SIDE 14 / DANISH FILM INSTITUTE / PRESERVE THEN SHOW
The Raw Material of Film History
CASPER TYBJERG
In the following, I shall try to offer some reflections on film preservation from the
perspective of a film historian. When I speak of “The Raw Material of Film History”, it
may perhaps seem obvious that I am referring to films. Yet film historians may draw
on many different kinds of source material when composing their accounts, and some
do not use films at all. In their book Film History: Theory and Practice (1985) Robert
C. Allen and Douglas Gomery write: “For certain investigations, film viewing is really
an inappropriate research method ...”1
One may well ask: Why inappropriate? Allen and Gomery indicate a number of
problems with film viewing as a research method. It keeps many kinds of research
questions hidden from view, involving social issues and economic issues. Distribution
and exhibition are important areas of study where looking at films may not provide
the researcher with any information he or she needs. Furthermore, there is the
suggestion in Allen and Gomery’s book that it is hard to maintain the necessary
scholarly rigor if one’s attention is distracted by the fleeting and glamorous images on
the movie screen. Film viewing, perhaps, is too much fun.
Allen and Gomery are very concerned to make film history more rigorous, but this
should not rule out trying to find ways of giving sense to film viewing as a research
method. This is often called “placing the films themselves” at the centre of research. To
exemplify this, I shall briefly describe two approaches that take “the films themselves”
as their object of research: constructivism and neo-formalist historical poetics.
But this gives rise to a new question - what exactly are “the films themselves”?
How stable is the identity of a given film? When we speak of a particular title, can we
be certain that we refer to the same thing? There are different ways of questioning
the films themselves. More emphasis can be placed on the fact that many films exist
in differing versions and variants. This will raise interest in learning the reasons for the
choices that have to be made between different versions, making manifest the
importance of critical documentation.2 Carefully documented critical reconstructions
may help to demonstrate the complex identity of many films. Researchers may begin
to examine the film as a material object, looking at the history of individual prints.
Questions of provenance and survival lead to a consideration of the history and politics
of archiving, and more widely, the place of films in collective memory.
As an approach to film research, the label “constructivist” has been used by a group
of German scholars to identify their position. In a programmatic article from 1989,
Werner Faulstich argued that the basis of film history must be the analysis of individual
films: “The writing of history as a reality-founded construct must in my view proceed
from what is given rather than filter and colour it through historicist-hermeneutical
PHILOSOPHY
Preserve then show 2.korr. 9/16/02 2:13 PM Side 14
or radical-contructivist glasses. And the one unquestionable given is for me ... the films
themselves.”3 When Faulstich and his colleague Helmut Korte call their position
“constructivist,” what they are referring to is what they regard as the constructed
nature of all large-scale historical concepts. They reject the notions most film histories
use to structure their accounts: nationality, periodization, movements, -isms, and so
on. These concepts force the films into a pre-existing interpretive framework, which
is why they call the use of such concepts “historicist-hermeneutical.” While they deny
that such concepts have any real existence, they do not adopt the “radicalcontructivist”
position of saying that everything is made up, either.
Faulstich and Korte applied their principles as editors of the five-volume Fischer
Filmgeschichte (1990-95). In their general preface, they write: “Film history to us
means first and foremost: the history of individual films, that is, product history.”4
The work as a whole consists of a collection of analyses of individual films, one for
each year (more or less) of the history of the cinema. Strikingly, there is rarely any
mention of the material used for the analyses, and where it appears, it is generally on
the lines of “According to the video edition issued by Atlas Film.”5 While there is an
insistence on the work as a given, the issue of whether the work can be
unproblematically represented by a videotape is not discussed.
By and large, these issues are not addressed either by the American scholars whose
work have been grouped together under the designation “historical poetics.” The term
was introduced by David Bordwell to describe an alternative to interpretive criticism
that studies “how, in determinate circumstances, films are put together, serve specific
functions, and achieve specific effects.”6 The approach has been ably summarized in
an article by Henry Jenkins for the anthology Approaches to Popular Film (1995). A
central point, he writes, is the belief that “historical explanations must start with the
work itself and move gradually towards its most immediate contexts rather than adopt
global or transhistorical theories.”7 Where the analyses of the constructionists tend
to focus on narrative structure, historical poetics are more preoccupied with stylistic
devices and their practical functioning, seeking to get away from interpretive practices
that seek to assign meanings to them.
When it comes to the understanding of the object itself, the clearest formulations
may be found in the work of film-philosopher Noël Carroll, whose work shares many
of the same objectives as historical poetics. He writes in an essay entitled “Ontology
of the Moving Image”:
The film performance is generated from a template – standardly a film print, but
it might also be a videotape or a laser disk or a computer program. These
templates are tokens; each one of them can be destroyed and each one can be
assigned a spatial location. But the film – say Broken Blossoms by D. W. Griffith
– is not destroyed when any of the prints are destroyed, including the negative
or master. Indeed, all the prints can be destroyed and the film will survive if a
laser disk does, or if a collection of photos of all the frames does ...8
Film historians, even those whose explicit goal has been to place the “films themselves”
at the centre of their research, have largely overlooked the problems posed by the
PRESERVE THEN SHOW / DANISH FILM INSTITUTE / SIDE 15CASPER TYBJERG
Preserve then show 2.korr. 9/16/02 2:13 PM Side 15
SIDE 16 / DANISH FILM INSTITUTE / PRESERVE THEN SHOW
materiality of films. A film exists in the form of prints; these may differ from each other
in a number of ways. There may be quantitative differences: multiple negatives shot from
different cameras (a common practice in the silent period), censorship cuts, and differing
versions for different markets where both images and intertitles may vary; there can
also be qualitative differences: genealogical distance from original, base (nitrate, acetate,
etc.), format (16 mm, 35 mm, etc.), colouring, chemical decay, and wear and tear.
A few scholars with a background in archiving have assiduously maintained the
importance of these differences between individual prints of the “same” film, none
more so than Paolo Cherchi Usai. In his Silent Cinema: An Introduction, the most
important points are presented as “rules”, set off from the main text in black-bordered
boxes. Rule 9 runs as follows:
Every print of a film is a unique object, with its own physical and aesthetic
characteristics, and should not be treated as identical to other prints of the
same title. 9
He writes this in the context of viewing notes, insisting that any conscientious scholar
must take care to indicate exactly which print of a given film he or she has seen.
Cherchi Usai emphasizes the differences between prints and, indeed, between viewing
situations. In his polemical and aphoristic little book, The Death of Cinema, he argues
that each individual screening of a film is essentially different from every other, and
that they should not be regarded as any more comparable than performances of
musical works:
The nature of the light source, the apparatus, the physical structure of the
image carrier ... and the physical structure in which the event occurs are
variables which have the power to determine the quality of visual perception
and its patterns. Given their huge number, each with its vast potential for
mutation ..., preservation of the moving image ought to be treated as an
equivalent to musical performance. As with the moving image, each aural
experience is in fact a unique event; both are subject to variations due to their
transfer to another object (musical score and matrix), their display (concert
and screening), their repetition (sound recording and viewing print).10
This is a radical position, which threatens to dissolve the “films themselves” into an
effectively infinite number of unique objects and events.
Still, Cherchi Usai’s combative pronouncements force us to consider the issues with
care. They certainly force those film historians who work complacently with videotapes
to think about whether they are not missing something crucial. They make plain the
importance of original materials, and they serve to keep the aesthetic dimensions of film
viewing in firmly sight, even in the context of the most rigorous scholarly investigations.
Another example of the passionate engagement with nitrate originals can be found
in the article “Nitrate Take Care” by Mark-Paul Meyer, who is equally insistent on their
vital historical importance: “In the case of a duplicate you are one generation further
away from the original.
PHILOSOPHY
Hidden behind the curtain of time?
(From “Atlantis” by August Blom,
1913). Photo: Nordisk Film/DFI
Preserve then show 2.korr. 9/16/02 2:13 PM Side 16
It is as if a curtain had been interposed between the original and the new print,
the curtain of time ... this curtain of time will always be noticable to anyone
who is sensitive to a direct relationship with the so-called pro-filmic, the people
and objects in front of the camera, as if you could almost touch and feel
them.11
Some people may regard this as fetishistic, and perhaps it is. Nevertheless, I believe
that it is an attitude that should be cultivated, not least in students. There certainly
are numerous film scholars to whom the importance of working with original materials
and sources is so obvious that it barely seems worth mentioning; but in my experience,
there are many places where this is insufficiently appreciated. I think much more can
be done to expose aspiring film historians to the lure of the archive.
The fascination of working with original materials should not be underestimated. I
remember how striking it was to see a roll of tinted nitrate film for the first time,
looking not black, but strangely multi-coloured, the tints having dyed the whole width
of the film strip, including the edges. The past becomes palpable.
Perhaps it is the strength of the fascination that the object can exert that has led
preservationists like Paolo Cherchi Usai to insist on the uniqueness of the individual
print, as he does with his Rule 9. Taken to its logical extreme, this leads to the
conclusion that each print is given the status of an independent artwork, like a painting.
That, however, represents a denial of the mechanical reproducibility of film and renders
any restoration efforts effectively futile, because duplication, copying, is necessarily
involved in them.
With paintings and suchlike, the artwork resides in a single, unique object; this is
not the case with film, as even Cherchi Usai admits.12 For that reason, even though
film restorers may find that there is much useful inspiration to be had from the field
of art restoration, it provides an imperfect analogue. A more apt one, I would like to
suggest, may perhaps be found in the field of codicology (other suggestions may be
found elsewhere in this volume). Codicology is a historical subdiscipline, a technical
tool, like palaeography (the study of ancient handwriting). It is defined, in a recent
introduction to the historian’s craft, as “the study of handwritten books as
archaeological objects, including the study of the materials (parchment, paper), of the
bindings, of cataloguing, and of preservation problems for manuscripts.”13 Another
formulation states that codicology “concerns itself with the technical preparation of
tools and the writing surface”14 including the way codices were made and the layout
of their pages.
The codicologist, accordingly, studies objects, but whereas the object the art
restorer deals with is the work of art, the codicologist deals with a carrier. It may of
course be argued that the film’s visual qualities are crucially dependent on the
properties of the print and cannot be faithfully copied, while a text found in an ancient
codex remains the same work if copied accurately. While there is some truth in this,
the process of copying is integral to the very nature of film. For instance, if an archive
possesses the original negative of a film and strikes a new print from it, it seems highly
unreasonable to claim that the new print is not the same work. I think the analogy
with codicology help us keep two things in mind: on the one hand, that the existence
PRESERVE THEN SHOW / DANISH FILM INSTITUTE / SIDE 17CASPER TYBJERG
Preserve then show 2.korr. 9/16/02 2:13 PM Side 17
SIDE 18 / DANISH FILM INSTITUTE / PRESERVE THEN SHOW
of the film artwork is not tied to a particular carrier; and on the other hand, that the
study of the carriers requires important technical skills that are not necessarily closely
tied to the skills the film historian uses in the study of the works.
Splitting these two sets of skills apart may well be regarded as deplorable and
unwise. Certainly, the most eminent theorists of art restoration insist that restoration
is (also) an act of art appreciation which must be grounded in a thorough art-historical
knowledge of the work, not (just) of the object and its properties. In practice, though,
the split is often a fact, even in the field of art restoration. In some places (like
Denmark), the technical skills required in conservation work are taught at institutions
completely separate from the university departments where art-historical scholars are
taught. This entails a number of obvious problems. Codicology, however, is not a
completely independent discipline but a technical tool, a domain of specialized
knowledge. It is connected to historical source criticism and especially to textual
criticism, understood as “the technique of restoring texts as nearly as possible to their
original form.”15 Textual criticism, with its tracing of genealogies and lines of descent,
its efforts to establish the relationships between different and often incomplete copies
of the same work, is as important for the film restorer as art restoration theory. And
codicology can be described as the branch of textual criticism that deals with the
carrier, the physical object, and therefore provides a good analogy for the activities
of those who work directly with old prints. As the codicologist deals with the “external
characteristics of the written heritage,”16 so the film preservationist deals with the
external characteristics of the cinematographic heritage.
Encouraging close attention to the artifact and thinking about what status it should
have is likely to awaken an interest in the handiwork of restorers. This is why careful
documentation is so important; it allows later restorers and historians not only to
learn what sort of technical work was carried out, but also to get an understanding
of the theories, convictions, even prejudices that shaped the restoration. Indeed, we
may discover that the methods and protocols of film restorers can have significant
consequences for the way supposedly “given” artworks become available for study.
The print is a cultural artifact which has a history of its own. As Paolo Cherchi Usai
writes: “The silent film we are about to watch is not an abstract entity... It is the
survivor of a complex, often random process of selection, not much different from a
Darwinian evolutionary scheme.”17 While random forces are perhaps the most
important, those that reflect deliberate choices have understandably attracted much
more attention. Recently, there has been considerable interest in the issue of collective
memory, “the ways in which present concerns determine what part of the past we
remember and how we remember it.”18 The reasons why things - including films - are
kept and shown are part of this. Many accounts emphasize how such present concerns
often lead to fierce struggles between groups with different opinions of what should
be remembered and what should be forgotten.19
A highly provocative discussion of these issues can be found in Cultural Selection
(1996) by Gary Taylor, one of the editors of the Oxford Shakespeare. As his title’s
play on “natural selection” suggests, Taylor espouses an insouciant cultural Darwinism.
He freely makes use of the vocabulary of evolutionary biology, speaking of “cultural
PHILOSOPHY
Preserve then show 2.korr. 9/16/02 2:13 PM Side 18
niches” and referring to artworks as “stimuli.” Taylor emphasizes the power of editors,
a term he uses to encompass all those whose work affects the survival of texts and
artworks, including restorers, curators, and so on. Their activities, in his view, reveal
political motives and have political consequences. Discussing the status of Griffith’s Birth
of a Nation and Shakespeare’s Othello, he writes: “...those stimuli could only be created
in a racist cultural niche; they could be perpetuated in collective memory only by editors,
by mediums, who were at least tolerant of racist representations of the world.”20
This attitude of suspicion and mistrust, determined to see unsavoury hidden
motives behind apparently benevolent acts, and while we may often think our
predecessors made mistaken choices, it is too easy to paint them as villains, because
that tends to let ourselves off the hook. Rather, we should approach their work
respectfully, with the assumption that they were animated by motives similar to those
that we believe ourselves to have. Only that way can we begin to learn from their
mistakes; and while we are probably as blind to our own prejudices as they were to
theirs, perhaps we can try to make them more visible to our successors and help
them undo the damage we cause.
For an interesting and illustrative example, I would like to turn to art history. In 2000,
the State Museum of Art in Copenhagen undertook the restoration of one of the
largest Danish pre-20th
-century Danish paintings: Kristus i de dødes Rige [Christ in the
Land of the Dead] by Joakim Skovgaard, painted 1892-95, an immense canvas
measuring 350 x 580 centimeters. The restoration was carried out in public, which not
only served to attract attention to Sjælebilleder, a large exhibition of symbolist art held
later that year at the State Museum, but also made the restoration work visible. Less
obviously, it also formed a new chapter in the contentious history of the painting, which
well exemplifies the way history, even the history of art, is often fiercely contested.
PRESERVE THEN SHOW / DANISH FILM INSTITUTE / SIDE 19CASPER TYBJERG
Preserve then show 2.korr. 9/16/02 2:13 PM Side 19
SIDE 20 / DANISH FILM INSTITUTE / PRESERVE THEN SHOW
The State Museum of Art refused to buy the painting when it was first exhibited, and
although it relented in 1911, it was long considered something of an embarrassment.21
It was finally removed from display in 1966, and since it was too large to store as it
was, it was removed from its frame and rolled up. For the next quarter of a century
no one saw the painting; there did not even exist a colour photograph of it. The Danish
symbolist painters of the 1890s were long out of favour with art historians, who
tended to ignore them in favour of the aggressively colourful modernists who followed
them in the early 20th
century. Only since the early 1990s has this really begun to
change. An important indication was the publication of a new multi-volume history of
Danish art which had an entire volume on symbolism and impressionism;
characteristically, its author was not an art historian but a historian of literature, Henrik
Wivel.22
Wivel is scathing in his attacks on the way the symbolists have been made to
disappear from view. He writes about Kristus i de dødes Rige: “...that this in every way
nerve-wracking masterpiece now, a century later, is still hedged about with prejudice
at the State Museum of Art can be seen, or, precisely, not seen, from the fact that it
since 1966 has lain rolled up in the museum’s basement.”23 The State Museum’s
handling of the painting had become controversial, and the worry that it had suffered
serious damage in storage (or would do so if unrolled) was perhaps a reason for the
museum not only to do something about it, but to be seen to do so.
Be that as it may, the museum made a point of putting the conservation on show;
the painting as object became the centre of attention. The conservation could be
followed by the museum’s visitors, and photos showing the gradual progress of the
work were posted on the museum’s website. After the completion of the conservation,
the work was documented in a very nicely produced little book.24
While it might not be possible or useful to give prints a place in collective memory,
it may nevertheless be worth trying to give the objects a history: to insist on detailed
documentation, to make public presentations that highlight restoration and
reconstruction work, to draw attention to the qualities of particular prints – and not
only their defects, and to display fragments whose very incompleteness draw attention
to the vicissitudinous processes that have allowed some films to survive and consigned
many more to oblivion. Perhaps this will encourage us film historians to reflect a little
more upon the precariousness of what we too often regard as given, the efforts
required to preserve and prepare the “raw material” for our work.
NOTES
1. Robert C. Allen and Douglas Gomery, Film History: Theory and Practice (New York: McGraw-Hill,
1985), 38.
2. The importance of thorough documentation in film restoration is well described in Michele Canosa,
Gian Luca Farinelli, and Nicola Mazzanti, “Black on White: Notes on Film Restauration: The
Documentation,” Cinegraphie 10 (1997).
3. Werner Faulstich, “Filmanalyse als Filmgeschichte: Prinzipen einer konstruktivistischen
Filmgeschichtsschreibung,” in Filmgeschichte schreiben: Ansätze, Entwurfe und Methoden, ed. Knut
Hickethier (Berlin: Sigma, 1989), 56 (emphasis added).
PHILOSOPHY
Preserve then show 2.korr. 9/16/02 2:13 PM Side 20
4. Werner Faulstich and Helmut Korte, “Vorwort der Herausgeber zum Gesamtprojekt,” in Fischer
Filmgeschichte, ed. Werner Faulstich and Helmut Korte (Frankfurt a.M.: Fischer Taschenbuch Verlag,
1994), 9.
5. Jürgen Felix, “Die Anfänge der Slapstick Comedy: Arbeit (Work, 1915),” in Fischer Filmgeschichte,
ed. Werner Faulstich and Helmut Korte (Frankfurt a.M.: Fischer Taschenbuch Verlag, 1994), 264.
6. David Bordwell, Making Meaning: Inference and Rhetoric in the Interpretation of Cinema
(Cambridge, Mass.: Harvard University Press, 1989), 266-67. See also David Bordwell, “Historical
Poetics of Cinema,” in The Cinematic Text. Methods and Approaches, ed. Barton Palmer (New York:
AMS Press, 1989).
7. Henry Jenkins, “Historical Poetics,” in Approaches to Popular Film, ed. Joanne Hollows and Mark
Jancovich (Manchester: Manchester University Press, 1995), 100.
8. Noël Carroll, “Towards an Ontology of the Moving Image,” in Philosophy and Film, ed. Cynthia A.
Freeland and Thomas E. Wartenberg (New York: Routledge, 1995), 77.
9. Paolo Cherchi Usai, Silent Cinema: An Introduction (London: BFI, 2000), 147.
10. Paolo Cherchi Usai, The Death of Cinema: History, Cultural Memory and the Digital Dark Age
(London: BFI, 2001), 103.
11. Mark-Paul Meyer, “Nitrate, Take Care!,” in Restauro, conservazione e distruzione dei film, ed. Luisa
Comencini and Matteo Pavesi, Quaderni Fondazione Cineteca Italiana (Milano: Il Castoro, 2001), 55.
12. Cherchi Usai’s Rule 10 runs: “The ‘original’ version of a film is a multiple object fragmented into a
number of different entities equal to the number of surviving copies.” (Cherchi Usai, Silent Cinema:
An Introduction, 160.)
13. Martha Howell and Walter Prevenier, From Reliable Sources: An Introduction to Historical Methods
(Ithaca, N.Y.: Cornell University Press, 2001), 56.
14. James W. Marchand, What is Codicology? ([cited 31 October 2001]); available from
http://www.artsci.wustl.edu/~smcarey/whatiscodicology.html..
15. Encyclopedia Britannica Online, s.v. “textual criticism: “Textual criticism in this sense is usually called
filologia in Italian; see, particularly, Michele Canosa, “Per una teoria del restauro cinematografico,”
in Storia del cinema mondiale, ed. Gian Piero Brunetta (Torino: Einaudi, 2001).
16. Bernhard Bischoff, Latin Palaeography: Antiquity and the Middle Ages, trans. Dáibhí Ó Cróinín and
David Ganz (Cambridge: Cambridge University Press, 1990), 20.
17. Cherchi Usai, Silent Cinema: An Introduction, 44.
18. Peter Novick, The Holocaust and Collective Memory: The American Experience, British Paperback
ed. (London: Bloomsbury, 2001), 3.
19. For some examples, all primarily concerned with World War Two, see Novick, The Holocaust and
Collective Memory, Edward T. Linenthal and Tom Engelhardt, eds., History Wars: The Enola Gay
and Other Battles for the American Past (New York: Henry Holt, 1996), and Claus Bryld and Anette
Warring, Besættelsestiden som kollektiv erindring: Historie- og traditionsforvaltning af krig og
besættelse 1945-1997 (København: Roskilde Universitetsforlag, 1998).
20. Gary Taylor, Cultural Selection: Why Some Achievements Survive the Test of Time - and Others
Don’t (New York: Basic Books, 1996), 140.
21. For the early debates over the painting, see Bente Scavenius, “Et omstridt gigantbillede,” in
Kunstværkets krav: 27 fortolkninger af danske kunstværker, ed. Ernst Jonas Bencard, Anders Kold,
and Peter S. Meyer (København: Fogtdal, 1990).
22. Henrik Wivel, Symbolisme og impressionisme, vol. 5, Ny Dansk Kunsthistorie 1-10 (København:
Fogtdal, 1994; reprint, 1995: Den store stil: Dansk symbolisme og impressionisme omkring år
1900). It does not stand alone, of course. Among other things, one should certainly mention an
important exhibition of Danish symbolist art held in 1993 which – according to the catalogue –
was the first of its kind (Hanne Honnens de Lichtenberg, ed., Symbolismen i dansk kunst:
Udstillingskatalog (Nivaagaards Malerisamling, 1993), 3).
23. Wivel, Symbolisme og impressionisme, 243.
24. Henrik Bjerre and Peter Nørgaard Larsen, Skovgaard-konserveringen (København: Statens Museum
for Kunst, 2000).
PRESERVE THEN SHOW / DANISH FILM INSTITUTE / SIDE 21CASPER TYBJERG
Preserve then show 2.korr. 9/16/02 2:13 PM Side 21