algerian Film Center is convinced he's
better off having Jacquin or Tenoudji distribute
his movies. That's the tragedy of the Third World:
it's always in a corner, always in a jam for money.
Everyone's in league against it, the way they've
all ganged up against the unemployed. The Algerians
produce Italian movies instead of movies by
young Algerians. They did give them some film,
but the kids used it to make irresponsible junk.
They'd do better, in such a case, to put a stop to
their production for a time and give the kids the
opportunity and the time to do a little homework,
research, and to see as many good movies as possible.
The crisis will take care of itself. Or they could
put them to work in television or in the processing
plants and the sound studios. It would be all that
more practical because no director, anywhere, really
knows what goes on in an editing-room or a film
lab. Everybody in film ought to get some training
in the sector closest to his own. Cameramen, for example.
They learn a little in school, but then they
never go on to get some training in the film labs.
As a result, the cameraman and the lab are never
able to reach an understanding. Let's say you shoot
a movie with a man who's a real master of light.
Let's say he's as familiar with Renoirs as with
Rembrandts. Fine. The print will be timed by a
man who hasn't the faintest idea what lighting is.
No more Renoir's than Rembrandt's. So, as a result,
the print will be too dark, or too light, but in
any case flat. Simply because the lab technician
neither knows what he can nor what he should do.
Or just the opposite. I just remembered that Matras,
when he was in Madrid, spent his time sending his
wife Mexichrome postcards instead of looking at
pictures in the Prado. You run into the same thing
at every level in film. Nobody's really been educated.
It's a question of education. Right here in
France, there's all you'd need to make really good
movies. But the men who are supposed to be directing
the work are lazy bums or highway-robbers,
They employ honest men, but they give them no
training and no responsibilities. The people who
do the actual work think they're doing it right. But,
the thing is, they're imprisoned in a whole system
of economic and aesthetic preconceptions. What
YOU have to do, then, is explain it to them. For
example, you can explain to a projectionist that
there just isn't any point in closing and opening
the curtains: film isn't theater . . . And if projectionists
are so badly paid, it's because no one thinks
the work they do is work of any real importance.
There's as little respect for them as there is for
the grips or the sound-men. A grip knows a good
deal. He can, often, talk much better sense about
film than his director. But he "doesn't count." And
as for the men who manage the sound, they're paid
even worse than the men who make the image.
Why? It's a result, once again, of a whole ideology.
SO they say, "Why should we pay the sound-man
as much as we pay the director of photography?
Film is the art of the image!" That's all wrong. But
30
the sound-man continues to get half what the
cameraman gets-and, what's worse, to think it
fair. If we start talking about distribution, we run
right into another problem: the distributors. Film
got off to a start without them. All it took was a
cameraman and a director. What did Lumiere do?
He took his movies right to the guy who ran the
Grand CafP. All right. But since then, distribution
has become a trade. The middlemen-the distributors-
are lazy. They don't make a move. But they
still keep on saying (and it's as much for themselves
as it is for us), "You can't do without us. It's
all got to go through us." But the only reason that
there are "distributors" at all is that everyone else
is too lazy. The exhibitors won't move an inch to
find the product to sell. The producers won't move
an inch to take it to them. As soon as that happens,
they need the third man-who robs them blind in
the end. . .
You want to: make different movies. But to be able
to make them, you have to work with people you
despise and dislike, instead of with people you like
and admire. The industry's rotten to the core: from
the point where the film is processed to the point it
must reach-if it ever gets there-to reach a public.
From time to time, of course, there's a hint things
are beginning to move. The Hyeres festival, for example,
isn't ideal, but it's still a lot better than
Cannes; and Montreal's better than Venice. You've
got to keep moving ahead. Film in Canada is an
interesting case. The National Film Board is a real
movie factory. They're making more movies than
Hollywood now. A beautiful set-up. But what happens?
Nothing. There's nothing to see. Their movies
never get shown. One of the first things Daniel Johnson
should do is nationalize the theaters in Quebec.
In Canada, too, film is subject to the imperialism that
prevails everywhere else. Those of us who keep
trying to make movies differently have got to organize
a fifth column, attempt to destroy the whole
system.
But some film is already being made outside the
system . . .
Yes, of course. Bertolucci isn't making American
movies. Neither is Resnais, or Straub, or Rosselllini,
Neither is Jerry Lewis. But even this different film,
good or bad, is no more than 1/10000 or 1/100000
of what's being made.
But is there still a really "American" film?
No, there isn't. There's a counterfeit that calls
itself "American," but it's only a very poor copy of
what it was once.
Would you work for an American company again?
Yes, I would. If that's what I'd have to do to make
GODARD
a movie. Or if it gave me a chance to make an expensive
movie, like Michael, Circus Dog; I mean, a
movie for which more money goes into the image
than into the actors' pockets. In saying this, though,
I don't compromise myself or my view of America
and the imperialistic policies of its giant film companies.
In the first place, there are Americans and
Americans, good ones and bad ones. In the second
place, there, too, they need a fifth column. You might
get it into their heads that they could make different
films too. You might even get them to want to. If
the movie you made were a success, you might,
little by little, get them to change their system themselves.
It would be hard. You keep running into
their imperialism at every level of production and
distribution. But you've got to hold onto the hope.
People can change. Then again, something is on the
move in America right now. You can see it among
the blacks and in the opposition to the war in Vietnam.
And as for film, the universities are beginning
to distribute movies; they're turning into real chains.
New companies are being formed. I sold La Chinoise
to Leacock's. Anyway, the world's a little bit
larger than America. But if I put the Americans and
the Russians together into the same bag, it's because
their systems are almost identical. They both treat
their young movie-makers like naughty children.
Every one of the Americans we really admire got
his start in film at an early age. They're old now, and
there's nobody there to take over. When Hawks got
his start, he was Goldman's age now. Goldman's all
by himself. Obviously, there are young men still who
do get into Hollywood, but none of them have anything
like Hawks's ideas. They've gotten what training
they have in structures that are on their decline;
they haven't had the guts to destroy them. It isn't
in freedom that they come to film; though it isn't in
any real poverty, either, aesthetic or otherwise. They
are neither explorers nor poets. But the men who
made Hollywood were poets-even gangsters, who
took it by force to dictate their poetic law. The most
courageous man in Hollywood today, the only man
who's managed to get out from under it, is Jerry
Lewis. He's the only one in Hollywood who's doing
something different, who remains outside its categories,
its norms, its principles. Hitchcock did for a
long time. But Lewis is the only man who's making
courageous movies right now-and I think he's aware
of it. He can get away with it because of his personal
talent. But who else can? Nicholas Ray is typical
of the point American film has now reached.
The case of the New York School isn't encouraging,
either. They're already buried. And if it's "underground"
film they want to make, it's got to mean
GODARD
they'd like to be buried deeper. I don't see why. The
Russians haven't helped Hanoi bomb New York.
Why do they want to live underground? There are
going to be more great American movie-makers.
They've already got Goldman, Clarke, Cassavetes.
We'll just have to wait, help them, even ush them.
I was talking about the universities. F;m's being
made in the universities-or, at least, they're beginning
to; there didn't use to be any film there. That's
important. Film's got to go everywhere. We should
list the laces it hasn't been yet and then say that
that's w fiere it's got to go. If it's not in the factories,
it's got to get into the factories. If it's not in the universities,
we've got to get it into the universities.
If it's not in the brothels, it's got to get into the
brothels. Film has to get away from where it is
now and go where it hasn't been yet. .. .
Where and when you get your start has a lot to
do with how you get started. No one in France had
been taking film seriously. Then people turned up
who were saying you had to, that it deserved some
serious thinking. For the same reasons, we had to
say, too, that there is such a thing as a "work." I
don't think now that there is. That's a point you
reach if you push your thinking on art just a little bit
further. There is no such thing as a "work," even if
there is something that's kept in cans or printed on
paper, not in the way that there are such things as
beings or objects. But, at the time, that was the thing
we had to do first: force it on people that there was
"work," even if you have to tell them now that
they've got to go a little bit further in their thinking.
In the same way, I'll say too that there is no such
thing as an "author." But to get people to understand
in what sense you can say that, you have to tell them
over and over again, first, that there's such a
thing as an "author." Because their reasons for
thinking there weren't weren't the right ones. It's a
question of tactics. . ..
Aren't you increasingly influenced by theater?
You've got to do theater in film, I think-mix
things up a little. Mix it all up. Especially the festivals.
I think it's absurd that they don't hold the
music and theater festival at the same time as the
film festival at Venice. They should have music one
night, film the next . . . You remember how it was
at Pesaro: after you'd seen a movie you could go
and hear jazz; you had a really good time.
But when you say that, you've begun to attack
one of the public's biggest taboos-against the mixing
of genres. You begin to realize the damage done
some thirty or forty years ago when the "theoreticians"
would decree that something "was theater,
not film."
31
There are a lot of movie-makers right now who'd
like to talk about theater: there's Rivette and
L'Amow Fou; Bertolucci and others. Persona, Blow-
Up, Belle de Jour are part of it, too. And Shakespeare
Wallah; that's a beautiful movie. I suppose
it means that people who've gotten the feeling
they're trapped by their means of expression want to
get out of it. I'm not talking about Bergman now;
he's been doing theater all his life; he's done more
theater than he's made movies. For a long time,
now, I've been wanting to make a didactic movie on
theater, about Pour Lucrice. At the beginning you'd
see the girl who'd act the role get out of a cab; she'd
be going to a rehearsal; no, not a rehearsal; she'd
be going in for an audition. Then you'd get into the
play. You'd see an audition, a rehearsal, a scene in
performance. From time to time, there'd be some
critique of the play itself. Some scenes would be
done two or three times: the actors would make
mistakes or the director would want to get something
just right. You could have the same scene done
by several actors: Moreau, Bardot, Karina could
each act the same role. And the director could review
the seven or eight great theories of theater
with the actors: Aristotle, the three unities, the
Preface de Cromwell, The Birth of Tragedy, Brecht
and Stanislavsky-but they'd be doing it in the play,
still. At the end, the girl you saw coming in at the
start would die: because Lucrece dies; you wouldn't
know where the fiction stopped, then. A movie like
this would aim to teach an audience what theater is.
Readings are just fantastic! When you get right
down to it, the most fantastic thing you could film is
people reading. I don't see why no one's done it.
Film someone who's simply reading . . . The movie
you'd make would be a lot more interesting than
most of them are. Why couldn't film mean filming
people reading really fine books? Why shouldn't
you see something like that on TV, especially now
that people don't read much? And eople who can
tell good stories, make them up-life Polanski, Giono,
Doniol. They could make u stories right in
front of a camera. People would %ten to them. If
somebody's telling a really good story, you can listen
for hours. Film would be going back to the
traditions and role of the Oriental storyteller. We
lost out on a lot when we stopped being interested
in storytellers. But the ideology that tells us what a
spectacle "is" is so firmly established that the people
who'd been spellbound by the story you'd been telling
them at the Gaumont-Palace would come storming
out in a rage; they'd say you'd tried to take them
for fools; they'd say they'd been robbed.
But you don't question spectacle itself . . .
32
No, I don't. If you look at something, it's a spectacle,
even if it's just a wall. I've always wanted to
make a movie about a wall. If you really look at a
wall, you wind up seeing things in it.
One gets the impression that there's an intention
to destroy the image itself at work in your sketch
Anticifation, to destroy it as the support for "realism.
It annoyed me that it was much too easy to identify
the actors. When I started shooting it, I still
hadn't thought of anything like it. It was only later
that it occurred to me to give the movie-you could
call it a "biological" look-like plasma in motion.
But plasma that speaks.
But the minute you do that, you attack an idea
that's almost sacred: the idea that an image in film
is sharp, clean, "solid" . . .
But an image is always an image, as soon as it's
projected. So I haven't destroyed a thing. Or else,
one idea of the image and what it's supposed to be.
1 never thought of it as destruction . . . What I
wanted was to get inside the image, because most
movies are made outside the image. What is an
image? It's a reflection. What kind of thickness does
a reflection on a pane of glass have? In most film,
you're kept on the outside, outside the image. I
wanted to see the back of the image, what it looked
like from behind, as if you were in back of the
screen, not in front of it, Inside the image. The way
some paintings give you the feeling you're inside
them. Or give you the feeling you can't understand
them as long as you stay outside them. Red Desert
gave me the feeling the colors were inside the camera,
not out there in front of it. The colors are all
in front of the camera in Le Mkpris. You're convinced
it's the camera that makes up Red Desert. In
Le Mkpris, there is the camera, on the one hand,
the objects on the other, outside it. I don't think I'd
know how to make up a movie like his. Except that
I'm beginning to want to. You can see my wanting
to in Made in USA. That's why people haven't understood
it. The people who've seen it think it's
supposed to be "representational," but it's not. I
must have put something over on them, because
they kept trying to follow it "representationally":
they kept trying to understand what was happening.
They did keep up with it, quite well. But they didn't
know they had: they kept thinking they hadn't understood
a thing. It really impressed me that Demy
was so fond of Made in USA. I'd always thought it
a movie "in song"; La Chinoise is a movie "in talk."
The movie Made in USA resembles the most is Les
Parapluies de Cherbourg. The actors don't sing, but
the movie does.
GODARD
Now that you bring up resemblance, is there a
connection between Persona and your last few movies?
No, I don't think so. And anyway, I don't think
Bergman likes my movies too well. I don't think he's
taken anything from me-or from anyone else, for
that matter. Anyway, after In a Glass, Darkly, Winter
Light, and The Silence, he could hardly have
made anything but Persona.
Persona is much more daring stylistically than the
preceding movies. The way the narration is "doubled,"
for one thing. . .
No, I think the shot you're talking about is, aesthetically,
just a continuation or a development of
the long shot in Winter Light in which Ingrid Thulin
confesses. But it's much more striking in Persona,
of course; it's close to formal aggression. It's so
striking as a formal device that as soon as you see
it you tell yourself "it's so beautiful; I've got to use
it in a movie myself." I got the first shot for my next
movie when I was seeing Persona again. I told myself
that what I needed was a fixed-frame shot of
people talking about their genitals. But in another
sense, it reminds me of the opening shot in Vivre sa
Vie: I stayed behind the couple during the whole
shot, but I could have gone round in front. What
he's doing is something like what the interviews are
in my movies; it's very different in Bergman, but,
in the final analysis, it always comes down to the
desire to represent a dialogue. And it has something
to do with Beckett, too. At one time I'd wanted to
film Oh! les Beaux lours. I never did-they wanted
to use Madeleine Renaud; I wanted to use young
actors. I'd have liked to-I had a text, so all I'd have
had to do is film it. I'd have done it all in one continuous
travelling. We'd have started it as far back
as we had to to get the last line, at the end of an
hour and a half, in a close-up. It would have meant
just some grade-school arithmetic.
How do you interpret what in Persona keeps reminding
you it's a motiie you're watching?
I didn't understand Persona. Not a thing. Oh, I
did watch it, carefully. This is the way it looked to
me: Bibi Andersson is the one who's ill; it's the
other girl who's the nurse. When you get down to
it, I guess I always rely on the "realism." So, when
the husband thinks he recognizes his wife, I think
she's his wife: he's recognized her. If you didn't
rely on realism, you'd never be able to do anything.
If you were on the street, you wouldn't dare to get
into a cab-if you'd even risked going out, that is.
But I believe in it all. You can't divide it up into
two; you can't separate the "reality" from the
"dream"; it's all one. Belle de Jour's really great.
GODARD 33
There are moments when it's just like Persona. You
say, all right, beginning now I'm going to follow it
carefully, so I'll know just exactly where we are;
then, all of a sudden, you have to say, damn it!
we're already there: you see you're already in it.
It's as if you decided you wouldn't go to sleep so
that you wouldn't be asleep when you went to sleep.
That's the problem these two movies pose. For a
long time now, Bergman's been at a point where it's
the camera that makes the movie, eliminating everything
that can't become part of the image. That
ought to be axiomatic for all editing, and not notions
like "the pieces have to be put together in just
the right order," or "there are rules that must be
observed." You ought, instead, to say that you've
got to eliminate everything you can say. Even if you
have later to turn it all inside out and say that all you
can keep is what is said. That's what Straub, for
example, does. In La Chinoise, it's only what's said
that I keep. But the result is completely different
from Straub's, because it isn't the same thing that's
said. Butiuel eliminates everything that is said, since
even what's said is there to be seen, too. There's a
fantastic freedom in his movie. You get a feeling that
Buauel can "play" film the way Bach must have
played the organ at the end of his life.
How do you view the notion of the "door-to-door
theater" Leaud picks up on at the end of La Chinoise?
I'm afraid it hasn't been understood. I suppose I
didn't make it clear enough. It's not he who's in
question; it isn't an individualistic solution. The way
I'd been thinking about it, I'd have had to show him
together with others. One would have been playing
a guitar; one would have been singing or drawing on
the sidewalk-the kinds of things hippies do in front
of cafes. But this time they'd be communists doing
them. They'd have been doing real work: they'd be
having to choose their text for the given situation,
switching from Racine to Sophocles or something
else. I really ought to have had more than one doing
it. There'd have been times when they wouldn't
know what to say; they'd have to talk it over, to decide
which was the right response. They might even
start talking to the people watching them, engage in
real dialogue. Instead of acting theatrical texts, they
could have recited some Plato. There shouldn't be
any restrictions. It's all theater; it's all film; it's all
science and literature. If you'd mix things up a bit,
we'd all be a lot better off. For example, the lectures
in the universities could be given by actors;
the professors speak like they've got mush in their
mouths, anyway. And you could profit from that to
learn how to speak a text, too, how to read it. It's not
just the conclusion you reach when you come to the
end of Descartes' sixth Meditation that counts, or
having to be able to talk about his system on an
exam, but the time it takes to reach its conclusion,
the distance you have to go-in other words, the
experience lived in learning about Descartes. I'm
not saying this is the only thing that needs to be
done; but, after all, when thousands of things need
to be changed, I think you'd do well to try changing
just one or two, instead of saying right off, once
and for all, that it's good or it's bad.
Do actors, like movie-makers or technicians, need
more study? Do they need more training?
Training, yes. The kind the American actors used
to get. If I were giving a course for actors, I'd give
them physical or intellectual exercises to do, nothing
else. I'd tell them, "now you going to some
gymnastics" or "you're going to listen to this record
for the next hour." Actors have so many prejudices
in physical and intellectual matters. For example,
when we were making Deux ou Trois Choses, Marina
Vlady came up to me one day and said, "What
should I be doing? You never tell me." So I told her
-she lives in Montfort-L'Amaury-I told her to walk
to wherever we'd been shooting, instead of taking
a taxi. "If you really want to act well, that's the best
thing you can do about it." She thought I was putting
her on, so she didn't do it. I think I still hold it
against her; just a little bit. She might have done it
if I'd explained it all. But she'd only have done it
once, and then the next day she'd be expecting me
to come up with something else. So it just wasn't
worth explaining it. I wanted her just to think what
she had to say. That's all. Thinking doesn't have to
mean intellectualizing. If she was supposed to put
a cup down on a table, I wanted her to think an
image of a cup and an image of a table. Everything
that's involved in just walking to the location every
day would have put her in shape to move and speak
the way that would have been right for what I was
trying to do. What I asked her to do was a lot more
important than she thought, because to get to the
point where you can think, you've got to do a few
simple things just to get yourself into shape. Everyone
knows that a dancer can't dance unless he trains
himself for it every day, does his exercises. But the
idea that actors need "exercise" too was already on
the decline among the actors in theater. Film actors
haven't the slightest idea of what kind of exercises
they ought to be doing. They tell themselves that
since they don't have to kick up their legs there's
just no point in exercise. Before shooting started on
La Chinoise, I asked Jean-Pierre LCaud to eat. I
gave him the money-I told him he couldn't spend it
34
over at the CinCmathPque-just so he'd be eating a
meal, in peace and quiet, ninety minutes a day every
day, not reading the paper, not doing anything but
eating an ordinary meal in an ordinary restaurant.
That's what he needed to do for La Chinoise. Exercises
like these are a little like a reverse yoga. It's
the kind of thing the surrealists used to call "practical
exercise." They are needed in every activity, on
every occasion. Actors don't seem to remember
they're being paid for eight hours of work a day.
Just like factory workers. The thing is, as soon as the
worker reaches the factory he works-a full eighthour
day; he can't cheat. Actors can and they dolike
a lot of others in the white-collar Brofessions.
An actor doesn't work an eight-hour ay-if only
because you can't shoot eight hours straight. All I
ask is that he do more work between the takes and
less during them. Because, if he's done his work
before the take, I can be sure it'll be good. It doesn't
do any good if he has to do his work during the
take. The trouble is that it's the hardest thing there
is to get an actor to do. But even so, when we were
making La Chinoise they got along pretty well. They
worked well as a group; together they did just the
right kinds of things to keep them in pretty good
shape for shooting. It went a lot smoother than
hlasculin-FQminin.Obviously, now, every thing I've
been saying applies to professional actors as well.
Neither the professionals nor the nonprofessionals
are prepared to submit to the slightest training. Anna
Karina's like all the rest on this point. I kept telling
her, all I wanted her to do was, every day, read the
editorial in the paper, Le Figaro or L'HumanitP,
aloud, calmly. She didn't understand either. Even
though little things like this have a direct influence
on one's acting. It's exactly the equivalent of walking
for an athlete, scales for a pianist, limbering-up
exercises for an acrobat. The big problem with actors
in film is that they're often so very proud. So,
they've got to be taught to be humble, the way the
humble have to be taught to be proud. It's as Bresson
says, "give and receive." And from this point of
view, I see no difference between the professionals
and the nonprofessionals. There are interesting people
all over the place. But Bresson talks about actors
the way the Russians talk about the Chinese. I kept
telling him, "They've all got eyes, mouths, hearts
. . ." And he'd keep saying, "No!" If I'd said,
"Well . . . when Jouvet was still in his mother's
belly . . ." he'd have said, "Oh well, you know . . .
Predestination!"
There's a much larger problem involved in these
exercises you've prescribed: it's a problem of education.
For example, the characters in La Chinoise
GODARD
have all emerged from the bourgeoisie, which hm
given them the education they've begun to question.
The fact is, it all lies in the way they've gotten the
knowledge they have. Their education is an education
in class. The way they conduct themselves is
determined by class; they conduct themselves like
members of their class. That's all made very clear
in the movie, anyway. On the subject of this "education
in class" that prevails here in France, here's a
thing I cut out of a paper the other day; I'm keeping
it because I'd like to make a movie on Rousseau's
Emile. Missoffe-he's our Minister of Youth, remember-
is on record as saying-it's in his White Book
-and I quote: "The schools must translate the structure
of society into its programmes: it must organize
(1)a long and highly intellectual training for children
appointed in the main by their family origins
to the highest posts in the direction and administration
of society; ( 2) a shorter and simpler kind of
instruction for the children of workers and peasanfs,
whose entry into the labor-force, it would seem,
requires no more than a limited training." No comment.
Tell us something about your Emile. What will it
be?
A modern movie . . . The story of a boy who refuses
to go to his high-school because the classes are
always too full. He sets about teaching himself, on
the outside. He observes people, goes to movies,
listens to radio, looks at television. Education, just
like film today, is an immense accumulation of techniques
that need to be re-examined, corrected. Everything
needs re-examination. What's going to happen
to the son of a workman who decides he wants
an education? Right at the start, he'll find himself
in a jam for money. We always get back to the
Third World's problems. The whole system of scholarships
is really immoral. They are supposed to go to
those who "deserve" them. All right, who "deserve"
them? Because the schools are recruiting right now,
just like the army, and the kid who doesn't answer
the call just hasn't the right to pass his exam, those
who "deserve" them turn out to be the ones who
always come to class, which means, then, the ones
who can always afford to come, who don't have to
be working their way through school. Even if the
ones who attend every class don't necessarily learn
any more than the ones who miss more classes than
not. Then again, no one knows what to do to give
people the desire or the time to learn. Then again,
the teachers are so poorly paid! I don't say it's simple.
I'm just saying that there's much too much
that's totally unacceptable, right from the start.
Are you saying the problem has no solution?
WEEK-END 35
No! Because, all the same, it's nothing like it is in is why we can't say or do anything together. A
France in America, Russia, or even Albania. In the worker . . . I have to repeat myself-a worker has
first place, they spend much more on education than nothing to teach me, nor I him. It ought to be just
we do. In France, the restrictions placed on funds the opposite. There ought to be a lot I could learn
are the result of deliberate policy. I refer you to from him and he from me, instead of its being me
Missoffe. And de Gaulle. He's just finished telling from my colleagues and he from his. That's why
the Canadians that "they had a right to form elites some people today-the Chinese, let's say, or, at any
of their own." There's the whole government men- rate, some Chinese-want to change it. The hope
tality, right there. Notice: he was careful to choose of changing it isn't utopian if you're willing to reckon
his words. He didn't say, "You have a right to train not on a few but on a few hundred years. Civilizamore
teachers, more researchers." No, he said, "elites tions last a long time. How can we expect the new
of your own." The thing is, they already have an civilizations that began with Marxism just a hundred
elite. Quebec doesn't need to be free to have an and fifty years ago to be accomplished all at once?
elite of its own. It's going to take a thousand years, maybe two thou-
In eastern-bloc nations, it's much easier to get an sand.
education. But some kinds of training are still re- As a matter of fact, the world's last Cultural Revserved
for an elite. A thirty-year-old day-laborer olution is just two thousand years old. It was the
can't ever hope to make inovies. He'd have had to Christian revolution.
have been to film school. It's only just starting to finish up. It's produced
The work a day-laborer and an intellectual do nothing but reactionaries. The industries of image
are quantitatively but not qualitatively different. are still its most trusted mercenaries.
We've never been placed on an equal footing, which [Translated by D. C. D.]
JAMES ROY MACBEAN
Godard's Week-end,
or the Self Critical Cinema of Cruelty
Week-end, in more ways than one, equals history, offers a very selective view. Godard, in
"dead-end:" not for Godard, and not for the this film, concentrates almost exclusively on
cinema, but for a particular type of cinema- two of the most flamboyant aberrations of conthe
cinema of spectacle-which is pushed to its temporary life-the bourgeois materialist in his
limit. Future generations (if there are any) most aggravated fever of accumulation and
may even look back upon Week-end as the consumption; and his double, the antibourgeois,
terminal point of a particular phase in the de- antimaterialist drop-out from society, whose
velopment--or, more literally, the disintegra- only alternative to the horror of the bourgeoisie
tion of western civilization. The point seems is more horror still. "This is a helluva film," reclear:
"civilization," as it exists in Week-end, marks the male lead in Week-end, "the only
is doomed to devour itself. people you meet in it are sick!" The remark is
But Week-end, in spite of its searing in- crucial to the understanding of the film, for
sights and its sense of the general movement of clearly Week-end is the negative and destruc

STAIRWAY TO HEAVEN (LIVE DANGEROUSLY UNTIL DEATH) 2011



THE FRIEND:- POULOMI PAL AS PLAYING THE ROLE POLO. SHE IS CHARMING, GOOD LOOKING AND HELPFUL FRIEND OF TASU. SHE DOES'NT KNOW ANYTHING ABOUT TASU'S BOY FRIEND. SHE IS ALSO TASUS BEST FRIEND EVER SHE SUGGEST HER NOT TO BE IN RELATION WITH ANINDYA BUT WHO CARES......SHE IS 26 YRS OLD NOW ALSO FROM GOOD FAMILY."KNOCK KNOCK KNOCKING HEAVENS DOOR"......
TASU:- SAYANTI SEN WHO PLAYED THE TRENDY ROLE AS TASU. THE MAIN LEAD CHARACTER IN THE FILM WHO WAS IN RELATION WITH ANINDYA, SHE CLAIMED THAT SHE IS AN ABNORMAL GIRL IN THIS F@#$%#^ NORMAL WORLD. SHE HAD SO MANY DREAMS IN HER EYES, BUT SOME HOW IT BLEW UP....... 
AYAN BANERJEE:- THE ADDICTED GUY, A MURDERER PLAYED THE MAIN LEAD CHARACTER ANINDYA WHO BECAME A RICH FAMILY BUT SOME HOW HE WAS ALONE THEN HE MEET UP WITH TASU.HE ALSO INVOLVED IN CRIME.FOR HIS HOME PROBLEM HE LEAVE AND WENT TO OUTSIDE. HOW HE BECOME ADDICTED?FOR, YOU HAVE TO WATCH THIS MOVIE.....
DEBOSREE SAHA:- THIS FILM IS BASED ON HER STORY. DEBOSREE IS A QUITE, INNOCENT AND LOVING GIRL. SHE THINK DAT ALL ARE FAKE IN THIS WORLD SO SHE CANT TRUST ANYBODY WITHOUT HERSELF.http://www.youtube.com/watch?v=8rHmgsBBSEI

Friday 27 May 2011

OUR ROCKRULZ TEAM

TEAM :-SOURADEEPTA CHOWDHURY, SAUMITRA CHOWDHURY, SOHOM NANDI, SHILAJEET DAS MAHAPATRA, SAURAV CHATTERJEE, POULOMI PAL, JINIYA CHATTERJEE, ANJANA DEV, SUPARNA DEY, POOJA DUTTA, SOUMYASUBHRA SINHA, SOUPARNA CHOWDHURY, ORGHO DUTTA BOKSHI, SAYANTI SEN, CHI MUK, FROGGY DAERVEER, TOMKE LEOPKIS, SREYA BANERJEE, AYAN BANERJEE, SOMOK BANERJEE, DEBOSREE SAHA, ANINDYA GHOSH, RESHMI CHOWDHURY, DEBJEET BHATTACHARIYA.

ASSOCIATION PRODUCTION:- GHOMOSH AND PRATISHEDHAK

PRODUCTION DEPARTMENT:- RAJU BAKCHI, BUBUN HORE, ROCKY SETH AND RIJU NAG.
FINANCER :- SAUMITRA CHOWDHURY AND KAKOLI SINHA.