Semiotics and the Analysis of Film 0485121514


135 70 12MB

English Pages 280 Year 2000

Report DMCA / Copyright

DOWNLOAD PDF FILE

Recommend Papers

Semiotics and the Analysis of Film
 0485121514

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

First published in the United Kingdom 2000 by THE ATHLONE PRESS 1 Park Drive, London NWl 1 7SG

This English Translation © 2000 The Athlone Press Originally published as La Semiologie en Question © 1987 Les Editions· du Cerf Publishers, noce: The publishers wish to record their thanks to the French Ministry of Culrurc for a grant towai-ds the cost of translation. British Library Cataloguing-in-Publication Data A catalogue record for this book is available /rMn The British Library

ISBN O 485 11532 8 0 485 12151 4

HB PB

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, clecuonic, mechanical photocopying or otherwise, without prior permission in writing from the publisher.

--

.

KING A~~R_Eo1s COLLEGE '

v.,

I

NGHt-:F, c:·-·r:~ ! - I.

:02424.-72.:J_ I 7q I.4-:3ot 1

_

Mrf

Types;t.by Acom Bookwork, Salisbury, Wiltshire noted and _bound in Great Britain by Cambridge University Press

Contents

List of Illustrations

vi

Preface

ix

I

Preliminaries

11

Signs and signification

24

III

The direct sign or the ~neutral' image

30

IV

'Ibc image and perceived reality

37

V

The shot

59

VI

Iconic significations

91

Indical significations VII VIII-' The inferences of montage Concerning syntagmatics IX

102

Codes and codification X XI ., Images and speech

140

XII,. Narrative structures

170

1

109

131 151

XIII

Symbols and metaphors

185

XIV

Rhythm

207

xv

Sense and nonsense

224

XVI

Image, language and thought

248

Notes Bibliography

Reference Works Index

258

270 272 273

List of Illustrations 1. An artificially composed world: Great Expectations, by David Lean, 1946. 6 2. Expressionism reborn in 1943 in Carl Th. Dreyer's Day of Wrath. 7 3. Contrast cutting - the rich and the poor - jn David W. Griffith's The Speculators, 1909. 10/11 4. The image reorganises appearances: John Ford's The Grapes of Wrath, 1940. 36 5. Medium Jong shot: Stagecoach~ John Ford, 1939. Wide medium shot, The Last Laugh, F.W. Murnau, 1924. Close shot: Lilian Gish in The Birth of a Nation, D.W. Griffith, 1914. Facial close-up: Falconetti in The Passion of Joan of Arc, by Carl Dreyer, 1928. 6()-62 6. Depth-of-field and shot-sequence: Citizen Kane, by Orson Welles, 1941. 64 7. The snaking line of people and I van in the foreground in Ivan rhe Terrible, by S.M. Eisenstein, 1944. 65 8. Details in a downward tilting close-up: Carl Dreyer's The Passion of Joan of Arc, 1928. 69 9. An uexpressionisticn close-up in John Ford's The Long Vc,yage Home:,. 1940. 71 10. A "deep field,,: the grand entrance into Babylon in David W. Griffith's Intolerance> 1916. 77 11. Effects of depth-of-field in Eric Von Stroheim's Greed, 1923; diagonal composition in La Nuit de San Lorenzo. 79 0 12. The fourth wall" included in the frame by a "mirror effect" in John Ford's The Whole Town,s Talking, 1935, and in Joseph Losey>s The Servant, 1963. 81 13. A downward rilt expressing the desire for power much m?re effectively than the usual upward tilt: Citizen Kane, 1941. 100

XI

aspects of it have been particularly rewarding) than of explaining why it has generally fallen shon of the mark. It was not my intention to direct my remarks to a small coterie of linguistic or semiological specialists or exclusively to film critics, but as far as possible to the maximum nwnber of lovers of the cinema interested in film theory. At the same time, although the teaching of cinema studies is becoming more widespread, as much in secondary schools as in universities, it may be that many students will find themselves ill-prepared to read a book like this. This is why, following the suggestion of my editors and seeing the value of their argument, I have prefaced the main body of this book with some preliminary thoughts intended to put the newcomer in the picture. There is no need to exp]ain words which have passed into common usage such as shot, s~qu~nce, angle, reverse-at,gle and others. The emphasis is less on techniques than on their application and their consequent meaning., and, in partiC\1lar, certain ideas relating to "film language'\ to linguistic structures and semiology. The reader familiar with this terminology will be able to uskip'~ directly to the first chapter without missing anything. (n.b. Because this book is a digest of a series of articlcs1 it should come as no surprise to find cerl&in repetitions., which I have chosen to leave as they are, since occasionally the same subiects arc considered from different angles.)

List of Illust.ratiom

vii

14. Curious index: a bed-cage on the plains of the Far West. John Ford1 s The Wagonma.uer> 1950. 102 15. The cup sequence in The End of St. Petersburg, by Vsevolod Pudovkin., 1927. 104 16. Relationship of details in D.W. Griffith's Avenging Conscience, 1913. 110 17. The cream separator sequence in S.M. Eiscnstein's The General Lin~, 1929. 114 18. As though space were being seen on two separate planes., related within the same frame: David Lean,s Oliver Twist, 1948. 123 19. Matching along the camera axis and in action continuity: David Lean's Bn'ef Encounter, 1945. 128 20. Superimposed characters ufrom beyond the graven in Victor Sj0str5m~s The Phantom Carriage, 1920. 143 21. The steps sequence; rhythmic alternation of various shots, in S.M. Eisenstein's The Batileship Potemkin, 1925. 174/5 22. Symbolism of form: the street in Henrik Galeen ts The Student of Prague, 1926. 189 23. Staircase symbolism: Metropolis, by Fritz Lang, 1926, and Nibelungen, by Fritz Lang, 1924. 190/1 24. Symbolism of light and shade: Benjamin Christiansen's Witchcraft through ihe Ages, 1921; John Fordts The Long Voyage Home, 1940. 192/3 25. Surreal symbolism: Un Ch-i"en andal.ou, by Luis Buiiucl, 1928. 200 26. Slapstick out of control in a slapstick film from 1911: Jean Durand,s Le Rembrandt de la rue Lepic. The "broken mirror'' effect in a slapstick film from 1913: Kiki dcmestique, by Ovaro (Italy). 230 27. uBy associating the objects he uses with those suggested by his actionsn: Chaplin~s Payday, 1921. 232 28. The Floorwalke.r, Chaplin, 1916. 233 29. The Pawnshop, Chaplin, 1917. 235 30. The GeMral, Buster Keaton, 1926. 238

Preface F:rom the moment critics staned 10 theorise: about the cinema, they were concerned not just with a new means of expression, but with a kind of language capable of signifying ideas and feelings where the meaning depended on editing, the types of shot and their relationships, as much as on the objects represented in them. Yet they were using the term language in a purely metaphoriod sense. lt was during the 1950s, however, following some socio-psychologi• cal work and research at the Institut de Filmologic that the idea of language was considered in an objective and concrete fashion only to be refuted immediately by certain thcoreticians:11 notably Gilbert Cohen-Scat 1 • In responding to his objections, I believe I was one of the first to argue that the cinema is effectively a lat,gucwe. I wrote at the time: It will perhaps be argued that film is a form of writing rather than a language, since the image which shows things without naming them has no phonetic equivalent. Clearly~ the cinema could never be a language if language is meant in the sense of a means by which exchanges of conversation take place. Yet, if the imag~ arc used not as a simple photographic reproduction but as a means of expressing ideas through an association of logical and meaningful relationships, then clearly it is a language. A language in which the image fulfils the rolc..-s equally of verb and subject, noun and predicate through its symbol structure and potential qualities as a sign. A la,,guage through which the equivalent of the data of the perceptible world is provided~ not via more or less standard abstract forms but through the reproduction of concrete reality. Thus reality is not being "represented,~ by some graphic or symbolic substitute. It is presemed as an image and this image is what is used to signify. Trapped in a new dialectic for which it provides the actual form, it serves as an element of its own narration2 •

Semfotics and the Ana/y5is of Film writing the above, I had not previously In 1963, when 1 . washs. between the cinema and linguistics for d h relanon 1p cons1dere t e th 1 believed them to be separate from each · ple reason. at · Iogtst · Ch ristian · th e sim bowever, a young sermo other. The followll1g year, . ·t1 d '~Le cinema langu .. me a manuscnpt enn e , c ou 3 Metz gave Th . ma· language or language system] . I was 1an e?" [ e cme . . gag d b th 8 eriousness and depth of this work basing itself unpresse Y e • th · "fy' . • • m · order to study in detail e s1gn1 on bngu1sncs . • I 1ngthstructures of film and applying a much more precise temuno ogy an ~sual> • · · de...:v 'ng mostly from theatre and stage convcntton. fil m cr1t1c1sm u 1 • • m H owevcr, l' t seemed to me that Christian Metz had a tendency . his research to take linguistics as a model ra~er than a simple reference assuming that verbal and film exp:ess1o~s were comparable, capable at the very least of being associated m an all-encompassing structuralism. This considerable work - which it subsequently became seemed to me to be on the wrong track. However great my admiration, I had no hesitation in telling him so, his error being based~ it seemed to me, on irrelevant linguistic considerations. The fact remains that, from 1966 onwards, the theories of Christian Metz and his imitators enjoyed a success which cannot be adequately explained merely by the scientific nature of semiology. Also, in response to the unexpected proliferation of oourses in this discipline in universities recently devoted to cinema studies, I had intended to write a book entitled ''"Le mot et rimagen [The word and the image] (already announced in the press), in order to point out how and why this type of research was unproductive and to ~ve .a _warning to the many teachers who, better qualified in h~gu•st1cs than the cinema> were wandering blindly through the nunefield of structuralism as they applied it to motion pictures. How~er, I had already undertaken a history of the cinema and e~en with the five volumes so far published I could see no end in SJght. Which is. why I started to publish a series of articles in Cine1;U1lographe which scratch the surface of these questions Reworked in greater detail the · I • .ch . . i se artic es fonn the basis of the present book, h w 1 1s not intended to b h · •. tl•on of my posmon and e c:,c aust1vc, • . merely a simple adumbra- such as it is int dmy ~uti~n with regard to film semiology erprete by hngwsts. Now that the excitement abou . . be-come less a questi f . t semiology has died down> it has on puttmg up a defence against it (certain x

3

°

I

Preli"minaries No one could have imagined at the start of the cinema that it could become an art.; or an industry. Or., indeed, a kind of language, a .. visual event" loaded with meaning. Before we move on to the first theories or principles, we should first of all remind ourselves briefly how the cinema became the artform we arc pleased to recognise in it. Before 1908

As we all know., the first public showing of the uLumiere Cinematograph,, took place in the Indian room under the Grand Cafe on the 28th of December 1895. At that time., the cinema was just a machine for recording and reproducing movement - remarkable in itself, since it was the first time this bad been achieved. A spectacle equally novel in being projected on a screen instead of in a theatre. However, it is worthy of note that L~ Arrivie d 7 un irain tn gare de la Ciotat, La sortie du port, Le gouter de bibi arc already instinctively ucinematic". Not only are the scenes shot ufrom life"., but the train which first appears in the background thunders straight into the lens; the embarkation is seen from the top of the pier; and Mme Lumiere feeding her baby is framed from waist up in a shot similar to what would be called later on the "American shot,. [translator's note: a term coined by French critics to describe a mid-shot wide enough to include cowboys' bolsters in American W estems J. The view-point is that of an observer seeing things from the best angle and on an appropriate axis. From the word go, the narrow confines of stage represenration were broken. Space took the place of the stage. The only limits were those of the edges of the image - which appeared as a kind of window on the world. Indeed, the direct possibilities of the camera, and the advantages of being able to move it) were discovered very quickly. The first

Semiotics and the Analysis of Film 2

en of the theatre deferring to preestah no t m h • film -makers were . d but simple photograp ers using the tool a• lished aesthetic og.rn8 , . ,

amateur might use his box Brownie. . . . an}rom 1896 onwards, Alexander Pronuo, :f'ehx .Mesgwch anc

. D bli·er who made many long foreign trips to show of Francis ou , b k ~, · ,~ f . . c·mematograph:t brought ac views rom all foui the L um1erc p . h l"d . corners of the globe. It was then ~at___ -~-nu_o__ on o I ay 1n I ta.I~ had the idea of putting his c~!8 m _a_ gondola. The_ camera sct0ftxed,, (as it always was up un~il _!9_09).,_ but the movemeni =gondola allowecf~-to filin- wide panoramic view~,such ~ha1 The Grand Canal in Vance (1897) was. the _first ever trackmg. shot;;_ insptred by his discovery., Promio subsequently attachec ms·camera to several different moving platforms, such as a railwal wagon, the bridge of a transatlantic boat and the MontBlanc cable-

~f

car.

In his own way the illusionist ~rges Melies~ ~~inking that the Lumicrc brothers' camera gave hlDl the opportunity to record and broadcast the magic shows which until then he had put on at tht.: Robert-Houdin theatre, was the first t~conceive of the cine,m!l_ ~s a show, the first therefore to introduce the idea of "'mise-cn~scene~, [translator's note! directing or,-literally~ putting on.stag4 Whereas Lumiere, capturing ccnature in the rawn and preserving its authenticity, was content to record it literally:, Melies was the first: to create _an. original spectacle with the aid of the, __rjnema. Though he was filming scenes prerehear-sed with theatrical technique, he introduce~ new illusions with which camera technique :~~lemented the de?ciencies _·of the. stage~.- And these techn1ques, ough they may be seen as those of an illusionist rather than the eleme_nts of a language still to be developed were qj,scovered and CXJ!lo1ted ~Y Mclies with matchless control. , 'What makes his art fundam n h~- . . ment of th t . cn_ta Y t eatr1cal 1s less the arrangeth e se ~ or pauned backmgs and the quality of acting than .. unec~~n!Y.mtth~n~~:y__oftlu_p_oi_r1:t_o/-vie.w_: camera rimdly set before :a uOID'l eatncal spa ..._ _ ; -' · .ea ~~tacleoccur.r~g in !~~nt_s~ alwaya ll'l_J~~ ~ai;n~ __ wa,x, ~C: :t wherever it should· be -~t, ra~cl'.' .thlµl revealing it. or t}nding · ·- - -· · occunmg « As Andre Malraux observed: ..,

fr:~

.me

as long as the cinema was u ducing movement it sed rnercly as a means of reproJI was no more art than photographic repro-

Pr~liminanes

3

; duction. In a circumscribed space - generally an actual or ima. ginary stage - the actors played out a dramatic or comical scene . which the camera did no more than record. The birth of the cinema as a means of expression (rather than reproduction) coincides with the destruction of this circumscribed space); from , the moment the film-maker imagined his story divided into 'r, separate shots, be had in mind, not the photographic reproduc-1 tion. of a play, but the recording of a sequence of individual 1 moments, the movemenc of the camera to and from the characters on the screen (thereby making them bigger or smaller when required), but more particularly of replacing the theatrical stage with a "field of view'\ a space limited by the screen - an are~ where the actor comes in and goes out which the fi]mma~er c;hoosesJ instead of being a prisoner of it 1 • 1

In the meantime, Melies' "field of view" was that of the theauical space and his limits of the screen those of the wings of the stage. His various locations may be different, but one by one they become overlapped in the same container, presented as a series of theatrical ~'tableaux>' for a spectator sitting in his seat seeing events from the same point of view. Clearly, there is a narrative continuity, a chronology between each of these tableaux., but as these follow on from each other there is always an ~'interruption,,, just as there is between one act and another on the stage - except that:, because the scenes are joined end to end, the curtain or uscene changeu is replaced with an instant transition from one tableau co the next. Yet this instantaneousness is not applied" to the continuity of the movement, or the action - or indeed the time. TimeJ movement and a~tio_n arc. presented in small discQtltinuou.t bits,-like successive jumps. -Filmed from an unchanging point of view, these are-identical spaces (however different their content may be) following each other in identical time sequences. The discontinuity of the tableaux is emphasised by the absence of any mot1ement in their space and timej mo:re than by an absence of any linking mechanism. Th.€ absence of linking mechanism - in itself the consequence of an absence of space-time mobility - produces "repctitionsn or distortions when the intention (as sometimes happens) is to present - consistently from the same view-point:, in the same axis and at the same distance - different aspects of the same action.

Semiotics and tlu Analysis of Film 4 • , .. oi-unae a rravers l 1 impossi.ble:J Professor MaboulQAi Thus, tn I.A, v J . • f S . a, car navels across (a model of) the mounta1?s o w1tzerlanc 't starts down a very steep slope. Right at the botton Sudden lY l h . . kn kin Mabouloff misses a comer and eras es mto an mn, oc g dow its walls. End of the tableau. . . In the next tableau, we are inside ~e same mn. Natur~lly, non of the above has happened· yet. Dmers are ~at at their tablei merrily chatting away. Suddenly, to everyone s great alarm> th car smashes through the walls of the dining-room and knocks ove the serving-table. Later on, editing will enable the event to be seen from tw different points of view, but within the same unity of movemen, since the possibility of seeing events simultaneously is limited b the universality of the view-point. Here the discontinuity is mor

marked because the same time is divided up by having to be spl: into successive tableaux. As for the set design3 any props which do not actually "feature arc painted on canvas and depth is created by a forced pcrspectiv( The vanishing-points all relate to a single point of view, which i that of the camera placed just outside and in the centre of the set

the same as that of a spectator sitting in the front row of the stall! immediately behind the orchestra pit. This unity of place of observation is what Sadoul and I terme, ~c upo~t of view of someone sitting in the stalls,,, any change c v1cw-pomt requiring the observer r:o move relative to the observe object rather than the set to change in front of the stationary spec ta.tor.

The recent discovery of a collection of short films made befor !:06 ha~ ~ven rise to arguments about the origins of editing an, 1;,; van:uon, along the lines that what was discovered betwec: &,d·is an. 1903 ought to have put an end to film-making i1 continuous tablea » h• h . end of 1906 Th" . hux ' ~ ic m fact continued nearly to th of points of· . is_ ISM~~ Pierre Jenn claims to prove the ''variet. view 10 ebes, film n . one tableau t th s with the explanation that, fro11 0 next, the s_e ts reveal ddferent · same objects Nowc th aspects of th · c camera m M' Ii· , fi. , many others beside _ remain . . e es bns - as in Zecca s an1 not to confuse the '" ld ed static. And yet we must be carefll the camera moving ~or th changing in front of the cameran 2 witl 111 shot). e world (ch.:mge of point of view ant

Preliminaries

5

On the other hand, the description of these films as '"filmed theatre,, is completely wrong, since neither Melies nor Zecca, when they were shooting their little fairy-tales or comic sketches, were adapting from the theatre - especially since the theatre is the art of speech and their films were silent. Melies would nevet" have been able to produce on stage what he created with the camera. At the level of the image his films were already pure cinema> but his direction or mise-en-scene (and that of others) was as though for the stage. Instead of "filmed theatre" we should refer to it as stage directing, in contrast to film directing whose beginnings were marked in 1903 in a rather remarkable way by Edwin Porter's The Great Traln Robbery. It is no exaggeration to say that, if the first directors were limited by the confines of the painted set.:, it is because they were ignorant of the capabilities of the cinema. Even the earliest films of some of them - notably the Englishman George Albert Smith and the American Edwin Porter - show evidence of the art of editing, indeed almost of the seventh art. Smith~s 1902 film Mary :Jane,s Mishap contains some very precise action cuts outstanding for its

time. As unlikely as this may seem:, between 190 l and 1902, when Edwin Porter was filming his documentaries - his upanoramic viewsn - he included panning shots over the natural settings and even the odd reverse-angle. However., when he came to shoot his ufictions,,> like Uncle Tom's Cabin, he returned to the conventions of the stage with the painted back-cloth and trompe-roeil sets. This tug-of-war between reality and the stage is characteristic of the earliest days of the cinema. The reason for it was not technical but, curiously, psychological. It was not the film-makers who insisted on remaining on the theatrical treadmill, but the public who imposed it. Ul!_cd tQ _st3g~__ repi::c;.~er1tation, t,l!_e _public found it. hard to understand - or-acc~pt --that a fictionaLstory Qr~ could be represented in any_Qt;her_w_~y. __Ih_~_ directors we~-tlierefore forced back t~. t_ttableaux'~•. Not only were_ the _actor 1_s movements not continued from one tableau to the next, but the

drama

uni_ti,-of yiew~point-me.ant-.that -the set· ~4- t~-_-1'i~~~Jl9_t_. Feaci:..'?.n, ~ involving inevjtatle.rc:p~titions and-constant.overla_p~...!.... The strangesto{ all that it was perfectly acceptable to have a variety of images - or shots as they soon became called - when it came to documentary uviewsn - doubtless because it was a "realn world

~as

. •,.., a1,u/ the Ana/vsis of Film S em,on.... w

6

ed where it was possible to move about. It then being rep~sent al 10 create an impression of reality by having b me qune natur ec~ d"fti 1 poinrs of view recorded by .the camera. yet various. I eren . too k a 1ong time . to "drama" the pubhc to accept' when 1t came ·entation of 'a fktmnal . d Jd worl cou conform to the that t h e repres . h. h · · · '"tru"'" reality Something w 1c certain cr1t1cs even data of .. · . • da'-'S find it hard to understand, 1f we take as an example now a J • • th . p . if Alain Masson's tbesis on the sub1ect m e rev1ew u.su, (April 1985-July 1986). He writes: Audiences at the beginning of the century were no more stupid than they arc today.[ ...] Suddenly confronted by horses with invisible legs, the audience of The Advemures of Dollie {Griffith, 1908) must have thought that, rather than legless monsters> wb~t they were seeing were horses whose legs had not been filmed. Obviously they understood they were seeing horses;, but they did

not understa_nd •why the horses were being shown like that, nor why the acuon had been divided up into successive fragments .

.An artifkially composed

Worlci: Grun Ex . pectat,ons, by David Lean, 1946.

Preliminarie$

7

ExprC'!lsionism, rebom in 1943 in Carl Dreyer's Day of \r-"rarh.

Particularly since in 1908 it was exactly as it had been between 1900 and 1908. Ye~. it_must not be forgotten that the fact of telling a story with moving pictures did not exist bcfor.e_tbe cinema. _...-;:;· Therefore known values had to be employed as reference, with the resulting application of theatrical forms to cinematic representations, since any new invention always creates scandal - and shock - upsetting habits, conventions and traditions. 1908-1918

While editing was being discovered in the U.S.A., Europe.an cinema, freed from the shackles of theatrical representation, developed under the heavy influence of painting. Because the theatre is verbal and the cinema silent, inspiration would be drawn from painting and not from the theatre. For the images to be meaningful, all that was needed was to ensure _that they express everything the pa.Tnting might express, any movement providing an added

value.-composing with images - that is, composing the image - with its internal structures, exploiting its surfaces, its lines and volumes,

-

Semiotics and the Analysis of Film

8

' th the. sym.bolism of shapes and objects th' light and sbade, Wl . d , 1· . ,..tion of the exper1mcntatton con ucted by n~ .... ~,s1· d th was e ire.. b . .al ...... film-makers heavily influenced_ Y p1ctor1 • . exp:essionism

Composing the image in fact consisted of composing artificially thi world being presented. . Of course, it would hav~ been ~erfectly possible ~o organi.t, reality without recomposing it. The simple effect o! placing a char acter or object relative to another, of structunng the conten relative to a frame would have emphasised. the objects shown b, giving them a temporary meanin~- It would hav~ been possible,; consider the world from a particular angle without neces9aril• altering it. However, this would have required the use of lenses not devel oped at the time, providing the opportunity to play with perspcc tive, introducing a greater variety of shot, a dependence on eclitin1 - considerations which were practically unknown in Europe. · An interpretation or stylisation of the world could only occur i it were recreated using the techniques of design. The camer: recorded a pre-prepared, aesthetically developed universe full o nuaning. A meaningl moreover, acquired from values outside thos of the cinema. Stage representation had been abandoned and replaced by th application, in a quite different but nonetheless definite sense, c the principles of stage-craft. Giv:en _the intentions of the Danes, these aesthetic principle were Justified. They allowed the surface values of direct reality t be transcended~ to display a quasi-dream world revealing ufror. within•• 1. d , • ts eepest meaning. A symbolic world whose meanin~ asswmng an exc1us1vc · · . . importance, provided value for simph mteni:1onally melodramatic stories but through which the grossl JlUlgnificd "fant st·-Am · • . a •'l.c:ll refleCtion of genuine anxieties real socu blems ,or undc:rly' pro th ' . . ture.. ing tru s could appear as though in m1ma DcveJopcd during the . d . ples gav . pcrio 1912-1915, these aesthetic princ1 one o. f g~ttostrGermdanf Expressionism (1918-1925)1 which ws •.... an s o the cin • More symbolic than d' . cmauc art. sionism nevertheless h ~scursave, the signifying values of expre~ l>S!choanalysis than w~th more to _do . with . the expectations . c editing seemed to be th commun1cat1on structures for wh1c e ne-cessary basic principle.

th:

Preliminaries

9

Editing In the primary sense of the word, editing consists of joining shots (or takes) end to end in the order required by the logic of the drama {story or narration). In this sense] editing is as old as cinema itself. It can also be used in the sense of joining together a series of cuts made to abbreviate a movement, change an action or link various sequences. But this sort of editing, practised mainly by Melies at the beginning of the cinema in order to extract the maximum dynamism from within his "tableaux 0 1 derives from the area of trick effects of which he was the past master, rather than the foundation of ngnifying relationships. Meant in this sense - the only appropriate aesthetic meaning possible - editing consists in the joining together of two or more shots such that their r.elationship determin.es a meaning belonging to neirher of the.m separately. But let us first examine how this came about. As more and more importance was placed on films and a greater variety of shots and framings developed, it became clear that, while building the drama, the simple linkage of shots created arbitrary relationships between the represented objects by giving them an allusive or symbolic meaning. Among the directors who became involved in the research and experimentation developed from this discovery, the most influential was the American David Wark Griffith, whose universally appreciated work is analysed in every historical study of the cinema. It was he specifically who used editing to involve the spatial, rhythmic and discursive forms of the narrative. 1. By showing events, actions., characters., sometimes in doseup, sometimes from a distance, face on or in profile, from the back or in three-quarter profile., from below or above, film placed the audience among the characters of the drama, within the space of the drama. By seeing objects from different, interrelated points of view, the audience was observing those objects as though actually moving around them. In this way the impression of spatial reality was established. 2. By creating - through the intensity of ~e drama, ~rou~ the various different points of view and frammg - relationships of duration between the shots, editing imposed a rhythm specific to ea.eh film. Moreover, by manipulating time and ~e . interrelated spaces, it allowed for all kinds of ellipses and abbrevaanons.

10

Semiotics and the Analysis of Film

3 The association of shots implied, suggested an idea which . .l · . r·. cerminology, might be called the "connoted signified'! 10 mgm:s 1c . b . or ·•r,·o,,, •~he meaning of d1e obJects s11own e1ng referred to as cmmo r,. the dem,ted meaning. . Long before semiology ever entered the arena, tt was said simpl. that "what matters is not the images but what exists betwecm th} images'' (Abel Gance), or else that "_th~ im~ges signify k--s: through what they show - whatever the :s1gn1ficat1on of the objects reprcsent1..-d - than by their ordering and less by that ordering than by their rhythmic and semantic relationships'' (Jean Epstein). Rhythm a,id sa'guificatfrm

Refined~ developed and extended by Griffith in The Birth of a Narion, the first master-piece of the worldts cinema (1914), then in Inwltrat1u (1916) - films ,vhich were not known in Europe until after the First World \Var - visual rhythm burst onto the screen with Abel Gancc's La Roue in 1922.

Contr,lst CU.tting - lhe rich

t~U:· David W. Griffith's Th-s speadarorj~

and the poo

Preliminaries

11

Replacing painting,, music then became the only valid reference, so much so that for several years all people could talk about was uvisual musict): HWould that film could be for the eye what music is for the ear'\ wrote one critic at the time, forgetting that these two organs have quite distinct sensory ranges. Which did not prevent a certain "avant-garde" from recognising the eventual possibilities of an entirely rhythmic cinema, dubbed, rightly or wrongly, "'pure cinema,, ... Over the same period, having also discovered rhythmic structures and cadences via Griffith's films, Soviet film-makers undertook research based on rhythm but, more particularly, on relationships for which they leaned heavily on dialectics and Marxism. The most outstanding discovery - purely accidental derives from Kuleshov~s experiments into acting: the arbitrary relationships between a subject "observing" and various different "observedn characters (supposed observing and observed) induced in the audience's mind various expected meanings. In other words, the audience bestowed on the "observeru ideas or feelings which it might or should have expressed about the observed facts consistent with an established socio-cultural logic. The quasi-linguistic nature of the cinema was thereby partly established.

12

Semiotics and the Analysis of Film

Firlt theories A from any aesthetic considerations3 more literary th~entific, such as those. of Canudo., Delluc; L,Herbic the shrewd analyses of a previously unknown American Psyi . t Hugo Munsterberg., who associated film understandinl : : formal strUctures of Gestalt psychology, i.e. direct perc understanding, the first genuinely scientific studies (at the I< regular science) took shape with Kuleshov!.s experii:nents. Jean Epstein was the first ~o emphasise the unponanc significance of the close-up, particularly the close-up of objec meaning and conditions of rhythmic editing; the symboli objects3 signifying values still generally considered and desii - following the lead of Delluc and Canudo - by the vagut phorogenics: 1 describe as photogenic any aspect of objects3 beings 01 which increases its moral quality th~ough its cinematic duction. [ ... ]The photogenic aspect of an object is the ef its variations in space and time 3 • Though not exactly a code, Eisenstein had in mind a k ci~-dialectic in which editing played a vital part. He writes:

The shot is by no means an element of montage [trans note: the French term montage covers the general princ editing, and Eiscnstein,s more specific application]. The a montage cell. Just as cells in their division form a pbeno1 of anoth~r or~er, the organism or embryo, so, on the oth 0 ~ the dialectical leap from the shot, there is montage. [I k~] loudly defends an understanding of montage as a linJ pieces. Into a chain. Again "bricks,,. Bricks arranged in to expound an idea. I confronted him my viewpo ~ontagc ~s a collisitm. A view that from the collision actors arues a concept.

with

1

From my point of . linka . F.ie-.. t • VlCW ge is merely a possible specit, ..... u., em goes on: If montage 18 ·

b to e compared with anything., then a phal

Preliminari~s

13

:montage pieces, of shots should be compared to th . f . . al b . e scncs o explosions man intern corn ustton engine, driving fo d . . c • .1 1 rwar its automob11e or tractor! .1.or> sum ar y, the dynamics of m ontage . . ~ d 4 serves impulses r1vmg .1orward the total fitm . For Eisenstein, then, meaning depends on the conflict between associated elements rather than an idea initiated by a narrative sequence. However, the signified becomes a signifier within a continuity whi~, like any other, is a developmenr (ideological, factual, dramattc, etc.), b~t a ~evelopment which, instead of being continuous, based on the linearity of the narrative, is discontinuous based on the collision of almost simulta,ieous elements creatin~ ideas in the same way as a harmonic development. Yet apan from re.al simultaneity as in "depth of field" (which developed after Eisenstein's cinema), this simultaneity is inevitably altered in continuity, signified rather than represented. What follow$ - from the effect of the non-linearity of the development - is a broken, fragmented continuity, i.e. one in which fragmentation is relied on to produce the meaning. In other words, in Pudovkin, the narrative continuity lends itself to montage and controls its effects, whereas in Eisenstein it is montage which constructs, controls a continuity which is both descriptive and discursive. Believing in his turn that "montage only becom~ constrUctive when it allows to be revealed what the images themselves do not show.,, Bela Balazs nevertheless endorsed the Hintellectual montage" advocated by Eisenstein, which "resulted in hieroglyphs or ideograms needing to be decoded like a puzzlen. He laid great store on the powers of the close-up and especially on the fact that framing effected a change to the relationship between the audience and the represented world. He wrote: The image is an interpretation, not a carbon copy. Nothing is more subjective that a lens5 • • • Rudolf Arnheim, a disciple of Kohler and Wertheimer {the

founders of Gestalt and the "psychology of form,. in 1912-14) _was the first theoretician of the cinema to establish the connections between film perception and gestalt structures. More convention~ ally than Bela Balazs he underlined the differences brought out by film between reality ~d its image. However, his ideas, which were essentially based· on the conditions of the silent cinema, eventually caused him to neglect and reject the talkies. As he wrote:

14

Semiotics and ihe Analysis of Film A •xwre and not a synthesis of fundamental llleth ·ft'ki~ will ultimately fail. It will then return the silent c1neni ~c:ls1 IA.I e • 6 its former perfect punty .

An assertion yet to be proved and one which doubtless never be proved. . . . For a whole decade, cr1t1cs avoided. any theorising 0 assumption that the first talkies pre~luded it. It was not ~nti~i the- war that theoretical essays with any degree of consisti started 10 reappear, starting with Gilbert Cohen-Scat~s Es.sa· les pn'nciples d'une philosaphie du cinema. Making a distin~ between the filmic (aesthetic expression) and the cinematic (tee cat process), in particular studying the forms of "film discou1 Cohen-Seat highlighted what was for him the basic differ between the verbal code and visual expression. Packed with ii csting insights~ his work is unfortunately rather neglecte, modem critics1 as is Alben Laffay's Logique du cinema (196 collection of articles on reality 1 fantasy and the great themes o silver screen) seen from a rather existentialist angle7 • Andre Bazin on the other hand enjoys a universal reputa Doubtless because he was the greatest critic of his generatio11 because, in that sense, his writings are of the greatest interest at the level of pure theory, his ideas are questionable. WI., strange is that the critics whose ideology is the antithes Bazin~s are those who praised him most vociferously. Bazin adopts the exact opposite position to everything v was advancedJ proposed and demonstrated before him and, s after him, He accepts montage in the sense of means of cons tion but rejects the ,.,arbitrary association of shotsu as an inev: "dist~nion of reality•\ Which is true. But art consists in 1 f?nmng the world, not imitating it. And it is upon this repr< tton - or reproductive faculty - in the cinema which : °?ncentrates in the name of transcendental realism which, c1ated ·th h · thcological convictions, brings him nearer the • wi . . 15 tential spintualistn of Gabriel Marcel or Emmanuel Mouniex n:ie bergsonian idealism with which he has sometimes been ciated. With the rational tba1 • d s imbuing it with e , the. camera captures reality an dedu tha any Ptetnous ideological or cultural values, ccs t the 0 titm image, · 1 the objective reprod uct'
reliminari~s

15

objects., is the surest means of knowing true reality reality bee-, ,~b ~ - ,~ . . . , ,ore know1c d gc, . e,ore. pcrceptmn ·, rcabty "in-nsclP,, .. For Bazin, the basis of c1nemat1c expression is and could only be this "objective" reproduction of the world and its objects•.. Bazin, ~dmining that the cinema is a language, refused to accept that the 1n:age may be reduced to a linguistic sign, i.e. through purely arbitrary symbols, t~ be merely the vehicle for concepts detached from concrete reahty, whereas Christian Metz laid the foundations for Hfilm semiology" by applying to the cinema the great building blocks of structural linguistics. But what is semiology? What do these terms used in linguistics mean? After this short description of the steps by which film developed into a code, it is obviously time to explain this terminology for the benefit of the newcomer.

Word and image Signification derives from psychology in that it associates an object, being or idea with a sign capable of evoking them. The notion of sign presupposes two different interpretations which often lead to confusion = natural signs and artificial signs. The latter arc divided still funher into repruentalion signs, which reproduce the natural qualities of things., and conventional or arbitrary signsJ which belong to a code. Semiology (from the Greek sima: sign) is the science devoted to the study of all sign systems - gestures, signals, symbolsJ etc. All signs include two values s a signifier (the sign itself) and a signifiecf' {whatever this relates to). Natural signs arc based on the relationship between a phenomenon and a common meaning generally justified by the qualities or inferences of that phenomenon. Thus dark clouds are the sign of rain, smoke the sign of fire,. near or far. The image reproducing a given reality - landscape, object or human-being - is the direct sign of what it shows, or a formal sign in that it is identical in form to what it reproduces. It may have a specific quality, but generally it signifies no more than is signified by what is represented. In these circumstances it is said to be ucoextensive with the signifiedu. By contrast the linguisiic sign is not a duplicare. It is fundam~tally conventional,. its own nature having nothing naturally in

16

Semiotics and the Analysis of Film

4 ~mmon with what it describes. or signifies: ' the word dog t biten as William James put it. no ' . . 1· . . According to Saussure' s d ~:fin1t1on, .th 9 e 1ndgu1st1c sign _ Of , _ is an arbitrary and unmotivated unit . A ouble-headed UIJ hich the signifier is the oral statement and the signified a co :lating back to the concrete reference (the object being descr~ There is however an foev~table asso~iation ~etween the sign the signified (the word chatr and the idea chair). It is the case modem linguistics has tended to reject Saussures bipolarii favour of seeing the word as a global signifier rc1ating both concrete (the object) and/or abstract (the concept) signified. context making the distinction obvious during reading. · Though the word - or moneme - is indivisible as a unit of v1 signification, it is still made up of phonemes, sound units devo meaning, which may be loosely compared with syllables (e> for phonological analysis). One of the essential qualitie language therefore is that it should have a double articulatic connecting the verbal units and connecting the phonetic co1 nents of the word inside it. Semantics (from the Greek simaino: to signify) is the stu( linguistic signification, i.e. the meaning of words, their pract1 unchangeable lexical meaning and especially the associative re/a ships between the signifier and signified. Signification at the of_ semantics is created by speech. The word dog may conju1 fa1thfulness, the word on'ent the opulence of Babylon. As . Greimas points out~ "the minimal structure of any significati de~ed by the presentation of two terms and by the relatior unmng them~•. It is easy to see that this definition is applicat montage. · 3 group o f words with . a single . • p·· b The syntagma 15 meaning . etwec~ th ~ word and the sentence. A short sentence such as· u walking 1s a synta A . · , sev: 1 gma. shghtly longer sentence may JO era syntagm.as. In the cinema where a shot is the equivale ::~r ~~e s~tences, the syntagma is generally understood Theofrsa ots mv~Iving a global signification, or structure. g«icntati · d · study of film h" on 1n. uced by semiology is of benefit JI indivisible se:n •eh lllltil then had been divided into very s However) it is of~:s - shots - and into parts - or seqw stylistic form which e. only at the level of analysis. Synecdochj (Ill general) presents a part for the whole

Preliminan'es

17

sail for a ship. In the cinema, it is the equivalent of a detai1cd close-up. The image When we re~cr to ~he.Ji.Im image, we mean a moving image and therefo['C a senes of ind1v1dual frames constituting a shot. Consequently., a shot or a film image means exactly the same thing, with this quaJification that a shot may have a particular description (medium shot, close-up., long shot., etc.), whereas the image has a more general, more flexible meaning. There may be a shot of a particular kind, but always a mMJing imags. Not only is the image a complex signifier., it is always individualised, personalised., differentiated. The image is always of tha! speci.fic dog, seen in thar specific place, from that spe~ifi.c angle, never a dog in general. Any other image would show it in a different place or from a different angle, yet always within a space where certain elements would be included in the frame~ the cinema contains no separate unit> no unitary, isolated signifier other than the close-up. Christian Metz pointed out quite rightly that it is not possible to compare the form/content relationship with the signifier/signified relationship, and associate or identify form with signifier, content with signified iO: In the inherent proposition that there is a kind of privileged kinship between, on the one hand, the effects of the signifier and the effects of the form and] on the other, the effects of the sig~ nified and the effects of the su bstancc, there is the potential notion that the signifier has a form - or is a form? - where the signified could not have one,; also that the signified has a substance - or is a substance? - where the signifier could not have one. It is possible in fact to define the fo~ and subs~~ce of si~ificr and signified by reducing them to the slfflple definitions bel~w· Por-m oif the si-i1;er: a series of perceptual patterns specific to .a ~ .. !!• · t' f their film; global strucrure of image and sound ; organJsa 1on ° -signifying relationships: ,, • e as the Substancs of the signifier; "content-matter of the miag

Semiotics and the Analysis of Film

18

representation of concrete values; sound-track (spcechJ rnusic). . u. . 1 the signified: thcmanc structure; structure of ..L ~orm o, fi 1. b' . u tionships of ideas or ee mgs; corn man.on of the sc elements of the film: . . Substance of the s,gnijied: social content of the theme of t1 all the problems raised b~ the film, except fo~ the spccifi ich the particular film gives these problems., 1.e.: wh I . . Form of ,h~ content: sty e, expression, way m which the 1 (drama.tic acdon, events described) is expressed, signified, . lised using the more or 1ess u spec:1'fic:,, t ech mques w h ich guish this story from what it might have been had it b• subject of a novel or a play. To stud)' rhe form of a film is in fact to study the whole film by assuming as relevant an examination of its organ its strUcture: it is a structural analysis of the film, the st

being as mu.eh a structure of images and sounds (form signifier) as one of feelings and ideas (form of the signifie4 On the other hand, when the "contentu of a film ii studied, most frequendy it is a study essentially conccrm the substance of the content alone: a more or less sh listing or description of the human or social problems ra the film, as well as their intrinsic imponancc, with no study made of the specific form which the film under coll tion gives to the problems. A genuine study of the contE film would in fact be a study of the form of its content: the film is not what is being considered~ but rather variot si:nera1 ~roblcms for which the film is a point of dei: ~•thout. ns re.a.I content being in any way confused witl !mcc this would be contained in the transformation coc unposed on these contents 11 •

Film semiology: its rea.rons For Ferdinand d S one area of semi e aussur~., the study of linguistics was most lingu. t ology· Now, in the rearguard of the stru.cn su~headin; 7::c .1,0 • regard sc~ology as nothing more of language As guaStics, the notion of the sign being the 1 · a consequence of which it became more

;f

Preliminarks

19

impossible to refer to semiology - even outside the realm of linguistics - without involving the structures of language. }laving foJlowed the work undc:naken by Christian Metz after his basic study, I wrote in 1965 in the second volume of my

A~sthetics: It is impossible for there to be a film grammar, for the very good reason that all grammars are based on fixed values, on the unity and conventionality of :signs. They can only govern modalities relating to these basic fixed values. Any attempt in this direction has ended in failure and indeed anyone who claims he can submit the cinema to the laws of grammar has a poor understanding of the expressive and semiotic conditions of motion pictures. Since it docs not operate with previously established signs:io the cinema docs not presuppose any a priori grammatical rules. Even syntactical rules arc unrdiable. They may be applied to a particular aesthetic or stylistic principle, but never to the language of film as a whole. ( .•. ] This is why I have grave reservations about the possible syntax for the cinema co which Christian Metz apparently aspires> which, he says, "has yet co be drafted, entirely on syntactical rather than morphological foundations... Since the absence of genuine signs cancels out the need for morphology and, if all syntax is syntagmatic (to use Saussure's terminology), it does not seem possible co govern, with any degree of accuracy, structures which are self-governing through their content and motivated solely by the (infinitely variable) meaning they give to the objects they express 12•

Which was a response in some way to what I had already indicated two years previously, specificaUy that: These same ideas can [thus] be signified in many different ways; but none of them can be signified each time by the same images. There is no link, no causal quality between signifier and sigm"fi.M otherwise the former very quickly becomes an abstract sign devoid of the living qualities indispensable to it. It is a fact that from 1966 onwards - particularly after the events of 1968 - the theories of Christian Metz and his increasingly

20

Semiotics and the Analysis of Film

Garroni , p·•er numerouS disciples (Umberto . . Eco, . 1Emilio 1y; Roger Odin ·ru· Gianfranco Bcttet1ru, m ta Pasol1 , • . . Ch , J~ 0 omm1que ateaul Michel "'Aumon t ~ Michel Marie, SA d La , '--Ol . "'"d others in the U. . . an tin America) enJo~ . Franee, ...... unexpected success. . . All this happened at about the tune cmema. studies and unication courses started to comm th . . be taught in universiu' ( 196s, those historians and eoret1c1ans capable of running c courses at university level could be counted on the fingers ( band (at the outside). For strictly administrative reasons i absence of .. doctors of cinema", the majority of the tcachini came from the literary disciplines where they had written doctoral theses. And since those who were genuine]y interes the cinema confined their attention to pure stylistic or cultural analyses, the arrival of semiology seemed like mann~ heaven. As Roman Jakobson put it: "The cinema seems to 1 especially important system of signs. I cannot imagini semiotic study not involved with the cinema ... n The reaction was that any theory or consideration outsic remit of this discipline found itself rejected as being worthle: unproductive, and anything before it as old hat or uninterc Evc::n now, in cinema studies based entirely on semiotics, · per cent of books quoted have a purely linguistic origin ... Moreover, the theoreticians of audiovisual communi thought ii possible., using moving images - film or televi: through an objectivity and precision beyond language - to c m~sages and relationships of fact and yet avoid persona] int tauo°:, or artistic judgment. With the proviso that I ~eanmgs could be given to the ordering and relations} images, in accordance with certain rules. Which is self-evi wron~ but provided them with the ephemeral vision of a pc SOlution through Metz' syntagmatic structures. th~l~rly C~istiw_i Metz never had the intention of transfo . ctnema into hnguistics. Without ignoring the gulf bt cinema and Jan th h to h guage, e purpose of his study was to see 0 exte?t ,it was - or should be - possible to apply. of -codifying 5Ystem which would be for film the equi e~ammar for language. ' , Even though he 881"d rely on 1· . . from the outset: '' Semiology can anc ingu1stics b · • ,, ' ut it must not be confused with Jt

cin:n:ta

Preliminariss

21

mistake - in my view - and that of all chose semiologists who studied the question (Garroni, Eco, Pasolini and others) was to depart from linguistic models and look for analogies in different functions Tather than start from scratch with an analysis of the smalJest uniu of the film message and examine how they worked, ignoring any notion of code, grammar or syntax. Christian Metz' work has enabled empirical methods to be replaced by a rational analysis, and has brought out the processes by which ( or with which) a film creates meaning, has revealed how it functions. Yet though semiology is capabJe of saying how someihing signifies, it has no way of saying why it does so. In linguistics the question never arises, but in contrast with words, images are never cre.ated to signify. They contain a curious function which turns film into a sort of discourse: whose structures 1 superficially similar to linguistic structures, arc totally resistant to any rules of language. Effective when it comes to analysing or Hdeeonsttucting"' a film, semiology is of no use when it comes to drawing up laws, codifications and rules applicable to all films. Its systematisations are all after the fact. Metz. muse have accepted this when he declared at the colloquy on cinematic research in February 1977: "I believe that, as a school of thought, semiology has had its day. It may, and even shouldj retire [ ...]." My work was deprecated by dogmatists who must have found it difficult to accept, but a certain panic set in amongst the ranks of the high-priests blinkered by their ucst.ablished" convictions. After the doubts, the first attacks were mounted. Twenty years after the event came the revelation that the visual and verbal had no point of comparison and that it had been a fundamental error to turn linguistics into the model for all semiology. As G. Deledalle wrote: It is understood that the notion of sign limited to linguistic sign

may be appropriate for textual semiology but is unsuitable, or at least is not the most suitable for a description of non-linguistic signs., such as a film or any other system based on iconography14 _ Following an article by the critic Raymond Durgnat published

in May 1980 in the review CiniaJle and entitled HThe death of

22

Semiotics and the· .Analysis of Film

Semiology,,, the English film-maker Lindsay Anderson The Guardian. in March 1981: The comparison of film with language and the at examine and interpret a work of art using methods .1 from linguistic analysis, are associated with a tendency ing as so often in France., which tries to examine the w from a scientific and logical point of view [... ], by replace a simple affinity with a Rule, intuition with a I The snucturalist movement in film criticism is ban: because it tries to replace interpretation> research into and human behaviour, with stylistic analysis. Then the philosopher Gilles Deleme, before publii remarkable work L'lmage.-Mouvement, said this in an inte: The attempts to apply linguistics to the cinema are < [•..]. Any reference to a linguistic model always proving that the cinema is something else and that, if i· guage, it is analogical or one of modulation. Which lel! conclusion that any reference to the linguistic model is to be avoided 15 • However, whereas Christian Merz abandoned. structun ogy to pursue a more productive study into psychoanalysi of his acolytes believed that if the taxinomic models pr where there is no such thing as image-verb., image-subject, image-adjective, where the briefest shot incorporates all these designations, it is not possible for a signification to be distri butcd by the structure. Which is another way of saying that the shot has nothing in common with the word. It is a unit of construction, but one which includes a whole series of relationships; a signifying unit, not a unit of signification; Thus, the shot is not the equivalent of either a word or a phrase; rather it is the equivalent of a whole series of phrases. Several phrases are needed to describe even the simplest close-up; and a great number arc needed for a description of a more complicated wide-ang1e4 • Thus if a single shot may be regarded as a statement, the linking of shots may be compared with the linking of phrases. Now) no .· grammar exists to govern the order of this linking - apan from Chomsky,s generative grammar. Yet even this is burdened with paradoxes. To be able to establish any rules in ordering units, the meaning of the units must first be known. The semantic system

semioti1;s and the Analysi·s of Film

28

which, according to Cbomsky, m~t be based on syntactical etennining the syntacncal rules on which it . tul end s u P d 1s Pre cated. · ·ct the linking of phrases 1s · entire · Iy d etermined by th I n f:a . th e lo at. .tt wou ld be more n"'t of thc narrative. Which means . .. urat compare film scructures ~1th narrative ~trucrures rather than thi No film unit stands on its own and any chan Of lanm•"'ge. e. "f . . h g~ absorbed into a separate s1gm ymg unit, as t ough passing f , one sentence to another instead of any grammatical transforniat~ Sigtiifying values

Yei, in many instances (notably direct antitheses), shots can like words because of their impact. Now., the meaning of word invariable whatever their semantic depth whereas what is denCJ by film is always someihing else. Words are neutral as lexical si in being identical to each other, whereas images, in being diffc1 from one another, each have a specific value, a personal qual Words have a distinctive quality only at the phonological phonetic - level. Pronunciation~ intonation, emphasis all co1 value and meaning on the "verbal material'\ but at the lcve speech rather than the written sign which alone may be contra1 with images in that its point of reference is without oral or equ lent intermediary. The concrete value of the images gives them a different mem each time. They never repeat themselves. If we were to incl only the 500,000 feature films produced in the world since 1' there would be over 60 million million images. And none of t1 bear~ the remotest resemblance to any other. Each has its ' spcetfic content, meaning and signification. Which brings us c :ely_ t~ the Fido Fido Theory proposed by G. Ryle to poke f1.1 gutSt1~: th e word Fido corresponds to the dog Fido· A, Schaff wrncs:

individual object tbe: or c1tmg us pot • 1 . d whol entia to generalise, we must d1sregar c system of abst th. . . • th co of hi ract inking established dunng e Slory, and th d tes_po0 ...1: • en remember an infinite number of wor 5 "-IDg with an uni· · d be'1 1

filf wfi~ ~equ~re the word to describe the

cna~.

n:uted number of objects an

P

Signs and Signification

29

There is no grammatical rule which could pssibly govern or dominate such a wide range of possibilities. Yet this is almost what the cinema does, without inhibiting the potential for making generalisations from the individual or for '~creating abstractions'" from the concrete. Having both the content of a single phrase (or several phrases) and the impact and essence of a word, acting either as a complete statement or as a sign, the film image avoids as I shall try to prove - any causative c;odification. Every image is endowed with a semantic content whose significations depend on a whole network of circumstances which are analysable only by giving each of them a temporary value, by dividing arbitrarily the signifying functions into several different levels~ in the same way as the tripartition of linguistic signs 6 • A. In essence, the ima~ maintains a reciprocal relationship with what it shows. It is not a signal~ which would imply a whole series of conventions) but a duplicateJ a consistent reproduction called gestalt-sign by psychologists, 11arural sign by linguists and direct SJgn by semiologists. With no other signification than what it presents, it is the degree zero of film expression. B. The first level depends on the way things arc shown., the angle., framing, spatial organisation of the field of view, in short the internal structures of the image. This signifkation is termed icm-i-ic or imaging. C. The second level belongs to the formal association of shots in the continuity, indications provided by the denoted objecu relative 10 the film -without these being symbolic or metaphorical, merely r8/ractive. D. Most importantly or most charaeteristically, the third level is none other than effuz-momage whose connotative meaning effectively depends on the association of two or more shots. Yet it can also depend on events interrelated within the depth of field. Symbolic, metaphorical or allusive> thi1; signification is essentially

rational. As we examine each of these forms~ we shall rry to see to what extent they approach or diverge from the structures of language, what ensures their intrinsic originality and distinctness, with the understanding that mise-cn-scene - the directing of actors - is the art of utilising, harmonisingt fusing together all these significations into one single signification - the meaning of eh~ film.

III

The Direct Sign or the ccNeutral Ima, According to the linguistic definition, the direct sign (n sign, or ge~tal~-sig1_1) is a ho~olo~us fi~e,. a kind of dui: where the ~ngn1ficr 1s coextensive with the signified. There is no such thing as a neutral image. The neutrality ... considering here is only a hypothesis to al low us to study ho image - photographic or film - is produced, outside the cont, the intention which provides its basis. And this image, limitc delimited by the camera lens, is of necessity held within a f the result, by re:ason of the techniques employed, being an preta.tion of what is photographed. There is no lens with an angle of view comparable wit bum.an field of view. The widest angle lenses may contain a si area but their short focal length and heavy optical distortion the effect of increasing the impression of depth and giving a curve to the vertical lines at the edges of the field of view. J other extreme, long focal length lenses uflattcn., any dis~ Only unormal" lenses (from 25mm to 75mm) provide a s effect similar to human vision. We know that our representation of the world depends o sensory capacities. Totally different representations are thci per~cclly conceivable and may be presumed to be as "true~' 1 reality ~prured by our eyes. However, to avoid for the tune any epistemological consideration we shall refer to the real our collective perception as being~ utrUe reality,, i.e. what i for us I h" h , · n w IC: context the effect of long or short lenses tn d eseribed as a d'1ston1on · . of reality or an interpretation of th < 0 f VICW, In another area

Li •



acco d' 'qua ties of hght are given more or less co1 r mg to whether th . . h , panchromatic film c una~e 1s formed on orthoc roma optical {add"t• ) · The same is true of colours. Recorded i ivc Proces perfect photog bi scs) co1ours are matched to pr odi

rap c reproduction. But only in close shots. Ir

~

Ths Direct Sign or th~ c'Neutral Image,.

31

shots - espccialJy because they are further away - the difference in axis of lhe lenses involves effects of parallax and produces unacceptable fringing. Which is the reason why the subtractive process is prcfcrrcdJ where the colours are produced by a chemical reaction on three superimposed light sensitive layers. However, lhis colour reproduction is no more than a technical artifice presenting the equivaknt of natural colour instead of a direct recording. All these permutations provide the film-maker with the basis for his incerpretationJ emphasising or altering them as he wishes. However, to describe these as "transfonnationsu when they arc generated entirely by the recording processes seems to me particularly misguided. In fact., the colours of a landscape change according 10 the hour of the day, the angle of the sun., the density of the atmosphere and other physical conditions, which leads us to wonder what is cbe "real" - or natural - colour of things. The extension of which is that the impersonal~ purely technical r~roduction of a photographic recording may seem as true as the appearances of perceived reality. Having said that, it seems to me that insufficient attention has been paid to the huge difference between the photograph and the film image, a difference not confined to the reproduction of movement.

Photograph and Jtill frame The photograph is clearly the "direct" sign of what it represents. In the photograph the sign is not only coextensive with the signified, it coexists as a separace entity both different and similar. The photograph of a person retains the impression of his presence. It constantly refers back to him. His going away merely reinforces the impression that this image: is the only testimony of what his physical appearance was at a particular moment in his existence. This is also true of each individual frame of a film. Every single frame may potentially be compared with whatever it representS; the film image, however., which is no more than the action of light and shade projected onto a screen~ acts rather differently. It cannot be recorded. without a support. It can only exist as the effect of the continual and regular substitution of one frame with another: and for the image to exisi ihe indi'Uidual fratms musl disappear. When the projector stops, the impression is lose.

32

Semiotics and tlu Analysis of Pa'lm

Although the frames are the elements composing th . they cannot be considered as a collection of 1,b.t ;) lllo,,j idUlle, . th 1 s ( .word is meaningless),. for at e moment the bits b - or c thC th . . -....iDJ.e ~ed from each other, e movmg image stops ntov'lll :sq a r • rived of movement th ey -stop b emg . · of moven-.g. :Bt·1 llnits dep . . ul . b •uCnt· , _, are units of dynamic arnc anon - to e considered , _. thC.J •n1 ~, • as ea· nu,.tic units though cerw y notJ ...m units♦ Which is why it is mistaken to regard the individual frsune ' la' equivalent of the second arttcu non o flan guagc., as have cl1il semiologists too quick to compare film language with :n language and arbitrarily attrib~ting _to it similar structures.~ simple fa.et that the seron~ aru~la_taon: may be sc:en as. oppos the first is for us a sufficient d:1stmct1on for the :rnoincnt: ii always possible to place the word and its constituent syllables phonemes) side by side, whereas one would be hard put consider the moving image and its constituent frames together. By its very stillness the photograph mark-s itself as a 1 whereas the moving image, by reason of its movement, its con, ual transformation, cannot be considered as the sign of the obj1 it represents. Unless it is every moment of the object repri:ser. at every moment of.its representation - which is precisely what individua1 frame is, without the movement which makes i moving image. The film image is no more the sign of a filmed reality than reflection in the mirror is the sign of my person (though it is fl~g sign of my presence). In this image it is I signifying 9eet!1g _me see myself. Exactly as objects signify rhemH/tJe5 proJ~g themsel'IJes onto a screen; they present themselves: d,uplicate which formalises their meaning but avoids being a ~ sign by being their living ecpression. ,

Whne rtality expresses irself in its imqe la T~ the extent -that the film image is uneuttal'' i.e~ not ted, to the extent tha . i , . by Wh , t it s snifies nothing except what JS _,,. at)n represents; to the extent that it re~oduus (and P•"""uce the mean. . r•

man .an· 1

s does B. idea or transpar tng appropnate to it., we may accept 8• cncy and concur with Roger Munier that: '

,l!jl&~

In the cinetna [ ] . th · · • • leaf really quivers; It expresses itself

Ths Direct Sign or the "Neutral Imagen

33

leaf quivering in the breeze. It is a leaf as may be seen in nature and yet becomes even more than this from die mODlent that, as well as being a real Jeaft it is also, first and foremost, a represented reality. Were it merely a real leaft it would have to wait to be signified by my looking at it. Because it is represented, duplicated by the image, it has already signified itself) proferred itself as a leaf quivering in the breeze. What fascinated audiences at the Lumierc projections, far more than the precise repetition of natural rhythm, was this self~cxpression of movement in the image. The leaf projected like this was, through the power of this auto-language, more "realH and meaningful in its quivering in the breeze than the actual tree-leaf. What was fascinating was not so much seeing the duplicate as being aware of the photographic power of the expression through the duplicate. Something was. being said which had no conceivable equivalent in nature. [ ... )

He goes on: Photography is reality-turned-into-expression. Rather like a word of the world. In the photograph the world states itself as world, in its undifferentiated self, before any abstraction or choice. It is pure revelation 1 • Bazin, for his part, reminds us

that:!

in photography,

for the first time, between the originating ob;ect and its reproduction there intervenes only the insttumentality of a nonliving agent. For the first time an image of the world is formed automatically, without the creative intervention of man, according to a strict determinism ( ...]. Photography enjoys a certain .advantage in virtue of this transference of reality from the thing to its reproduction 2 •

Expression achi(if}td by artifiu .However, to say that things signify themselves by projecting themselws onto a screen is merely a :figure of speech. A way of considering a meaning which the things MqUire as though accldentaUy through being represented on film for, though it may be trUe

34

s,miotia and th• Anal,;vm of Pil,n

that light autontatically ~rds the image of things on i ck this same image proJccted onto a screen is not . ~~~ of reality"1 lik~ a pho~graph, bu~ realitJ, itself Pt-e~ moving i,noges, a duplicate,. ~s &t wer~, with the sp_ccific ad qualities of whatever makes 1t a genwnc re-produclion. · In fact, the camera has an undoubted e_ffea on what it sh well as the framing, angle of shot and lighting, its simple graphic quality is already an interpretation. Beside whic the camera,s point of view is necessarily posirioned, the i~ ever presents one aspect of' the world., an aspect imposed vision. What we are seeina,; on the screen is what an already seen1 which has an automatic effect on the dat1 directly perceived reality. The world is no longer a'VaiJab. rlaat particular chair, seen from that parricular angle., henceforward all conceivable chairs - and all conceivable · chairs". The image is thereby referred back to the COD suggests the idea through a form at the same time as it m object 11unreal" by rejecting the reality of which it is the i order to present itself as an image. In other words, it is pc less between reality and fiction than between essence a1 tence. It evokes an essence through an existence as though an absence through a presence: it reinforces the prescnc chair shown but, as an image,. it affirms the absence oft which nevertheless I can see in it. It affirms a real datu very unreality retaining only the forms and appcaran

univ-erse "'without substance;'. HJ'i' • al~' r • •rgrn sa.lity or "ucondar.yH reality}

~nse~ently, to refer to the ima.a--e as "'revelatiol'l consider tt as a • . 0 • deepest . more mtcnscly perceived reality, exper1enc Ba . · mean~g, and not as a ntranscendental0 · reality in ' un means n (th h

ear oug we need to know exactly what t ances, tr~n~? of th~ physical world, to all intents and aspect of 1·.. Id ~rcCJved reality since the latter is < ,. eabsm ·st • Pt'eallmed evokin exi s to the extent that an m us,~ is th~ con g Platots Ideas; an In-it1elf whose "re• rev-• . . l -.&ed by the crete cameracxn-s• ra.- lOD and whose existentta I Clearly lhe ·cam tnere)y through its objectivity. ~ .eni, Which is nothing more than a mean). Th

35 without memory or consciousness, is capable of :recording what Bazin calls "virginal reality". He writes:

Only the impassive lens, stripping its object of all those ways of seeing it, those piled-up preconceptions, that spiritual dust and grime with which my eyes have covered it, is able to present it in all its virginal purity to my attention and conisequently to my love3 • But this is no more than a notional truth. What I attach to things in the way of accepted, codified meaning, utilitarian, cultural or social signification, I discover in their image (or attach to it) so that, though it may be devoid of any subjectivity, the reality recorded by the camera loses its supposed virginity at the very moment it present$ itself to my eyes. Anything expressed de Jacto by the moving image is not necessarily connected with film signification, i.e. with the arbitrary meaning ascribed to things through the composition of the images 9 the organisation of the narrative structures, etc. As we have said, this is the dggree zero of film expression. However, we must clarify certain definitions, in order to avoid any confusion or contradiction. Though there may be no such thing as an intenrional expression or signification, what does exist at least is an objective qualiey deriving not only from the objects filmed, their meaning ''as objects", but also from their reproduction. A secondary signification dependent on the way the objects are recorded and formalised. A signification which is not appended to the primary signification, as is sometimes claimed, but envelops it and alters it, for it is obvious that the representation alters what it represmrs. Necessarily and fundamentally subjective, the message - even a simple report - conveys only a mediated reality. The impersonal, the absuact generalisation are the property of words. I defy anyone to translate into audio-visual terms: "Every day at the same time the marquise went for a spin in the woods'\ for an image: can never translate the indefinite article. All I will ever see is that particular marquise, never the marquise or a marquise. And always from a particular angle, in a particular context. She must be given a body, the time of day must be specified. She will come out of a panicular mansion or apartment

Semiotics and the Analysis of Film

36

~-.

-.;,.•·~·:., .. :-4.._

.

~.., -I

_.,.-1 --:~ •





_.

The image reorganises appearances: John Ford's TM Grape~ of lf"r,

block, she will drive a particular car, a Citroen or a Rolls be seen from cJose to or far away, in tracking or static ~ So many different methods of saying the same thing, < but formalising the same action in different ways confei specific meaning, an allusive or analytical, informative or interpretation, according to a slow or fast rhythm. W : 0 ~dabout way of saying that information., even the mo znev,iably be.comes a kind of personal discourse. The documentary, the smallest news item) whether good or already fJJOrks of ari., or are on the margins of a work of 31'1 The anoving · miage · will never be could never be, a 1 1 angua~e potentiaUy capable of ~rganising itself tov aesthetic pn.,..0 • · e· __ ., se. 0 n the contrary it is an artistic · deve1oped, in w . . ., . 1 . tha ays sumlar to those of language. We mJg say t, whereas v b l 1 th :2i it sigrufi er a anguagc expresses only to e e expresses~s, film language signifies only to the e]{ten1

IV

The Image and Perceived Reality Up to now semiology has been confined to more or less linguistic preoccupations and very Jittle attention has been paid to imagery as such, i.e. as perceptual information. To my way of thinking, this provides a convenient starting place, specifically to establish what relationships there are between the image and the reality of which it is the image; also what we mean by reality. Before we consider whether there is such a thing as an uimprcssion of reality"., we should define our terms and agree as to what reality is. What better definition could mere be than that which obstructs our senses? Can a concrete reality exist independent of the consciousness we have of it 1 ? Obviously there can, but can this be seen as an absolute? In other words., does this reality exist uin that formn previous to our consciousness of it or is it a subjective construction., an "effect of consciousnes.sn? The response is different according to whether one is an idealist, materialist, spiritualist or positivist. For the Kantian idealist, though phenomena are no more than simple representations, it is clear that they are representations of a reality which transcends them~ a reality given to each individual mind> impossible to isolate in fact but which remains in principle the genuine ~'thing-in-itself". Such that Kant draws a comparison between perceptible reality - or phenomenon - and an intelligible reality, the object of reason andl by extension> the absolute reality> which he calls noumenon. Thusl Noumenon merely takes the place of Plato's Ideas> with slightly less mysticism. Though the concepts are different, the kinds of relationship also, the uobject-in-itself" i~ turned into the "in-itselr' of the object. What cannot be perceived by the senses is rejected only to affirm it in an "essential" ideal postulated a priori. Even sol idealism finds its psychological expression (not just at the metaphysical level) in the thinking of George Berkeley who

Semiotics and the Analysis of Film

:38

the impossibility for the individual to forsa,~A h' . • J . A,.._ 1s 1 dual consciousness. Pierre anet writes:

pastu1ated

Idealism, taken as a whole, requires definition: any sys.te reduces the object of knowledge to the subject of This is how it has been formulated: Esse est percipi; the b things consists in being perceived by the thinking subject2 _e

kn:!

From this to argue that objects are merely constructions. mind was but a short step, dangerous to take, but happily nonetheless. Berkeley's position, however, was a long way solipsism and his philosophy, which seems nowadays to t most up-to-date of all the classical philosophers, does not dt the ridiculous position into which his disciples unfortunate it. In point of fact, Berkeley never denied the reality c external world; he confined himself to proving that it was uni ab1e except through the representations we make of it: "Idi passive objects) do not really exist in the mind except received via the scnsesu 3 • In other words, &se est percipi s not be translated as "the being of things consists in perceived by a thinking subject'> but ufor the thinking subje, being of things consists in being perceivetf', which has a quite, ent meaning~ we would all agree. Husser! writes: .~ an absolutely universal rule, in no possible perceptioa !s, no possible consciousness, can an object be presented immanent reality4 •

He goes on: All consciousncss · · consciousness of... •s

r1 ·

In other words . . a matcn· 1 fi ' consciousness d0es not exist "in itsel ; it a onn of li . · , does not exist wi ~ea ty. The object of which I am con. consciousness _ tbin my consciousness; it is a datum c and cons\UnJnat ~?ne otber ~an perception itself, fully re what is pe · e · a perception which "knowsn itself thl rce1ved. C.On . . . b. 0 Jcet of which . sctousness of the obJ. ect coalesces wit . one 1s con · • ·1 Perceived reality d scious. The: object is correlauve W1 an Perception. ·

The Image and Perce,"wd Reality

39

Modern science and philosophy invite us to consider that what we declare as real is merely a structuring, an organising of our sensory faculties formalised hy what we are conscious of, also by what we are conscious of our own consciousness without ever being able to appreciate the organising functions by which ou:r consciousness construct$ itself as it makes us perceive what we call reality. A reality which is not an illusionJ but is real and true only for us. In other words, if our sensory capacities, our perceptual frontiers were different, it is very probable that our world would also be different. Reality is not an ''in-itself" but one of the many aspects which phenomena may adopt through a particular consciousness of which we can only ever know the surface; it is a umateria1° world constructed by our senses. A physical reality which transcends the percept is perhaps an illusion. For the physicist, it is the manifestation of Energy. For the mctaphysician, it is a manifestation of Spirit. Now they each may be merely two complementary aspects of the same Absolute, the distinction being merely one of category of understanding. Having established this, let us examine things from the angle of perception. We know that visible light is only a tiny part of the electromagnetic waves whose frequency varies from one or two cycles right up to millions and millions of kilocycles, from Herzian waves to the frequency of the pl'oton. On a scale somewhere between 380 and 770 million million kilocycles, visible light extends from the extreme red (the limit of infrared) and extreme violet (the limit of ultraviolet) ends of the spectrum. At either end there is a threshold beyond and below which the human eye is incapable of perceiving light vibrations. If, for simple convenience, we term semory kvel the: area between the minimum and maximum thresholds, we realise very· soon that all our perceptions, whether in space or time, of any rela:.. tionships of intensity or movement, and visual, auditory, tactile impressions involve a similar sensory level, limited as they are by the two extreme thresholds. Now this sensory level serves as a framework. Acting as a grid, it enables us to record certain events, which we call phenomena or objects, but remains unaffected by many others which ''slip the net''. At the same time, it would be wrong to think that these "objectsn might exist Hin thcn1selvesn as we perceive them, that they arc simply drawn out of a context which remains alien to· us,

Semiotics and the Analysis of Fil,n

40

. . we who srructure the forms through which they 1s . b. ,, ap · us, wh o "make them mro o Jects •

for

Jt

Form and substan~ Let us imagine someone lowering a pipette into the sea it to draw up a sample of water. He_ ~en l~ts out the contct:1 by drop: the drops t?u~ created onginate 1n the ocean butt not exist as drops within the ocean for the reason that the 0 not a collection of drops of water. They are correlative both ocean which provides them with their substance and to the which structures their form. Let us recall Bergson s dictum about the runner:

1

1

A thousand successive positions are contracted into a symbolic attitude, which our eyes perceive, which art duces and which becomes the universal image of a n man5 • We know that a flash of red light of a second 1 s duration cc so many synthetically captured basic pulses that it woul 25000 years of human life to be· able to perceive each om: rately; and that cenain vibrations 3 at the ultraviolet end sp~c~) inyolve a greater number of frequencies from the ~uttmg parucle over one second than there have been s1 smce man first appeared on catth. Now, we have only to imagine a ball bouncing discontin and randomly at a rate of two hundred bounces a second to thath t>bhlo;""evc:r sharp our vision, we could never isolate each c a..1..__ I s separate PoS1t1ons. · · We could never expect to rem 111 ore UJ..iill one o twO '• fr . 0

matic a . r as ecze frames (supposing a quas. together~ys:mof movement). Yet our minds immcdiatel for u.s re c erent positions in a more or less broken line , l present th 1 second . Tu·Is represe movement of the ball during the course . b , over a .Period f . entation is neither true nor false; the a, tions but in tho .nme~ have been in each of the remembercc , e interun · ·1 I words, a diasrarn f ' I t w1 I have been in many others. n a shape structu 0d th e movement will be merely an abstract_: l)ercei\'ed P the constant exchange of energies, responding, corresponding and balancing: ea.eh other out, remain beyond my perception. Which does not mean it exists any the less "at the same time and in the same space,,: it is the piece of metal, which I perceive as a body from the surface of which visible light bounces back, which resists my touch ae we11 as my vision> and presents itself as an object, a tbingJ by virtue of the resistance it offers all my capacities for understanding and the fact that my consciousness links, in the form of a smooth and homogeneous shape, the finer points of its impenetrability. And so this "statc'' really does exist, but not as an "in-itself'', not as independent of a number of the other interdependent states without which it would not exist. Now, it is uniquely this state which I perceive, to the exclusion of all others and whichJ alone, isolatcd, uabsuactedu, forms the "'piece-ofmctal-for-meH. The piccc of metal is therefore not a part with a separate existence within the whole3 but a reality structured by the assembly of the several phenomena - or physical realities - which make it up. Consciousness neatly avoids what it cannot perceive and constructs an image by creating tighter and tighter links between the elements which impinge on it through the senses. Yet the nonperceived elements arc as much an integral part of reality as the perceived elements; the "objea" is therefore merely a slice of the totality of the physical world; but reQrganised and structured by consciousness into a coherent entity. Correlative with this world and the: act of consciousness, it does not involve an "'in-itself". There is no possible transcendent uessence" since the phenomena from which it derives arc different from what it is and since it exists as an object only by virtue: of the act of consciousness which uproduccs it as an object»~ as the logical construction of perceptible data.

42

Semiotics and the Analysis of Film

The real a11d reality Perceived reality is real in the most concrete and tang·bl . 1.or "' us1 re Iatmg . to the cond· I e. se11•'ii the word but it is a rca1uy or , . l k llJons

our existence, our phys1ca ma e-up. . .~ Of coursei objects are the expression of a phenomenon Whki Precedes them and excends beyond them; but this phenoni . · existence · " as o b.Jeet s ,, , ts . no t an •~ essence~, of wh· eno~ alien to their , lci Th d they are merely the outwar appearance. ere as no "ideal" fon: preceding their representation and no more relationship bi:twce the stimuli and the perceived object than between a building ao the materials with which it is constructed. However, there neo:i the involvement of a builder. The builder in this case is pcrcti: tion, which is able to build only within the limits set by our sensr.:r. Thrust in the midst of a reality extending beyond the limits 1 human understanding, our consciousness organises, with 1~ means at its disposal, a reality which it is able to dominate b(.'Cam it is the consequence of its efforts. In other words, the direct da1 of consciousness are the effect of a constant and ceaseless mtdi~ tion. This arbitrary reality is for us the only true reality; it scrn direct to us only because this mediation!' dependent on our sen!;01 levc:l, it is what forms the actual act of perceiving. Instantaneous! str ucturcd by this process, our reality is true/or us, direct/or rts. For beings endowed with a sensory level different from our owi tbe reality of the physical world would appear totally alien, sm tbat their "universe1 ' wouJd have nothing in common with ~1 own .. The ('object' 1 structured by their consciousness U$II p~ysacal events of the same order but whose perceptible da m~ght be diffen~nt would bear very little relation to the one'' ~~ 1 remember. What right would we have in judging it less r4: •n ltscJr'? An un reality '"for usn might very well be a real1'tv· J'~ rhem. The obj f th • le: true th cct O C:Jr consciousness is not more tn1e nor 1 - wh .;:: our own. It is rnerely one aspect of the "physical wor'

po:ssi~J

presupposes as many different aspects as tbere

3

e sensory levels. Sensory data ( · ·at'~ ( which a p • sens.at1ons, stimuli) are the "raw rnateri 1o articular fo b . Yet ta, meaning is rne . rm estows a particular mean:ang. i;; conscious of i'tantAnngful only to the extent that consciousnCS·tl . d consciousness . . s tO • extent that it ..___ can only be consciou r.,l'\.~omes. c • d esn11

onsctous of this form: Form an m

The Image and Perceived Reality

43

are the "wayu what is perceived appears to comciausnes.r. Meaning is therefore inseparable from fornh but form is not an a priori to which the sensory data must submit: on the contrary, rhc data are what deternaitlc it. To say that these darn are "directly informed'• (that isJ given meaning) is the same as saying (as we have just done) that the structure and meaning proceeding from it are previous to consciousness - which is confined to "recording" the product of the perceptual act. However, though pre-conscious, this act is already a creative acr for, although the "spontaneous structui-e" is uwhat appears to our consciousness"; this structure is not presented to our perception; on rhe contrary it is comtrucu.d by it, though it has the ground laid for it by the sensations in that thefie sensations arc U,m"ted. Forming a dead!,' defined "field" by virtue of this limitation, the stimuli offer a network which perception formalises into an organic uwhole". Pcrcep[ion creates associations, connections, structures. Perceived reality is the form of our perctptitm, which is predetermined> i.e. "framed and limited .. by our sensory level. 1·0 perceive is to construct a world; to be conscious of it i!'I to present the world as an object.

Visual perception Vision is of particular interest to us. Firstly at the level of direct images called ureal" (or, incorrectly~ retinal) in order to distinguish them from mental images - photography, painting, etc. As Bertrand Russell notes; From the physical point of view, whatever I sec is inside my head. I do not see physical objects. I see effects which they produce in the region where my brain is6 • Clearly) the image forming on the optic nerve exists within the organism. From the physiological point of view, the neural process corresponding to the perceived subject occupies pan of this organism; to begin with, the back of the retina where the image is formed. But this image is nor what I see; it is what i..,,; sem. 1 do not see the effects produced in my brain by the act of sccingt I surrender to them. I do not see my vision, it is gii,•en to me. What I sec is

Semiotics and the. Analysis of Ft'lm

44

the result of that vision, whi~ appears in my consciousne because the retina d' $$ 31); bCCOm es immediately extcmahsed , • • > ISpJa , the properties of bram tissue among wh1ch the ea _Y1T11 11 :xtemalise perceprio~s - does not presen_t the forms ::c~)·_ ~ • s·de or even alongside my eyes but outside them. No . t:11) m ae time because I am mcapa . ble o f re lat1ng . Wi at th to '•me~• th . sam , . . 'h emfo1 mation I am expcnencang and whac comes to me .,fro • u· • • n t th (tJ th outside", I relate •~ m _intention o e cause of my "seeing' What is seen, an~ 1s an rmage, !11en becomes ui?at I ste which • tum into the obJect I am lookmg, at. The subJect of my u· ••o becomes the "objcctu of my looking. It becomes objectified in that "object" existing outside me, according to a process which 1 still quite vague but which my tactile perceptions help to brin into focus. It could never be said that consciousness "becomes transccn-de in the dire,rion of the object", for this would be to see the object: already existing in forms attributed to it by perception. Consciou ness, simply, "knows itself, within the object - and through t} object. When I see the red colour of the blotting-paper in front of 1n the red is not objective in the physical sense of the word. It is subjective construction presented to my consciousness (since tl mediation of the percept is previous to it). Consequently, the red objective ''for me". Unable to relate it as a creation of my will, can only receive it as an external cause. It is thus spontaneous and definitively perceived as a quality of the object. And what true for this quality is equally true for the object. The ~lour docs not in fact exist as such in the physical worl It is a vibration of a certain frequency transmitted (reflected) O\' a surface. The subsequent stimuli" are interpreted as the ~•colO' red'' b · · ..,..._. > ut •t 18 a "representation,, of reality and not reality iise J. o1s colour red · • · di ect 18 0 n1 Y true (real) for us because 1t 1s r presented to us The h . . . 1110 relation hi · P ysic1st 1s right to say that there 1s n 0 . betw s between these vibrations of light and the colour t~, itselfc_en c colour and the word uredn used to describe it - whi· 1s merely a "si ~, f. However if co ~ 0 its perceptible qualities. .. process rath' thnsc1ousness is merely the product of a struccur11 ' ct an the • the effects Prod d . Process itself, we may well wond c r h< . uce in our b ta.Jn . b y the act of seeing appe ar to o consciousness.

l

The Image and Peruived Reality

45

In attempting to provide an answer, we should examine the act itself. Let us not forget that the retina is made up of one hundred million cells whose messages end up in the brafo in a visual zone which is explored, ten times a secondj by an associative :z:one, duplicated by a psychic zone, the whole system enjoying the benefit of an extraordinary network involving approximately thirteen million neurons. It is by means of this "seeing machine .. that the stimuli become translated into image,.. As we have just described, there is no such thing as a retinal image, if this is interpreted - as it was thirty years ago - as an image produced on the back of the retina. as though on a mirror. The wor-k of Henschen and Wildbrand have proved that the retina does "projecft onto the brain. To describe the process more accurately, there are two retinas, one on the: pedphery which receives me impression of light rays; the other situated in the calcarine sulcus of the occipital lobeii, which is symmetrical to it forming the cerebral retina. And each of the two cerebral retinas is related to the two eyes: it i's th.er~ that the impressions become conscious. Establishing the: formal links between the imaged data and forming the image into an organic whole, into a form, the cortex fulfils perception. Thus consciousness defines itself in what appears to it. And if it is to be compared with the surface on which the image appears, it is tantamount to saying that it is the brain itself, i.e. the whole upper level of the sensori-motor mechanism which acts as a control while observing it. To put it another way, using as reference:: the recent work of David H. Hubel and Tonten N. Wiesel7 , the retina, containing over ten million rods and twenty million conc:s, transmits to the brain via the optic nerve signals which are more or less intense according to the number of photons received. The optic nerve carries out an initial process by the fact that its fib:res (estimated at over a million) assemble the impulse into several hundred receptors. A second process is involved at the level of the chiasma, i.e. the point where the optic nerves associated with each eye:: meet and cross so that each cerebral hemisphere receives the nerve-impulses from both eyes. Which c::xplains why binocular vision is simultaneous and why it is possible to see in three dimensions. In shon, it is the cerebi:-al image formed in the cortex (striated cortex or zone 17) which acts as support for- the visual image.

46

Semiotics and the Analyru of Film

,,..n..:ch is constructed from received impulses and is Wh W IU • th •_ ~ . ~fC and any correction :e1at1ng to e 11.uormation Provided other senses play their part.

Real images and film images: movement and illusion Though the perceptual image ("objectified" visual Pere inseparable from the perceived objects:, the direct image shall continue) through habit and convenience, to refer· t retinal image - and which is "flat',) can be seen as diffen me objects shown as images. In other words~ the landscar view presents itself as a two-dimensional image - thougl three-dimensional reality. Or, to put it another way, I cot a window between the landscape and me; it would then a me through the window as though projected onto it lik image projected onto a screen. On the other hand., from the moment we exclude everyt visual perception, we perceive the film image in the sami we perceive objects. The image of a landscape presents itsc vision in the same way as a real landscape stretching out bi eyes. Narural1y I am not able to sit on the image of a chai the same time nor am I able to sit on a real chair if it is p the other side of a window. It is therefore as though rl: projected onto the screen is a real world as seen th window. With this important difference: that the image es· between the elements included in it a series of rclationsh iJ do not exist in reality where there is no frame by which at of segregated space can be removed from the world., to be ered in isolation. in the context, of film images, we always ref~t smn of movement 1 • Doubtless because this perceptH'l ~o tbe discontinuous linking of a series of images which ""Tc e ~ed and whose replacement time is less than that 0 J>Crs1stencc t:: a sth f da fourth f utteen o a second in silent films, an If m: a second since the advent of sound. frames _v:ent were to be broken down into a greater n'L jerky pan or 100 per second - the rendering of rapid act have to be d . smoothed out but by the same to k en lllanufac:tured as a "reconstituted'~ movement! tl on. Movement does not present ttsc

c:

•~ii :t.

s~w~

c=:-'1

The Image and Perceived Reality

47

image: it either is or is not. In the image of Clbjects in motion there js perception of real movement. As Rene Zazzo describes it: Movement is not in addition to the image. It suppresses it as an image, instantly altering it into reality. At this le\•el the feeling of reality is not a construction of the mind or a product of the imagination, it is an immediate reaction8 • We might add that the perception of movement over a quick discontinuous sequence is less the effect of retinal persistence than a relative inertia of the cortex, called the «phi effect,, by Wertheimer in l 912'J. The perception of real movement is similar. Without delving into ir-rclevant scientific considerations, we may say very generally that continuity docs not exist in the reality of the physical world the continuity of wave movement is only ever the continuity of alternation. In other words~ what we term continuous - what is continuous Jor u.s - is always a non-perceived discontinuity. The formation of the television picture provides an example: we know that this image is not the effect of a whole $eries of frames being projected bul the transmission of a series of succc-ssive dots. The television tube comprising 816 lines (or 625 according to the system) and each of these lines constituting a thousand contiguous dots) it is the succession of 816,000 dots in one twenty-fifth of a second which reforms the image across the whole of its surface. Remaining visible from the appearance of the !ast dot by virtue of the calculated inertia of the receiver, the first dol of the first image is replaced by the first dot of the succeeding image, and so on. Now, unable as we are to perceive any discontinuity less than a twenty millionth of a second, all these dots are perceived as a global unit replaced by another at the ratio of twenty five ever)' second 10. As well as this impression of movement> the "impression of reality" has provoked the most varied and contradictory comments.

Perspective and the "impression of reality" We know that it is from thr: data of the camera obscura that the theoreticians of painting, like Picro della Francesca and Alberti,

48

Semiotics and the Analysis of Filrn

starting from the .. d c fin e d the laws of perspective~ Jd, h'1s o b servauon-point . Pos1t1(J . ter in front of the wor i prun . . • , .e. the . providing the geometnca1 interpretation of the 8 view . d "th b J. . Pe for there is only one centre, that of the person looking or whatever takes his place (the camera). It is the observer who chooses where to place the perspective effect and, therefore, the point of view, for there is no such being as a spectator who is not the subject of what he seC1li. In photography, pcrspcc:tive is recorded onto film in the same way as onto the retina. With this important difference: that the pencil of light rays crosses an optical system constructed on the model of the human eye and that the image docs not extend beyond its registration on the light sensitive emulsion. The camera perceives nothing, it records mechanically but, when seen by our eyes, the photograph forms a retinal image to which the brain reacts exactly as in normal perception. However, because the photograph is inert, the vanishing-point and dimensional relationships of the ob;ecrs mean that, though we recognise the pcrspeetive effect~ we do not experience it. Space is "flattened" in an image where the data, ins.tead of being represented as in painting, are registered by rhe light, yet stratified and set. However, once they arc projected as moving image-s, the movement creates the effect that depth is experienced., felt as it is in

dire~t perct,ption. And the image appears instantly to become separated from its support, indeed does become detached; it is no longer a photograph projected onto a flat surface; it is a uspacen I am perceiving. The film image presents itself to my eyes as a "spatial image", similar to the real space stretching out in front of me.

Semiotus and the Analysis of Film

50

M Micbotte's observations., the consequence of a stud; in experimental psychology, confirm this idea. .A~r:at deat e sal-'s:

As soon as a technique can be successfully applied t the constituent features of an object from the surface .~.h~IJ;lr, · o f dimens1ona · 1·1ty immediately · wv tch a as its support, the nott~n ass'Un the obvious and sometnncbe_s e:cienbunexpected f~atures of rcaJi This resul~ may be. a 1ev ! severa! different methoi among which there 1s one (particularly mteresting fro . ) wh"ch • . setting . Ill C point of view 1 consists m up an intera . between the constituent features of the object. The anti~~ between the movement of the shape and the immobility of 1 screen acts as an agent of separation freeing the object frorn 1 surface in which it was included. It is3 to some extent, imate alised' and assumes an independent existence; it becomes a 'tt poreal object'. One very simple test, used for many yean now (reference to can be found in the work of Von Recklinghausen dated 1851 proves our pojnt. The shadow of a solid object made of wir( parallelepiped or cube, for example - is projected onto a scree Observed from close to, the shadow gives a similar impressi to that of a simple perspective drawing traced onto the scref but, for the object to become real, all that is required is to SJl i~ and, in certain viewing conditions, it actually becomes im~ siblc to distinguish the moving shadow from the metal obJc itself. This experiment is. important in that it reproduces precise what happens in the cinema where the behaviour of the cha a~ers, their gestures and changes of facial expression~ even ~ !:tmpl~ ~ansfcr of inanimate objects., must obviously eventu• ma surular effect. 1

M. Michotte adds that: it is_ ca~y to ascertain this if the film is suddenly stopped ~urii proJccnon Th · . ears it 10:s its reality ~d 0 • unpress1on of relief suddenly disapp. ic bna: with fla · gives way to the unreal volumes of a sirnP t l'Crspective11 ~

In a sense the :fiJ •

'



In Image· is comparable

. n in with th" rcfle b Y Vlrtue of the projection wh1'eh ' direct f0tn:t g Pro~ected~ connotations frequently appc£ s or specific •. . ~ the origin of th . qua11t1es of the denoted obJectS, e ideology which emerges "insidiously" f

~

The Image and Psrcaivtd Realiry

55

obvious but imperceptible manipulation, which fools no one but those who are blinkered and one or rwo rather misguided critics ...

From crin'cism to Critias Moreover, jf certain truths arc pushed beyond their boundaries, the consequence is sophistry worthy of Critias. For instance, basing his remarks on Godard's rather silly aphorism: '"the cinema is not a just image) it is just an image 16". Clement Rossct has this to say: The area (of just imag~s) relates to a cinema where the cinematic: expression of reality is not expected to be found except where it already exists, in the subject of reality, in preformed ideas, a pre-representation ready to be filmed. The justness of the :image is proportionate to the nature of its own illusions· of reality 17 •

I am not sure that my "representation of reality" is an illusion arising from a preformed idea. I rather think that it is the formation of a concept. If the image of a tree is formally identical to the tree it represents (whose reality is obviously at the limits of my perception~ controlled by it), if I recognise the one through the other3 then I would say that the image is just. And> indeed, it would have to be, otherwise I would not recognise the tree. Rosset goes on: The image is the expression, not of reality, but of the signification preswned to be contained within il [ ...]. I cannot sec how such a signification could be presumed. It might be an Haddition" or ..modification", the image being a duplication before it is - and in order for it to be - a signification: either I know the reality in question, in which case I have no need to presume or presuppose its meaning; or else I do nOl know it, in which case I have to create an idea of it - an illusion - which my perception will eventually correct. Rossct concludes by saying: As always, reality appears alongside the specitlc (just an image), the illusion alongside its duplicate (the .. just" image providing merely an illusion of the reality which it .. justly" claintfi to evoke).

Semiotics and the Ana{vsis of Film

56

camera 'WOuld be obi· I f this reality did .not exist, the . f . 11 or unjustly - 10 give me an image o 1t. Which in, obviously be no more than an aspect of reality, and ni conceptual "in-itselr'. But~ wherever I am~ I personal] ever see one aspect of the world. Clearly, this kind of sophistry is no more than word~ the proviso that the cinema "is capable only of turning unoriginal duplications and unsurprising representatioi is to recognise im.p]icitly that they are accurate rela, reality), it would be ••totally wrong to evoke this rcalit Now> the word ncvoke,, is ambiguous. The image has 1 to ~cevoke,,, since it represents (re-presents). E\rocati it may be associated with the word which describes it and therefore act - contrary to all other shots like a linguistic unit. With this important difference: though the word "'gunH for instance may describe a concrete object and signify the concept with which it is associated, the close-up docs not poss~s the dcnomi~tive or demonstrative function ascribed to it by Christian Mct2. The close-up of a gun docs not say uhere is a gun" or "this is a gun'>. It merely slww:s that rhe gwi is ther~! ihat particular gun and none other. As with every other kind of shot, although in a much more meaningful way~ the close-up assumes a specific character only in the context which determines it. Yet if it becomes eclipsed by what it reveals, it proves nothing. If it signifies anything, it is immediately the idea suggested by the associations which the gurt has with the events described in the sequence of which it is pan. From which we can see that though the close-up acts like a sign, it leaps over genuinely linguistic significations in order to reach directly the narrative or discursive significations where the association of sign and signified is always accidental. This is doubtless true of most shots, but the significations of the dose-up have a symbolic quality which distinguishes them from all other shots whose quality is mostly allusive, suggestive or simply descriptive. Meanwhile, it was this facility for possessing a constant "potential as a signu which surprised and excited the first theorists. Moreover, as wc:11 as the ideas and feelings of which it is the ephemeral sign, ch.e object presented in close-up inevitably draws attention to its perceptible quaiities> to everything which makes it different. lt appeals to the emotions but these can only be felt> experienced by seeing it. In isolation it becomes a .. whole,, relative to the frame which sets it apart, with.in its own constituent pans, the internal components which subdivide it, whereas, in wider shots, it is submerged under the vast number of relationships between the objects and the whole of which they form part. The close-up thereby presents a tactile~ sensual impression of objects. It concentrates on the object, on its foi-ms, all the rccognitive and dynamogcnic operations relating to the knowledge we have of it; and this before it makes any appeal to the intellect. Of

68

Semiotics and the Analysis of Film

objective thr all sbots:1 it is the· most concrete,b"most • ough s the most abstract, most su Ject1ve through what . W , SbO • • th It S Any emotion is contain~ m e rcp~esented object,. but als fiorms of the representation and the ideas suggested _ 88 . . h a other shots, obviously, b ut 1n a muc more relevant a cially, more meaningful ~ e r . Thus, in The :'. Potemkin, the pince-nez hanging off the· steel hawser give, emotional level, the impression of a disaster - but of a ti. . " represent ., the d octor who hI disaster. Not on ly d oes it thrown overboard (or rather his absence) by showing the the whole, but it also symbolises, not without irony, the co the old order represented by thi$ officer. Since there is nothing intelligible in the cinema which been given through the senses (visual perception evoking a tions, tactile and others., relating to a given object)., them; cism to be levelled at structural semiology, beyond the bib it has frequently taken, is that it considers significations on level of what is intelligible, totally ignoring what is fell instance rhythm which offers many possibilities for analysi disregarded by stru.ctural analysis.- As though temporal : ships.do not rcpresenc one of the most basic elements of fiJ: tu.re. So much so that, as the raw material of·art transcendi communication, emotional qualities play a considerable significations, the signified being almost. always filtered, e ~Y the emotion~. as much by what cannot be expressed as JS expressed. Shots do not just present a· difference. of scale or a mor ex!ended spatial field~ Using· the represented objects to co suitable form, a "specific qualityn each of them acts diffei ~e ' . . percept and, therefore, the consciousness:11 the emon i~tellcct. The same script, shot first in wide shots and thcr. ~ose-ups, would provide two entirely different films, ever ey ~ould be following. the same dramatic development, 1 narrative sequ • . ence. The story-might be the same, but the s1ons, the idea 8 d fi . . , an eelmgs expressed would be quite Would have a diffi · · the rhyrhm . erent \ralue., a different meaning - partic W s were also changed e have only .,0 • • • • .-1 "' a shot 10 · • ·· unaginc Dreyer's Tiu Passion OJ Jo wide-shots th · h ts · would ha th ra er than close-.ups and mid-s O • vc e. sante Zllcaning. 1.

TI~ Shot

D~raila in

ia

69

dO\'\-nw~rd cilting close-up: Carl Drcyer's TJ,e Panion of joan of Arc, 1928.

This example enables us co cover a sub-section devoted to facial close-ups used - mainly - to give emphasis to a reaction which might be imperceptible but reveals the psychic state of the character, such as trembling lips or fluttering eyelids, a sudden movement of the eyes, etc. These close-ups always follow in a sequence of wider shots and are, as often as not, the enlargemem of an expression ~gun in the preceding shot.

Facial crose-ups Towards 1915, when the use of close-shots (of actors, not objects) had become widespread, their integration within the sequence was very different from what it is today. In the most important films (those of Griffith~ Ince and twenty or so of the rop level of directors), transitions to dose-shots were made in srraight cuts. An action begun in one shot was followed through into the next. Matching movement had not yet been discovered, therefore changes in the camera axis had to be used to disguise the jump-

Semiotics and the. Analysis of Filn,

70

ts and overlaps but the dynamic continuity remain d.

cupoo·r films the solution was simple: a subtitle insene UJt . . f ed b shots softened th~ trans1uon rom one ~o _another, such tha difficult to exploit any sort of associative effect or ,, . t montage~,. It was artother processdialtoge1:1ter in standar:~ In these films, made for a mass au ence, mstead of a conr movement from establishing-shot to close-shot, the sa,.. in . an d wit . h.m t he same r:ime. ..e R were repeated in the same axis If, for instanceJ there was a courtroom scene in which a , had to give evidence, the scene would be recorded by two c: set side by side. While one was filming the whole scene, th( would be simul:aneously shooting the woman in close-up. Which would give: A. The woman stepping into the witness-box, raising hCJ band to give the oath,. glancing furtively at the accusec screen)l looking anxious then settling herself; Fade down an in/iris out to: B. Close-shot. The womari makes the same movements. Fade down and iris-in/iris-out to: C. (same as A.). Continuation of the movement: the 1" settling herself. The judge asks her a question. D. Subtitle explaining the question; etc. The close-shot (or close-up according to the context) was fore merely a way of showing a closer view of what had al been seen, drawing attention to something the audience • have missed. It was merely an addition included., a.s a shot,: body of the film, not included as a movement within the dyi continuity of the action. explains the c'onstant rnarkin unacceptable repetition of time which we find so irri nowadays when we watch reruns of old films. The; face as the "mirror of the sour) was a pet-theme for 2 ~f va_nous theoreticians towards the end of the silent era: ,~te _is psycho-anatyticaJ1J, Jean Epstein wrote. And Bela B t is a mirror in hi h f the s, rcfl.ect d h w c the source:i the essence o only ~ow, th e face is able to reflect this "state of ~ to pa . en Its expressions are related by the audience to an thin ss1lns. w~ose effects we imagine they translate. As with e g c: se, 1t 1s purel .. , In SOllle iil Y a qucsbon of relationships. d c up in the fir ent films the close:_up (the reason it was caJle st Place) was surrounded by a circular n,ask; .:ai

Which

;h ·

Thtt Shot

71

An "cxprc~ioni11:tk•• r:lose-ur, in Jc,hn Ford's Thrt Llmg Vt>yage H.fm1!1, 1940,

ounide the camera. Its use was abandoned around 1920, however. From then on, faces were recorded ' 1full-scrccnn and, even later on~ framed alongside a significant detail. In Drcycr's film~ shot almost entirely in close-ups, we can see that Joan's face is placed almost invariably between pans of either faces (chopped off by the frame) OT in front of characters (monks or soldiers) shown in midshot. Except for the sequence where she is alone in her cell, there are very few shots in which her face is totally isolated. It contrasts therefore with other similar close-ups (Cauchon, Massieu and others), for the filin is a confrontation of faces. Here they arc never enlargements introduced into the narrative sequence for the sole purpose of emphasising the actors' talent whose close-ups are unwatchable because they interrupt the narrative flow contributing nothing but a worthless additional expression. On the contrary, the disruption caused by the close-ups of objects are dynamic disruptions. They ere not perceived as disruptions because they gmerate meaning. In effect, a facial close-up tk,es not :signify; it expresses and . almost never acquires the quality as a sign which the isolated object assumes. Except, that is, in The Pa.ssion of :Joan of Arc:

12

Semiotics and the Ana.lysis of Film

ted and juxtaposed in a kind of abstract representa . contras . h . if t11 r: "al close-ups become, as 1t were, t e sign o what the ,aa ' film har . . 3' E This is what makes Dreyer s , so c acter1st1c and, sorn, certain of Bresson's or Bergman s films. Pans and tracking-shots

Amongst camera movements; the simplest and oldest (1 obviously the pan, which is the same as the vision of sc standing still and turning his head to the left or right, or t which is the same as someone lifting his head up or dow1 camera remains :fixed and pivots on its axis. · The term 1racki.ng-shot has various different meanings: e is a shot "in motion", i.e. filming the countryside from a 1 train, c.ar, cable-car, etc. The camera rcmans fixed and mov, the moving support on which it is set. This type of trackir is as old as the cinema itself ( The Grand Canal in Venice; 5 Promio in 1897). More generally, we mean it in the sense of '~truckin camera on a platform mounted on rails or rubber wheel camera moves in relation to one or more characters, ahead c or behind them, getting closer or further away, etc. This · camera movement, used in conjunction with the actors 9 ments or any moving object independent of the camera, w. for the first time - apparently- by .Griffith in 1909. The shot which tracks in between static characters (rest theatre), the camera picking up the behaviour of some of th acters in the drama or representing one of them moving rel, th e others, is rather more recent. It was used for the first t Mu~u in The Last Laugh, in I 925. ,, Smee the advent of talkies, cranes (1929) and ujib-anns ~ve enabled all movements in space to be combined, t widespread use of portable cameras during the 1960s (a advent of '' steadicam") has made a marked difference to tl of film . · s provi'd•1ng a freedom of movement previously uni ~le. A freedom which has also produced effects as intoler UJ.Cy are unsuitable Ma . ,_.., ,,de to translate the speed with which we tum our bes.di .,,~s work against th . . h we , tum our heads eir own intentions. In fact, w en and our eyes take l/6th of a second to move

Ill..

Thit Shor

73

field of .view 2 ~r. 3 metres away t0 one 7 metres awayJ any objects caught 1n our v1s1on a~e not seen as though uzipping" from one to the other. Mor~ver, if the field of view is very close_ 50cm 10 1 metre - the obJects arc reduced, leaving as ir: were a tTace behind them. Even so they are not blurred. · Firstly because, in not being limited by a frame, the visual field is unaffected by its isolating effect, and by its associated sr:rucrure. s~ondly, and more importantly, because we are unable to move our bodies (or eyes) faster than the minimum time for visual perception, that is, the perceptual threshold which varies {according to the lighting of the field of view) from between I/50th and I/ 100th of a second (nothing to do with retinal pcnis1ence which is associated with it in a different way and is related to a short-circuit in the cortex called the uphi efl'ect0 ) . In other words,. to perceive images as if in a zip pan, we would need ro spin on the spot at a rate of something like 20 or 40 rums per second. The camera, on the other hand:io recording at a rate of 24 frames a second, will record only 4 frames in the same time (7 metres in l/6th of a second). And because the shutter speed of the camera is not fast enough relative 10 the speed at which the camera is moving, each of the four frames will be blurred, which is what gives the zip-pan effect. I am teaching grandmother to suck eggs. Every member of the camera department knows this. And yet there are still cameramen who persist in panning the camera at speeds based on the human body. Trying to be true to lifeJ they arc consistently wrong moreover, they startle ·and upset the stomachs of even the sturdiest of audiences~ A similar speed would be conceivable if it were possible to record an even greater number of frames in the same time. Eight frames in 1/16th of a second for the same: distance would make quick pans at least watchable. Filming - and projecting - should therefore be at 50 f.p.s., which has many advantages, since the reproduction of movement is smoother the more ~c movement is broken down into a greater number of fTillllcS. This camera speed, technical considerations apart, ~as proposed by many cameramen at the beginning of the talkies. It was o?IY rejected for reasons of economy, since it would have meant u.smg twice as much film-stock. However, it would have allow~ silent .films to be projected "normally" by tripling each frame 10 new

prints...

74

Semiotics and the Analysis of Film

Editing against the movement of the camera Moreover; the development of film style during the . di . cour1 first twenty years o f th. e cenniry, us~~g e ting rather thall movement, has surpnsed many cr1t1cs and theorists . knowledge of the histo1?, not .onl~ of the cinema, but-:: tions of work - producing, d1rcct1ng, photographing _ period. We should let Christian Metz take up the story: 0 The history of the cinema between 1900 and 1915 i striking observation: the indirect technique - more ques in some senses:, less immediate - has played a more,., than the other; montage, and its corollary, the shooti 11 [translator 1s note: literally; 'technical breakdown1 ]> hav1 to liberate the camera in a more decisive way than movements themselves. Of course, panning and trackin, ments were not dismissed, but the most important consi lay elsewhere. Camera movements do not really exist as Griffith's films, although he exploits certain of them i11 wonderful way. The great events of these fifteen years represented by the first appearance of the panning shot dramatic effect (E. S. Porter~s The Great Train RobberJ nor the famous camera 'displacement' in The Birth of a but the first use of cross-cutting (F. Williamson's Am China MissionJ 1901), parallel editing (E.S. Porter's : Convici, 1905), changes of angle (in Porter,s films), and notion of inserting a close-up into a sequence of lor (A.G. Stnith1 s The Liule Doctor, 1900), then of exploi ~•ose-up for expressive purposes, rather than as a simpl in d~.atl' (Griffith'sJudi"th of Bethulia 1913; Thomas In Fugmve, 1914): in short3 the effec•"' montage7 • It' L,;t La: is ~o accident therefore that Andre MaJraux an< l"b ffay ci~e the shooting-script as a corollary of the CSJ11e ~ erated} whereas Bela Balazs9 and Jean Mitry10 consi c montage Tb • th ·s a Patad . ·. . ese writers choose not to reveal that 1 ea ox:. it 18 not camera movements which have n mera more mob'l T fi 1 e. the · hey write as though the trans ·orn CUlcniatograpb · th ol the succession in~ e cinema ccntered on the pr , tionat inodality of var1.ous images1 much more than on_ • oftbe itnage itself- such as 'camera move

of

The Shot

75

This may well have «gone more quickly to the core of the problem", as Christian Metz suggested but, at that panicuJar ntomcn t, no one could have known this. . . The truth is much simpler. As paradoxical as it may seem today> montage was used in preference to camera movement precisely because ir was much easi~r. For reasons which ought to be explained, before the 1920s, tracking-shots represented on average only 10% of shots. All (or almost all) shots were static. Since each one involved a separate t*c, a sepai-ate fragment, it was normal for them to be spliced together according to the order required by the dramatic action without taking account - before 1906 - of any signification produced by the specifics of the ordering. It was c>nly as it was put it into practice chat it was realised that this "asso1;iation of shots in a particular order;' produced a meaning, an unexpected signification. Which was immediately exploited, developed, refined by Griffith as well as a dozen or so othcn1 during the period 19101914, and enabled critics to discover rhythm and montage. Why were there so few tracking-shots? It was not very difficult to mount a camera on a dolly. However, any movement involved altering the focus, and cameramen turning the crank-handle (with their right hand) would have found it virtuaHy impc>ssiblc to operate the focus ring at the same time with their left. Moreover, controlling the focus would have required using a through-thclcns view-finder. Now, at the time, the view-finder was used only for "framing'' purposes. It was closed during the take, so as not to fog the film. A 0 sighting mechanism'\ parallel to the filming axis, enabled the actors~ movements to be controlled, but not the framing or the follow focus. Which was of no significance since the shots were static. In addition~ with the exception of the very rare and very brief movements "in depth'', if the camc:ra moved (midshots or long-shots) it always maintained the same distance between it and its object. From 1915 onwards (1912 in the case of Ince> Griffith and one Ot' two others), the camera-operator had an assistant beside him to alter focus according to reference-points chalked on the ground during rehearsals. Towards 1923-4, the new cameras \\.'ere equipped with light-tight (wit.h rubber eye-pieces) view-fmdcr-s adjustable to the eye of the operator, allowing him to frame up through the lens during shooting and, from 1926-7, the widespread development of automatic cameras, whose intern.al electric

76

semwtics and the Analysis of Fi'lm

liminated the need to rum the crank, freed th .. kin g-s h o t s b ccame more ... {)Pera Om then on., nac hands. F r th h d h null\ ,,..._'" dollies and, of course, e an - eld camera t e1 .._,.._..,..es, fil h , %k 1 Which is how, like it or not, m aest et1cs comes to rely 011 1 . . •es _ themselves dependent on research proniptcd b . n1ca1m . 'd , )' Lt tions which may be descnbed as 1 eo1og1cal. The sa.nic is tl'l dspth of field. motors e

Depth of field and the deep field we must be careful not to C_?nfuse wha~ is tcrn:-,c:d depth oJ (or use of depth in ~e s?anal field) Wl~ an mterpretati< space) with the field wh1ch 1s only deep_ relanv~ to the given ar Depth of field is a means of expression which appeared in with new lenses of short focal length, so-called ~~wide-an (16mm and 18mm), used for the first time and developed b cameraman Greg Toland in Orson Welles' Citizen Kane. A deep field is produced by 50mm lenses used in the cinen in photography, from the very beginning. Fixed to the most of box-brownies, they allowed focus to be maintained over a ' field", i.e. from 3 metres (subjects framed from head to through to infinity. If a technique is to be judged by its complexity or its e1 then the most extraordinary movement within a '~deep field' produced by Griffith in 1916 in Intolerance when Cyrus's a

entc:r Babylon driving the crowds before them. Yet it is ot that what is shown is a descriptive simultaneity. The char have no other connection than of being there and being inv 1 despite themselves, in a global event. Now., in Citizen Kane, ~e are being shown is not just two or more characters · simultaneously, but reacting differently to the same cause. However, though the space contained in this way corres or ~e~s to the visual field, its image in no way conf~r ormal vision. In fact, the wide-angle distorts perspective (rr tbe source of some significant effects) and focus over the field 15 · bar ' a..rt:1·fictal. If we focus on an object in front of us, it a] bs ,!'_~ut everything beyond it soft· if we shift our focus ac"l!,.1.vund · bc "d li , then this "'IIK:COmes sharp' whereas the o b)ect up cated" th Certa. rough the effect of our binocular vision. .

:ore

:,,

ut camera,nen tried to translate this peculiarity 01

Th~ Shor

77

A ;'deep 1ield": the grand entrance into Babylon in David W. Griffith'!! bitokrtm~. 1916.

into film. By altering focus gradually, distant objects at first soft became progressivdy sharper as the foreground objects became blurred - and vice versa (with no duplication, the camera being monocular). This technique also became fashionable (berwcen 1935 and 1940), but eventually excruciating. In poinr of fact, in reality, this perfectly natural process becomes more or less impcrcepti ble because of the chosen focus-point, whereas the image:, imposing on us both points at the same time, one of them blurred, the other sharp, gives it an annoying emphasis. On the other hand, a photograph composed with "depth of fieldn comprises a totality which the eye stes according to separate perceptual fields and presents an image u,~ifan11ly in focus. ln other words, we see a representation relating with the same sharpness objects which appear as near or far only by the effect of perspective. Which is the same as saying that the inaccurate and inre:rprctativc reproduction which maintains focus over the tota1 field presents a "correctcdu equivalent of normal perception where-as a consistent translation ends up being a false representation.

78

Semiotics and the Analysis of Film

. ed li'ke a torture on the audicnce,s eyes, this . Inflict. thankfu 11 Y di sappeared ··n itu . 1 blurriness and sharpness . n . television where electronic cameras are a•except an . f . da . uno1 . the instantaneity o v1sua1 accommo tion. 1 achJ eve · dep th o f fi e Id, th e effect is as tl an image registenng 5 ~e is related according to two separate s~ots imposed pah ther within the same frame - creating a kind eac o .er. • ·1 th < blishing a associative euect s1m1 ar to at of tnontai :tai:nontage within the shot, for though keeping focus whole field underlines the h~mogeneity of the spacc3 draws attention to the separanon of the areas which 111 (closer foreground/more distant background) and enal normally would have been shown in two successive st combined. The field and what lie1 outsz'de it

Remarks about what lies outside the field and the endh sions about the notions of continuity and discontinu mostly from epistemological platitudes. If the follo~ Pascal Bonitzcr is not a platitude, then I do not know wb The cinema utilises as much what it does not show does show [and) cinematic space is made up of a sp;i the-field and a space-outside-the-field, of what is seen is not seen.12 •

)

Through its limiu, the field dearly implies and define: outside it; but what lies outside it is not excluded. It ii incognira~ another world; simply uwhat is not seen", '1 be s~en in the succeeding shot, which is always there I 00 :side but to one side. And not neutral as a non-pre actmg more or less directly on the events in the fra through 1·nfiormation suggested by looks Hoff-screent' w acter · · effect on the behaviour of the " . may . ~, have a d ecis1vc . Withid~ tbe frame; or through noises which inform us unm e 1ate surround' 1927 B, 1 B 1. mgs and create a meaningful atm~ respect et: i~oazs ~s. ·alrea?y referring to uindirect v11 effects and nnation outside the field perceptible thr more meaningful than merely suggested, '

Semiotics and the Analysis of Film

80

..,...ber among the best-known examples: Edna' .. 01ninion the comp1etelY n akc d young woman s Ill~ Puhlic r .t "th OUt..0 bottom of the screen, w1 the movements of th th ate fh bd . e:rna hands moulding the shape o 'cher o Y. And, .in the s9-. ......,e f . the only part of wh 1 we see as ntght falls b · tram d. h . . C latfonn lights reflecte 1n t e carnage windows as it d P .station then again . as tt . p ulls away wt"th Edna leav· ra the ,. th . ~, Th •nE effectively "climbing onto e tram . c murder of th artist in Variety., which happens off-screen where what symbolic reasons) is the marriage-bed. The aeroplane she by Charlie Chaplin in Soldier Charlie> where we see Chru aim at a Fokker bi-plane., then follow the expression on l the direction of his eyes, as the aircraft falls from the sic, there are many more examples. As many in Chapun~s filtn~ the first from 1914 to exploit the suggestive power of of events) as in many others, for instance Renoir's Natta qt length by Noe] Burch, whose effects were already common1 1926. Nowadays they pass unnoticed. Burch points out that there are not only the four ofi possibilities on either side of the frame, but also the "fourt - face on to the camera, or the painter in the case of a p Howevert this other wall acts as such only in a limitei enabling what is uoff-screenu to be related to the field c The best known example in this respect is Velasque: Mtninas where, within the painting:11 the mirror reveals tl: looking at his children framed in the doorway. Mirrors cnab]e action placed in the axis of the earner included ~n this way~ the action becoming the image of an reflected in the mirror set in the field of view. A striking c of "~1!/0 n scrcenJ is in John Ford's The Prisoner of Shark ~ nuhtary guard is shown in close shot by the prison gate : ~to a. glass case the names of prisoners condemned to mcluding that of Doctor Mudd. Having disposed of che 1 pfapMer, tbe soldier closes the glass case in which the angui st 0 rs Mudd·19 " · field th suddenly reflected., introduced thereby 1 as ough "ea tu edi pb outside the fi Id .P r by it, while remaJnJng e ofv1ew reme,u

w:

7

J

Yet beyond

)

q







• •

the camera) wou~~rwn distance, the space where the s!'ecr: stand cannot be considered as "outside t of Vicw,n 1 . . t is the l . P P ace to which ·Someone looks., un

Semiotics and the Analysis of Film

82

rtain distance without which that someone look· ~e kcd at himself. Which is effectively a definition •~g bei ·~thout which the image would be incapable of be~ the m b . ~% Though its limits are not to e compared with the visual fie 1 frame behaves in the same way as our body, creating . l material distinction between the Self and the non-Self. •n ft!: · We might imagine a landscape painter wanting~ thro h . . 1· • Ug crazy impulse to push rea1ism to its 1m1t, to reproduce ev included within his visual field. He would realise that in him in the foreground, would be his painting on i;s would then have to include them in his painting. But ( painting repre~ented ~ithin his c~vas, _he would h~ve to ;et the canvas which he 1s represennng himself as painting. A the painting represented in his painting:,. another painting senring that painting. And so on:, in a regression to infinit the advertising posters on which a black worker is shown hol tin of cocoa on which a black worker is shown ..., etc. In conclusion we would agree with Noel Burch that obvious that any camera movement involves the space outsi field changing into the space of the frame, and vice versa,, 14 • What is also obvious but has been rather neglected by CI what corresponds at the temporal level to events hap outside the field. Ellipses:, for instance, are the equivalent "indirect vision,, described by Bela Balazs? As Pascal Bonitzer observed, it is clear that the space outs field is in no way imaginary in the sense that it cannot b rather it is imagined; the characters in the fil:m do not stop e when they leave the field of view, their existence being cap consideration only at the level of fiction. It is also ttUe tl "cinematic image is haunted by what is not found within i that what happens outside the field of view has a "quf uncertaintyj even angst, which confer on it with an em dramau· c power,, However, when critics refer to the ""mcot ness of the film image" we have to ask them to define their Whetber on film or not, all images are incomplete since tl only ever ~c Particular representation of a particular fragr :~ow~rld, inasmuch as, if something is incompiete in the fi . . wing shot completes it with an incompleteness comPl its tum, and so on in a continuity which can only ever be i cular look at a particular aspect of the world. It is also rruc

i

ea;

?

IIIIIM.c..

The Shot

83

the level of this continuity, the image, no longer con.,•d d · the context o f the represented space ·but "' ere as a trarnnent in e--. . . . as an e1ement in a signifying sequence, ensures its semantic autono . i • · h. h · h 1 my v1s-~-v1s the: meanmg w 1c 1t e ps to produce. Moreover, some clever minds would have it that the · _ . 'd th fi Id ·d · e1nema whether 1ns1 e e e or outs1 c tt - is a ~'mysti"'~--" ti ·11 . h h ..:,~ orcc. th Dou~tl~s _c 1 usion ere as a_ ~ore formal, more insidiously ureabst basis from the fact that 1t 1s a duplication of reality, yet because film shows us what seems. to be there but is there no longer, it confirms the absence of things behind their formal presence - and this is, must be, a snare and thlusion. Yet are those who would have it thus so naive as to believe that only film images arc illusory? 1 will not go so far as to cast doubt on the reality of the world but, as I have already suggested, this reality is only ever a darum of our senses picking up fragments of phenomena which our consciousness turns into a reality which is ucffectively real" for us. When we.lookup at the sky, we know {but overlook the fact) that no star. actually exists where we see it, with the exception of a few thousand whose light takes between ten years (Sirius) and four hundred (the Polar Star) to reach us. The others - hundreds of millions - are much further away. At the speed of 300,000 kilometres a second> it takes a hundred thousand years for light to reach us from the Milky Way and more than two million years to cross the distance separating us from the Andromeda galaxy near our own. Further ·out, the galaxies stack up at between 50 million and 15 trillion light years distance. When we observe the Andromcda nebula, therefore, we are not seeing it "as it is,, but as it was two miUion, two hundred thousand years ago. Where is it, how is it today? We have no idea; indeed we do not even know whether it still exists. We will know it so to speak - only in two million years, time. Thus when we look at the sky, what we are seeing is the past, a distant past. The vault of heaven is peopled with ghosts: a rather more consistent delusion than that of film!. .. But is it not also a delusion to extend the idea of the present to the whole universe? Theillusion is concealed in the hidden depths of reality.··

~ks ·to.camera As I mentioned above, "the space where the camera woul_d stand [~ {~].implies a certain distance without which someone )ookmg (tbc

84

Semiotics and the Analysis of Film

looked at".. This qfourth w a11 ., spectator) himself becomes . marks an unbndgeable separation between two , scene d " ~, . un1,

Communication cannot an must not exist, without the fiction by revealing it as such. ~cceptable, at a pinc~:l theatre _ in comic rath~r than ~tra1ght plays - the ~1aside dience" is intolerable m the cinema where nothing ll'lust au du~ ,, d' . I 1.or an au 1ence witnessing a sup 10 have been contrive true event rather than knowing that the story is purely ima! Even as early as 1909~ Frank Woods (one of the first~ critics) was writing in essence: A good director constantly asks his actors not to look camera., and good actors tty not to. Many of them succe~ is the simple fact of not looking at the camera enough? ~ there be a complete indifference to the presence of the can: By turning his face to the camera, the actor betrays tl that he is acting, shows there is someone in front of him, I from the audience, whom the actor is addressing. The ir sion of reality disappears instantly and the hypnotic il capturing the audience's spirit vanishes 15 &

This was the case with the Rigadin series ( 1909-1914), the actor> not bothering to preserve the impression of 1 instead emphasised the artificiality of the story by maki1 audience a kind of accomplice "let into the secretn. This kind of aside to the audience, still common in comii made before 1915, disappeared a long time ago. It ha:s replaced, however, by what nowadays is called - rather dun the look to camera, a look which demands various conditio Marc Vernet asks:

'

In what context is it possible to refer to looks to camen question begs a whole grey area. in evaluation of the direci looks ~~ the way they are filtned. The expression HI, cam~ 18 a poor one since it attempts to explain in tei filnmig an effect produced when the film is projected, v: flM~ator ha~ the impression that the character in the nart"E fioo tng straight at him in his seat in the cinema. So, thr, erent spaces arc b · • th tive uni emg lined up; the filming process, e verse, and the cinema auditorium. Nothing surprii

The Shor

85

that, one might say, since the "look to camera~' has . · th' 1· preascly the effect of prod uang 1s a 1grunent. Yet., cause can beco meexc confused wi'th e ffiect . Also.· d oes th'1s a 1·1gnment actually occur for the spectator? ept For the moment, let us confine ourselves to the film• "I k n ing process. T ~ a chieve a oo to camera the actor must look into the Jens without any other actor or. object coming between h Im • H d and the cainera. e must stan quue close to the camera so that it is· possible to see th~ direction of his look in the image; he must therefore be shot m close-up (or mid-shot) as front on as is possible. His look must also be focussed on the focal plane of the lens or camera (he must bring the ••rocusn of his eyes forward) in order to give the impression that he is concentrating on something or someone - otherwise his look appears emptyt with no counterpart or purpose, stuck in no-man's land. [... ] Moreover, it is worthy of note that what is called "look to camerau is the union of visual data (the look "inco the lens,') and sound data (a speech or comment). It is obviously very tempting to explain in terms of a look information deriving from the voice and sentence structure:, even gestures (an ann raised to the camera). Ii: is easy to sec how a parallel may be drawn between a look and a "voice-over", which is a purely vocal speech to the auclicnce 16• But looks such as this occur rarely in the cinema. They are much rnore frequent in television where, without any contrivance, a· narrator, in referring to events illustrated on the screen, speaks to listeners rather than to spectators. Such is the case with Alain Detaux whose look is noi focussed on the focal plane of the catnera, precisely in order to avoid giving the impression that he is looking·at someone in particular, instead anyone looking at him. In fact the only justifiable look to camera is when the aaor (n~t ne~ssarily in close-up) talks to the audience during the film in Or.tier -to poke fun at the film in which he is acting. Or when t~e .iubjc:ct is .the film in the process of being made (as in F~an~ms T..t:uffaut's La Nuit mnericaine [Day for Night})) where the director ati~\liis crew wish to give the impression that they belong to tbe ~:.world as the audience.J thereby dismissing the drama tbey arc ~t:t,l1~ ·pr~cess of filming as an ob~ous fiction. a.ins ·•~,re ts another look which d1rcctors are almost always at P •

,,

I•·

.

0

86

Semiotics and the Analysis of Film

to avoid, for the reason that they confuse it with th ra and that is the look towards the. camera i e e 10( ,ame , di ,1 hich . . ' · · to\\ra d ce "alongside the au en.ce , to w 1t 1s equa}( i 1 spak as to the space outs1'd e the fi e Id o f view · y POssi to the righ tor I , Ioo . . the charaetcr 1s ta1king to someo c th is instance, either . th . . f l . ne not and needs must give e impression o ook1ng at hirn w· focussing on the plane of the camera; or else he is simpl 11 at what is before his eyes, people, houses, landscapes soliloquises like Be~ondo in A Bout de souffle [Brearhle~s)-'. which, in a revoluuonary way, proved merely that looks du towards the camera were in no way, or were not necessarily, directed at the camera. In the case of singers singing directly into the camera (i example quoted by Jim Collinsll referred to by Marc Vernet)i American musicals, when "various shots are seen wher audience plays a narrative roleu it is no more than the cqui, of shot/reverse-shot. The camera takes the place of the auc instead of that of the other character in a two-handed scene.

Ywi

Technique and ideology

It is not the purpose of this study to be polemical and I absolutely no intention of downplaying the influence of idea ideologies - on the development of the cinema. But.t as I sai1 reference to perspective, this docs not mean that we ~ constantly be putting the horse. before the cart, at which Louis Comolli was a past-master when he was writing his tl1 in Le.s Cahiers du cinema 17• According to him, the reason the use udepth of field,, wa~ or less rejected between 1928 and 1942 was not techni acstbetic, but economic and ideological udictated by the in• and JJ?WCr of the ruling classesH. It 18 , obvious that technical considerations were not ei :pons,bJe, technicalities being subject, as anything else, stan ~ of any given society. However, it becomes nee 10 ~ 1 Peak plainly, to sort the chaff from the wheat with resJ c oFsc-ups, perspective and depth-of-field. or a start we sh uld • · ~'1 of depth-of-field,, .0 . ~o1nt out there has never been~ e· includ' th ' if this 1s meant in the sense of huge wid · ing c whole depth of space. There were bundri

The Shot

87

cxain.ples durthmg th~1..thi_rties. di~, .on the other hand, it is meant in con m.ons developed in Orson W ll t th-e s_ ~se ofa-, e •acsu:,enc I there was an u eclipse,,. . c ess Dl_8_,then> e.u.ccnve y., since the ma• •ry f bi befi Jon these condsnons were unrea 1sa e ore the development of sh ort 0 focal-length lenses:, i. e. before 194218 • • • The tn1th is that there. ~s a particular fashion during the years !928-40 to film the maJ0nty of scenes shot in the studio with lenses with wide apertur~s and, th_ereforc, limited depth of field. This bad to do firstly with the widespread use of panchromatic stoek from 1927-8. Insensitive to the blue end of the spectrum dils stock made it imperative to replace arc-lamps with incandes: cents whose spectru.m tended towards the yellow end and burned with intense heat. Since their power was less than that of arclainps, cameramen were forced: either to use more of them, which meant that filming had to be interrupted constantly and the lamps switched off in order to reduce the unbearable heat in the studio (make-up would start to melt...); or else to "open the lens upu to its. maximum in order to expose the film correctly wiih less available light. Which was the choice most often made. As a consequence of which, mid-shots were sharp against an out-of-focus bad;ground. Talkies also put an end to arc-lamps, because they had a tendency to "spark0 , making sound-recording impossible. However, from 1932, silent arcs burning yellow light (using ~gsten arcs) enabled normal conditions to be resumed. But cameramen - and directors ~ having become used to creating "hazy,, images and to use many more close-shots relying on editing; continued along this track ubecause it was the right way'•, ·and also because it allowed them to play with rhythm and shotfil_

..

r

-COntrasts.

_ •Thus there was no compulsion, except for a very short period, at tI,i~:Jeyel purely of technicality. And even less so, at the level. of i~~~9gy•.. The same is also true of close-ups. Why is it that (with

ve.qr:- ~-

exceptions) there arc no close-ups in any film before l~l2? There was no problem in bringing the camera ~n closer nor Of-'liting "portrait" lenses similar to those used by snll photographers~· ··What happened was that, as in the theatre1 char_accers had to_ be shown from head to foot. Henri Fcscourt, who due~ed for ~µ:ttlont from before 1914., relates in his book La F_0 • et l~s ~ttlgn&J that Leon Gaumont made him redo shots which aca-

s,miotics and the Analysis of Film

88

dentally framed only half an actor's body. Gaumont used to · him: '~ou should know that heads canJt walk b l ~ 19 y~ sclves....I" . It was much the same for every country 1n the world. Idea Just as it was in the first years of the century for shots {i 11 er according to different points of view,. it was a~ c eaeh Oth . ~uat1 culrural tradition,. ideology suggesting economic-political a • I irreJevant to the question. As with uacking-shots, the reasons and conditions governiIJ cinema's development are often unexpectedly much simpler . more comp]ic.atcd. Note relating to the close up It is odd to realise how far certain critics.) even the most e: enced, will sometimes go in following dubious misinterprcte For instance, in an article in the Belgian review Gros Philippe · Dubois quoting from some rather confused rema1 Sadoul on L' Arrivie d'un train en gare de La Ciotat, obser\'es

This train, which approaches from deep in the field of vie· moves without stopping through the whole space, quite li1 makes us feel the strange forces associated, not with dif sizes marking its journey like the compartm.ents of an 1 structure which it fills bit by bit, but with the constant tra1 from one to the other, with a release from the system, a er, over. Through its continuous movement, The Arrival of a bypasses .all articulatory structures, avoids the imposidoi "~~ of shot sizes,, seen as an a priori optical-technical s: wilh ns various degrees, its distinctive criteria, its pree5ta b patterns.

himButt ~s _is Precisely what depth-of-field is. Which then o mveagb against: A shot present.d d. . as such, clearly identifiable isolate , seJ discrethte 10 tbc linguistic sense of the word tllat is cut of e o e:rs by a b . ' . which b arner marking a distinctive frontier Y that fact sets . · trt system:. the .sy t ,, up a preestabbshed, stnct, s s em of a Scale of shot sizes n. C\,l

The. Shor

89

But this c 'sea.le': has ncv~r existed anywhere exc . ,1 _n'll..,,a.rs" cstabhshed during the thi · b ept in film ~... ---~ . mes Y theoreticians for want of classificanons. As I have said a hundud · d" · · th . . rimes before these arbttnll'Y 1snncaons> ese divisions into segment > ~ . s are on 1y a shorthand use d ,or converuence sake. There are a.,. h . · .. many s ots possible as there arc points between the lens and i... '"-"ty th . be:" k d b th . a.u1.u1 ; e r~1 clisdncnon ing mar e y e spaual margin .. in focus)• b vinue of the foeal length> the lighting of the subject and the size the aperture. Philippe Dubois goes on: p

J

If there was a genuinely distinctive effect created by Lumiere's film apparatus (and all historians seem to think there was) it was the ef/1ct of the dose.-up: impressing the audience with surprise~ or shock, or even terror, tO the extent of making them jump out of their seats and run away from this increasing, overwhelming, all-consuming image. To quote Sadoul again: uln The Arrival of a Train, the locomotive rushing from deep in the frame towards the audience made them jump bade for fear of being crushed'~. Now, there are no close-ups in the film in question. The closest the locomotive comes. is into medium-close-shot leaving frame on the left. The closest any of the passengers come is inro mid-shot. :what frightened audiences was not the shots getting progressively bigger but the movement of the locomotive getting progressively closer. And it was only the initial audiences which were alarmed. We must not forget that before the existence of the cinema, it was not possible to record movement - the genuine movement of real objects. There was indeed something novel and therefore sw-prising in their reproduction, particularly the setting of a tr~in. However, audiences soon became used ro the idea and the notion

that there was a r~- shooting. The signifying modalities exist in the interplay of lip.t: ·llll.d shade, in the relationships of volumes, lines, surfaces, w~ch the. -camera records according to a symmetrical or asymmctri~lattangement of a limited and constricted space. -~e, -information provided by the frame in general lakC$ on ~ i l l g . entirely from the context,. which means that iconic signifir:a,tlo#,s. work at their best when the image behaves as a separate a~~~ous unity, for instance in Expressionism where formal ~•~hips are emphasised by the permanence of the frame: In f~_:.~c:s~ become rather less evident in moving shots, but sin: ~J=ialtc;rati9n of the image always relates_ back to the ~ e W,AAt . :the audience feels is a transformation of the plasuc a ' ~Y..be ·in a less obvious sense, but no less definite. •

,Iii'

.



t'::s:

:--l~ted and representation dam · tall attached to its frame, Any ~n ll the screen gives the -~~=es ot Srrucb"-' basing his work partially on Kuleshov,s data th. nd ted Ii.• was ., . . • f A at E: stein developed bis theones o montage. s he wrote: isen.;.;

Pudovkin defends an understanding of montage as a 1. · A · ub ric . k s ,, . B r1c~s · arranged ink4ge • .0f Pieces. Into a cham. gam., . "th in S.Ctic 8 to expound an idea. I confironted h un w1 my view of rn · . ' th fro th 11' . • Ontag~ as a colhs1on. A view] T~ ~ _c : 1s:_n of two given facton arises a concept [... . e pob_mt ~s ) atf e coh~ulation (perhap• we had bct:er ~ay, the com manon o tw~ 1eroglyphs of th; simples series ts to be regarded not as their , sum, but as th e:11. f al th dim product; i.e. as the v ue o ano er ens1on, another degree each, separately, corresponds to an object, to a fact, but thei combination corresponds to a concept3. [translatorts note: In th last sentence Mitry paraphrases Eisenstein. A literal translaiio: of the French is as follows: The juxtaposition of two fragmcm of film resembles their product more than their sum in that tb result of this juxtaposition presents a quality (or meaning) ru produced by either of the fragments in isolation.)

Excessive generalisa,tions These ideas, soon generalised into theories by various critic took on the appearance of a rule of syntax according to which 1 association of any two shots A and B, necessarily involved, by del nition, a signification C. Which is fa]se, since the meaning implied by the association shots A and B essentially depends on the nature of the shots, 1 whal they represem1 according to a process which cannot reduced to an A/B = C formula Th~ Primary responsibility of montage is to ensure tbe filn narrattve flow; in which regard Pudovkin is correct as far gcnder~l Sttuctures are concerned (but not specific instances). Jf 1 or ermg of h . ·t ,.vo1 s ots endows each sequence with a meaI1lng 1 • ha not Ve organ·18ed d. . Jaung dfcrently, it is a dramatic meaning :re th e events on . bowei frequently the screen rather than connotations wb1c~, or :me phoricat Th Y may occur,. are not necessarily symbohc · ey may be confused feelings, vague impressions-

The Inferences

of M

omage

113 th e golden age of Soviet cinema I c _ . h b' > OJnpared thi.• . le sentences w ere su Ject verb and s effect With 51~1:iing only relative to. each other. Since then eo,n~Jemcnt ha\re nt d a much more obvious comparison b tw , semiologists have 1118 e and those of the different elements ine e~n rel~tionships of shots . h . a semantic once or twice t . e companson has assumed th strucrure, buut1·vslent of a word. Now, as we have seen th he shot to be tbe eq fr th l . ' e s ot is not . f meaning (apart om e c ose.-up). The mistak d a unu 0tein and the theoreticians who followed him fataellma~ e ~Y Eisens d • a-. ' Y ,ascmaced b , an its euect, was to concentrate excl ,· 1 ? th e dose-up . . "fi . . d . us1ve Y on this unamb1~odus s1grul catJonth, .an to interpret e,•ery shot, be ic close shot or w1 e-~ngfe, as ~o ing more than a global signified related to a single urut o meaning. At the level of the close-up it is certainly true chat che ass • _ · 'fier A an d a s1gn1 · "fier B brings about a signified 'ociaof tion of a s1gn1

Jn

I

I r

t

connotation C. This is true of the ''cream separator sequence'' in The General Line, where close-ups of the apparatus in action are cut against close-ups of peasants standing and staring in wonder• ment. It is equally true of the association made by a dose shot showing a piece of action and a detailed close-up of the same action, as is the case with the example ah..,·ays quoted from Tlrt! Bartleshi'p Potemkin. In close-up we sec a upince-ncz,, dangling by its braid from a steel hawser. What could this image mean, isolated from its context? Nothing, apart from being a pince-nez dangling by i~s braid from a steel hawser. Which is precisely what we see. Bu~, ,c · th hip's happens that the pince-nez belongs to a doctor Smimov, cs . SUrgeon. We have seen him toy with it throughout tbc prccedi~ . ha m to cbaractensc s:'4ucnces, to the point where this obJect s co. c h' b ha\'iour. hsm as forming part of his habits, his idiosyncra~ics, 15 e It has become a kind of indicator of his personality. M • . of the sat·1ors of oreovcr, we have just witnessed the muum been thro""11 "The Potemk.in", during which their officers ha:c by his fecr, overboard - among them Dr. Smirnov. Hungb P 1iung o\·erStabb , h ·use een ' . bo cd~ bundled up like a pa.reel, he a_s 1 w~ ha\'c seen hnn ard , t.n spite of his shouts of protestau_on: struggiing and losing his pince-nez in tbe r,ggJ~g. The pince-n~ u 1rntnediately, this image assumes a ,m,at:~ly, signifies his ,,represents., Dr. Smirnov or, more accuand rather conceinp~ absence,., • N o. th.mg remains . o f this arrogant

114

Semiotics and cl,~ Arralysis of Film

The cream !iepani.tor s.:quence m S.M. Eigcnstein's Th~ G,meral Lim:~ 11

tible officer than his ridiculous pince-nez foolishly dangli~g • 18 · precise en d °f a rope. The pan replaces the whole; but it t . . 'fi f mos ms1gm cant detail which reminds us o the chai inviting our contempt of him. There is more: through his p~ hi~ tank> Dr. Smirnov, a sampJe of the ruling class and pro~ a:1stocracy, "represents'~ that same class. So much sot: 11 Ptncc-nez comes to signify in a single moment the down n

The l,ifere11us. of A1ontag~ 115 . •i'c literally "thrown overboard.. The . rgelS , . th , . • re IS 110th • t,oll 1 , . symbolically, an a ridiculous !.Ubsta·,. : ing left or , c ass, • • • Ute 1m 1 · ih1S 1 sness and stupidity of what il represent" · P YUlg the . I ..,, worth es ·e have sa1d, t •~ spec'i.fic languagt ,,, r,,, _. A,s '" , . · 'J •< ~m~ma oi · . ,g w objects wath the obJects to r.vhid, n•e reJe B 4 u~,r:( .,, re/er-;;~ificd idea, the represented obiec1 is L'\"lnfi;~C'it{n\:lrnt~\'e!r the c as an object. Consequently, though rhe • , ~~ ~m instanc • ~ . •mast· \lgmfi • . h what it shows, 1t can ne\•cr ~have like- word. cs mroug . . . l · s uanspartm 1 ·m their signifieds:• t 1.e image 1s a wavs relati\•c to \\•h al 1t. S1lOWf, \\'l • And it is because tt acts as a _scrcc:n fo~ what exisls beyond th~ th~u n can a~surne a s1gn1 • 'fi1came.... ning of the represented • . fi obJects . . 0 outside its own s1gt11 cation. 1n lhe abo\'c example -1 · no .eh . h . "fi , 1 ~ not the pince-nez wht 1s ~ e s1gn1 er but the relationship of chat object with the. preccch~g shot. The image of the pin1;c--nc-z assumes the quahty of a sign only because we imui1i,·cl\' obicaif,., in it the idea suggested by the relationship of which it i!.'onc ()f th~ elements containing the greatest potential signitlcadon. Thi1- is how it comes to behave like a sign or a symbol. And ye1 1 thi~ smacturc, containing the smallest possible syntagma (two shots joim-d by an associative structure) is a signifier only r.i1itlrin rlrf i:omt.u of the film. By itself it is devoid of meaning. Somc:onc with no knowledge of Eisenstein,s film would find no meaning in it iU all. The: sequence: officer thrown overboard/smoking chimnl!y,itlag flapping in the wind, etc. would not tell him any more or any less. In ordc-r to decipher this association (for there to be an association in the first place), he must see all the previous sequem;es, for it is these alone which give the object a meaning by showing 1hc pinc..:-nr.:z a'!> specificaUy belonging to Dr. Smirnov. It is thi-s "current. llf meaning" :i1 carried by the film, which turns these two succesl\L\'e ~hots into a coherent entity. Without which the}' arc mcrc_h' two !ndividual shots arbitrarily brought together: there is no dasL'l!~nibJc connotation (or rather there is every poi.siblc connoranon depending on the taste of each indi\ idual \•iewcr). h i" tllU\.. ibt COnt • ·h gi , \"S •hem . d1e ext which determines these structures, w h.ic c. .. ._ • nn. t· which U1~, t'VWer to signify by loading them with a meamng or_ . . . ·_ act ' .dcnUl\.. lt') ~s.SO . ~ agents. In isolation, the syntagma lost--s tts 1 '· Ctativc mearu . di 'Wb ng sappears. . br:\'l..,Jld and '"d ercas in linguistics the syncagma has meaning · alter or .... CJ>end • • mar nor 01 1

eom 1 ent of the context which may · a bC\·on.J thi: :P etc ·it:i1 there is no such thing as n film syntagttl ·

116

Semiotics and the Analysis of Film

---·' enti"' through which the syntagmas are fonned

8I l ea~ ~J' • f • di . cttve structure, thc th WOl'ds, by v1nue o, its pre oer . 1 linguistic syntagma con~m_s ~n inte~ ~ssociation Which

meaniniii whereas the s1gmfymg association of the fi.1.Jn 81 (making it what it is) almost always comes from outside. · ... and excessivs signifiers

Thus, though the unamln"guous signified is (or may consequence of the rc1ationship between a medium she close-up (the relationship of wide shots involving a nu1 different signijieds), this meaning coul~ never be containc◄ a succession of close-ups each possessing a more or less s quality; without the need to use intermediate descriptiv The 4 •Gods sequenccu in Eisensteints October is significan respect. Showing successively a Baroque Christ> then the ancier Gods, Hindu, Mexican, African, etc., Eisenstein claimed was 44 discrediting the idea of Godn and, by extension, th◄ of die concept. According to him:

While idea and image appear to accord completely in statue shown> the two elem.en.ts move further from ea1 with each successive- image. Maintaining the denot: "God~\ the images increasingly disagree with our co: God; inevitably leading to individual conclusions about nature of all deities". His analysis has mc.rit5 but it is one. made after the fa, cditiog~bench. Films, however, are made to be seen, and immediately decipherable - for the most part at least. · they appear in the film~ these divinities tell us nothing of 1 They are perceived as a collection of ornaments or statu re_~ds any connoted values!! these are as~ociated in O\l wt~n the same global concept: the idea of religious c beliefs; and ridicule of these beliefs and divinities. The spectiuo.r may wen recognise these as distinctive - eJ denoted · · the u . - si~, he can never be forced to see them 11s 51 discrediting of the idea of God,, In 0 ther words, the same signification ··• is ascribed to cac 1

The Inferences of Momage 117

. The whole is perceived as r,oneatin th •,11n1fiet"S· d"ffi · -,.. g t: sam rJtc s!rr· under various _ 1 ~nt gu~scs. e c:sn1oi in this is using unages like words without reali . 121 once agae~arate signifiers relating to a meaning an a sing ';hat --' arc s r • PPropnatc ",oivs bereas images, re1atmg to a concrete reality .-.~A r w d d ' .... .., sepru-ace C()flceP ~eh Jevel of the enotc and not at the Je1,•cJ of . 1 al e eh • a \ a ue I oo Y t,earin3 on e concept to which the denoted ...t'. . dnfflellt .1-:( . I ) . h 1u ,r~ nnote no1.u.1.ug dor very .1nt e m t .cmscJ1,•cs s· It ,·srcLers. 1 esco g Ims . ~i.is instance:, the enoted 1s a symbolic rcprcsentatfon cear .A ha( ,n µ1, E" . W)u1 c co~otative value. But 1senstem assembles a series of readymad~ a bols which, taken together1 suggest no more than each one srrn1·fies in isolation. Thus, though the global signified 9S$0Ciating s,gn sinl"'lifiers - th e nonon · · periec;..-rJy r o f 0 Godn· - 1s well defined th cs e oa~ ·n d . d"fti ~ rhc separate si~ e assuming a. J er~~ value in each is not. This means that instead of what Eisenstem 1ncended - separation _ we have comparist:m. His "dynamisation of the subject'' turns these into a series of abstract signs whose meaning, confined •.vir.hin the representation, becomes stiff and lifeless (if it does not di.§.appcar ..icogether behind it). When the principles of HmontageH were first applied, one or two film-makers went too far 3 on the basis that,. once a rule has been established (this is a well-known fact)~ it is used to explain and resolve everything by relating everything to it. 1"hus, a "montage released from arbitrarily selected attractions, independent of the actual action" should have made it possible to signify by means of symbolic or metaphorical effects. Now, Eisenstein's theatrical experience already proved to him the difficulties and limitations of SUch a method. In the cinema the main objection is thac such 3 technique is valid only inasmuch as it uses Jiving elements (in tbe drlllllatic sense of the word) from which it takes ir.s emotional power at the same time as a concrete symbolic signification. It :op~ being Valid when it acts with symbols arbitrari{\' seleaed ~d . . . li ed b.) 1·r · Eisenscem f PPhed to · reat·tty instead of being 1mp rcqucnt1y zo. Jl • • · · h" disc::o\"crr, cndin .1~ mto this trap by over-systemaosm~ 15 , 0 eJcll'lcg u~ in a kind of abstract formalism, forcang bis chose In hn.ts into an ill-fitting suait-jacket. ~....is first fihn S . 'k . metal founl,U,' Pl.lt dow tnke::1 which traces a stn e m a h ,wing 5 c~~d in Workers : . by the Tzar!Jos soldiersJ he contra st~ an. abatt . emg shot down with shots of a bull bemg. ut . s.ignif01r• ......._e --. • • . '- though tJie idea 1:s .., n o;uect IS unpreSStVC, uUt

sht

Semiotics and the Analysis of Film

118

ied the narrative logic is somewhat distorted In d , . th l h . or er- t symbolic expression., e s aug ter-house scenes included in an event totally alien to it. are ' One might argue that the scenes in the slaughter-h . 1y 1s . m . th ~ conce1vab · e same area as the .&.actory) are a ou those in the factory, that the plight of the workers is : r and purposes the same as the bull's and that3 howev-~ · the metaphorical association, it is still manifestly justifi:. as it may, it has no shock value. On the other hand, in October, where he sketches out . on he would term cinedialectics, Eisenstein became pr exclusively with the 'idea'., and frequently overlooked, authenticity of the events, at least the logic of their linkit association. Such is the case with the attack on the Winter Palace i second Soviet Congress, where the Mensheviks' sp compromise are intercut with shots of women playing 11 idea (literary in the extreme) is to give the Menshevik 11p tone and appearance of lyrical whining., mind-numbing b However, though the idea is valid (if perhaps a trifle fc are left wondering what the harps and harpists are doi~ into the objective, concrete reality of the meeting. In the same film, just as Kornilov prepares to march grad at the head of the White Army:i an image shows t sining arrogantly and forcefully (one might say 'NapoJ on his horse. The impression is· reinforced by the angle i - an upward tilt. Yet Eisenstein does not stop at thjs; he an image revealingj from the same angle;, an equestx which confirms and reinforces the ludicrousness of th AU well and good. But what is this statue doing in tJ logic of the action and, more particularly., on the Russiar: W"f!,Isewhere, we see Kerensky in one of the huge ~ . Ulter Palace. He rehearses a speech~ walking to ~d bu anns about and looking at himself in various mirro the shots revealing the tiny man lost in the vast proport ::dd_s to the impression of overwhelming solitud~ s -th him. However, on one of the mantle-pieces in i:tan~ a sta tue, a bust of NaPoleon. And predictal arro ~w~ to contrast· it with- different shots of ~cte~

-u:::

pn Y m front of the mirrors. This -association ,

The Inferences of MomQ8e .

f

119

i,ecause the statue 1s presenr - Particularly . peatedly passed backwards and forwards in tncc lI. apmcs nolbing without a point of reference. In this ordtt of idea, the simplest propositional fonn.s, such u P11n iJ hil, quid(y, and even aborter propositiom. such as r1tt aor ~ clodta,o,-/f nau down may be considered II the 11D11lat 'Yntagma. Dill Before appropriat~ the cinema, were 111 fnlsmcnts conta1n10g shots relaang to a moment . .,d used U> called sunu. Tb.is term deriving tbc th~ - r k judged ~ b e the laid rather than the .ra.>'ffll ~uatc for the purpoacs of scructUral anal~ grouJ)I of mat ~~ the: notion of sc,nou, such u ~ • ~ TbdC mo< fre_,,Qllna a aianffying link.age, like lingWSDC umlSQlri._n ~ tiy. ~ t a ha\Tc alao been termed syntap,01 bJ wtucb {«- plGPO" . ; : : of thcii- •imilarity with the word grouP' -"' 1)111JCD" arr 1 {altboup this would unply lhal sctJ)C ,od die - one lll1d the same. the one rclatina to fona

A~_.,

me-~:

sesnio1?~

fron.1

was~~;;-~-



132

Semiotics and the Analysis of Film

11 con'"•ent, with any scene .capable of being broken down in•o one 0

several syntagntas, just like a phrase).

r

Main objections An initial obj~~n, however, ~~es to ~ d : howev~r much it mav necessarily signify> the assooattve organisation of shots is " · ·1ar _to .th ose contro11·mg the associationnot subject to any rules s1m1 of words. It is similar to the linking of phrases whose order derives from the logic of the story. The shot, moreover (which is the smallest segment of film), corresponds not to one but to several phrases. Which means that the film syntagma is in fact a .syntagma of synragmas. lt is no longer a group of units of signification (words) but complex signifiers (shots) considered both as units insofar as they are syntagmatio, components and as utterances in reference to the signified. Obviously, the structure of the syntagma remains the same, but the way it behav~ is different, especially since, in the shot, the signified does not depend on the linear succession of elements which are separate and capable of being broken down, but on a dynamic and indivisible whole, a global movement occurring in a limited space according to a predetermined time sequence. Moreover, if there are no rules governing the linking of shots, there is no such thing as a film synragma except for the causal entity which gives its meaning and makes it what it is .. If we take, for instance, the cell fonned by two shots in The Battleship Potemkin: a. the sailors throwing an officer overboard; b. a pincencz dangling at the end of a steel hawser. . . it is clear that this cell (which is the smallest possible syntagma: two shots joined by an associative relationship) carries meaning only in the context of the film.

~ we. saw in the previous chapter, this relationship, considere~ out11de its associative context, signifies no more than what it shows - no connotation can be read into it. . Wher~ the linguistic syntagma possesses an internal assocla?on its gfammaticaI organisation, the association which gives e&nlng to the film syntagma (which is what makes it a syntQglllS,), ash: ~ve said, almost always has an external origin - but one ~ ~o~ of the internal structure of the film, its core, ratbet nguig to an external body outside the film itself.

:rou~

. ?

Concerning Symn,....,H . "15rnat1cs . . . 133 The secon d o b 1ect1on is as follows· Ch - . . •bes two basic groups within the nitian Metz disringt11S h' h ......nt synra . one of w 1c , more or Jess codified . gtnat1c forms eh ea . f · . h . , involves th , ducuon o meaning. a-c ronolog,ca/ synta,grn e same pro as and chri 1 . l synragmas. -ono ogi. ea The first group is divided into parallel s . d . ;ymag,nas (gover . vents which are contraste JUX:taposed or comp· ed) nmg : ; ,10 [translated as ~bracket, by Michael Taylor] ~:nbra"b • d .th. th 'J ,,,as ~vYcrn ing events as~o~1ate di w1 . ~ e same global concepruaJisation) • The seco~ ml tob eshcri~tt-vethsyntagma.s ~governing events co~sidercd succc~s1ve y ut avmg e same ex1s_te_nce in space and time) and narrative syntag~, themselv:s subdivided into linear narrative syntagmas (~overnmg consecutive events in a series of autonomous shots, episodes or sequences) and alternating {translated as 'alternate' by Michael Taylor] syntagmas (gO\'eming the simultaneity of separare events). All this is true. But., without wishing to doubt the importance of Christian Metz' work, we might observe that:

di.ff.-...

j,

I.

I

I

S":na- 311

I. Alternating syntagmas govern non-simultaneity as well as simultaneity. It is possible to use these to link two events separated in space but occurring at the same time or two events occurring in the same space but separated in time. The alternation: Here I scttrewh~re else / here / somewhere else a1so presupposes the alternation: today I yesterday/ todaJ' /yesterday .•. 2. Parallel~ comparative, or associative syntagmas also derive from the same form of montage, variously described by film technicians as alternating montage, parallel montage or cross-cuuing. The technique is one of making, shots alternate according to a A/B/ ~/B/A/B ... , etc. sequence. When there is a juxtaposition of ev~ts ~1milar or with the same meaning (simultaneous or 0th e~·is~}, interacting at the level of the diegcsis, there is an asSoa~tifJe synragma. On the other hand when there is juxtaposition of differ, . · il ·zy then ent events but showing a factual or ideological still ~ 1 tbere is a comparative syntagma. Thus, in Storm (}1)er ASia: ~onks clean the statue of Buddha. The English officer shaves. he statue is polished. ~:e 0 fficer>s wife powders her face. Tb e Statue is decorated with various adornments. klace. e officer's wife does up the clasp on her pearl nee

134

Semiotfos mid the A,ialysis of Film

The: officer puts on his uniform decorated with his medals. The statue is dressed, etc. 3. In the same way, descriptive s~tagmas "?d narrati.,..e syn mos (the difference between them 1s superficial) derive from same type of /i,iear montage A.B.C.D.E., e~c. The hurried, brc nature of so-called non-linear montage derives entirely (in this of syncagma) from the elements bei"! contr';"Sted in this formal 1~. position. In other wordsjl what differentiates these syntagnia their content~ the nature and character of what is represen rather than strictly codified organising principles. It is cle possible to relate these forms to various large structures, as Cl tian Metz has proved~ but to those described above (which are most frequent) many others can be added. Specifically: false alternation such as Bclmondo"s escape Pierroz le Fou, shown in various different aspects, when he has yet left. Imaginary alrernation, like the sequence in Midnight Coi where we see: a. the young man in his room writing a letter tc uncle; b. the uncle on his ranch receiving and reading the lettc the young man still writing his letter, then tearing it up 1 thinking about it. Dual alternation where present and past both become absor combined, confused within each other, confirm or contradict 1 other, as in Lenny or Star 80; where the actual facts (eyewit reports, interviews, investigations) do not serve as a commen on a completed action constituting what actually happens in drama, but constantly reactivatcs it. The list of example endless:, but this structural schematisation, convenient at the l of analytical study, is valueless from the moment an attem1 made to draw general rules or de.finite codifications. Moreover, these cJassifications, relevant only in signifier/sb ie~ relationships, at no point bear on the rhythm which, as we l said,. contro!s relationships of time. Now, though rhythm ma~ p~ovidc an mtelligiblc meaning to the syntagma, it does pro~11 with an emotional value not to be underestimated, film bein much ex~ression as signification. Yet rhythm also ex:ists by ~ of what u turned into rhythm~ and not as is the case in musi poetry, b Y virtue · ' of pre.established structures or rules. ~e may say, in a general sense that im11lication is one of basic • • ' jf notr the most 1mpor . condinons of film eXpress1on,

,

Concerning SyntagmaticJ

135

. . n is not a closed system but a generath• f . ·b'J' s·1m1·1ar to semantic indu,....· . e U11ct1ot1 Ilfl.P1icat1° Jy open to poss1 1 n:y. .. .. 8 rant . l'k . , .. .ion, tt 11, co~~ t 1•0 that> ,ust 1 e semantics itself, it is Jimircd b . •tfcreO . th . synrax, d1 d' ·oned by 1t, whereas c soJe base-s of •mplication} .ar h; con iu . 1 d th . . e t c. . 1 psycholog1ca an o er prancrplcs which we inr.end ~~, ro examine later on. . . . Be this as it may, •~ 1s not 1mporrant that che rc-suhing significa. should be obtamed through montage (alternating1 paran .. z uons . . b .. t comparative, c':'ntrasung., 1mear, rokcn, ere.) or by its absence (panning, tracking, cam~ra moveme?ts, depth-of~field, (..°tc.,>. since! the different syntagmatic forms exist only br virtue of whar is howrt and not because of some Jaw or other. s "Alternating montage tells us nothing about whac shoujd be puc into the images/' Christian Metz tells us 2 • Now, in my \'iew _ which the evidence seems to support - it is precisely whar js put into the images which governs their structure. Only obje