by David Bordwell. Routledge, 2007.
512 pages, 7 × 10 inches. Illustrations.
112 kb pdf]
[Chapter 3 | Three Dimensions of Film Narrative / 2 mb pdf]
[Chapter 4 | Cognition and
Comprehension: Viewing and Forgetting in Mildred Pierce / 2 mb pdf]
[Chapter 10 | CinemaScope:
The Modern Miracle You See
Without Glasses /
1.7 mb pdf]
[Chapter 11 | Who Blinked First? /
149 kb pdf]
Poetics of Cinema (2007) gathers several published articles,
one going back as far as the 1970s, and several new ones. They move in many different
directions, across the history of film, but they do conveniently highlight several
areas of my research. Most important, all of these areas of inquiry put a film
or batch of films at the center of the process.
This automatically sets me off from some of my peers.
Today there are many scholars studying the film industry and probably many more
seeking to understand filmic reception by studying audiences in their sociocultural
contexts. For those researchers, all the films on earth could dissolve into nothingness
tonight, and they’d
get up next morning and keep going without losing a beat. Film history without
films doesn’t work for me. I need my movies!
Some people think of me as
a theorist, and I suppose I am to some extent, but I have always tried to keep
theoretical reflection tied to the concrete facts of cinema—as films, as
experiences, as part of a historical process. The first section of Poetics
of Cinema is the most evidently theoretical.
The introductory essay introduces the idea of poetics, a frame of reference that
helps me study the things that matter to me most.
One of these things is the aesthetic
dimensions of cinema, particularly the ways in which form and style shape meanings
and effects. This goes back to my youth, when I fell under the spell of literary
and art criticism. My earliest publications were in the area of film criticism,
and my starting point for almost any project is an analysis of how the films
in question are put together at the broadest level and how they work on a moment-by-moment
basis. I suppose an art critic would speak of structure and facture;
a literary critic might speak of architectonics and texture.
Less grandiosely, I’d
say I study large-scale cinematic forms, typically narrative ones, and small-scale
A second area of concern for me involves the historical
circumstances that create aesthetic traditions, such as classical Hollywood storytelling.
Once we bring formal principles to light, how shall we explain them? How do we
think they arose in history? What factors in culture or the film community can
we invoke to explain how these films took the shapes they did? This has led me
to examine national cinema traditions (in my work on Dreyer, Ozu, and Eisenstein)
as well as technological and production conditions (in The Classical Hollywood
Cinema and The
Way Hollywood Tells It). I’ve also posited some historical continuities
that are transnational, such as the tradition of depth staging (in On the
History of Film Style and Figures Traced in Light).
interested in explaining how the aesthetic dynamics of films are designed to
elicit effects from spectators. In the mid-1980s, I found that the most satisfying
explanations for why films had the forms and styles they did involved the ways
in which these aesthetic elements engaged our minds. Every gesture, every cut,
every camera movement is designed to shape our uptake. Our experience—perceptual,
cognitive, emotional—is guided by form and
style. It seemed to me then, and still does, that a broadly cognitive conception
of how our minds mesh with movies is well suited to explain many aspects of cinematic
These three areas of inquiry—film aesthetics, the
causes of aesthetic traditions, and cognitive explanations for artistic design—are
reviewed in the volume’s introductory essay, “Poetics of Cinema.” Ideas
from these domains are put to use in the following essay, “Convention,
Construction, and Cinematic Vision.” There I analyze some functions, sources,
and effects of shot/reverse-shot editing.
After this pair of theoretical essays,
Part II concentrates on problems of narrative. I offer a general theory
of cinematic narrative, teased out along three dimensions. This overview is followed
by some case studies: an analysis of narrative comprehension in Mildred Pierce,
a discussion of the tradition of “art cinema,” and
a discussion of “forking path” narratives like Run Lola Run and Sliding Doors.
The last essay in the section is a new, long study of what I call “network
narratives”—films with several protagonists whose actions both diverge
and interlock. I discuss several films in this essay, but the most detailed analysis
is reserved for Nashville, Magnolia, Favoris de la lune,
and Les Passagers. This essay gives me the opportunity to test out several
of the ideas floated in earlier pieces.
Part III is devoted to stylistics. It
gathers essays on Andrew Sarris’s
theory of style and the techniques of classic Japanese and Hong Kong films. Another
essay examines the very distinctive style of the German director Robert Reinert.
The longest and newest essay is devoted to the stylistics of CinemaScope. Here
I try to trace how different principles of staging were manifested in the early
anamorphic movies. Like Part II, this part doesn’t neglect the way aesthetic
choices affect spectators. For instance, one essay proposes a cognitively flavored
explanation of why actors so seldom blink.
The essays collected here don’t
sample all areas that intrigue me. For example, I’ve written “meta-analyses” of
the conventions of critical interpretation (Making Meaning) and of the
historiography of style. Still, this anthology is a fair survey of the kind of
done over the last thirty years. Many of the topics in the book are taken up
in entries on the blog maintained
by Kristin Thompson and me.
main page on Amazon suggests that it’s available only
in horrendously overpriced hardcover, but there is a more reasonably priced paperback
There are some errors in Routledge’s design of Poetics of Cinema.
For reasons nobody would explain to me, some pages are more densely packed with
text than others. More important, the designer didn’t lay out the chart
accompanying the Mildred Pierce essay correctly. (It’s also inaccurate
in the first printing of Leo Braudy and Marshall Cohen’s Film Theory
and Criticism of 2009, but in a different way: That version leaves off everything
after shot 4. What is it about this harmless chart?)
But I made mistakes too, and here are the ones that have
surfaced after the book’s
first fifteen months of life.
p. 30: Not Spiderman but Spider-Man.
p. 35, lines 6–7: Mistaken repetition of clause referring
to running time of introductory shot of Rope.
p. 126, second sentence from the top: Should be “All
that matters is that a scene calls forth in us a mental schema—people tell
one another about an event that has occurred—and that triggers only one
p. 133, last sentence of first paragraph: Should be “If
the implied author is the set of overarching principles of design governing the
film, we can simply talk about said principles, even, or especially, when they
create problems of unreliability for the spectator.”
p. 221, fourth line:
Should be “Michael Haneke.” Thanks to Peter
Parshall for the correction.
p. 243, third line: Should be “2:37 from
to Peter Hourigan for pointing this out.
p. 248, third line: Should be “Cesc
p. 350: Not Oscar in The Little Foxes but Horace.
here is the Mildred Pierce chart, currently on p. 144, as it should
Pierce: The opening scene and its replay
1. 5 sec: (extreme long shot)
house at night, car visible alongside (Fig. 4.3). Dissolve to:
1. 12 sec: (medium shot)
goes to the car and tries to start engine (Fig. 4.12).
2. 4 sec: (long shot)
and car. Two pistol shots heard (Fig. 4.4).
2. 4 sec: (medium close-up):
slumped over steering wheel. Two shots heard (Fig. 4.13).
3. 8 sec: (medium long shot)
facing camera, looking off left (Fig. 4.5). Third and fourth pistol shots
3. 5 sec: (ms)
Veda fires four
times (Fig. 4.14).
Monte is hit, staggers forward (4.6) and
falls to floor. A pistol is tossed into the frame (Fig. 4.7).
4. 6 sec: (ms)
Monte is staggering
forward and falls to floor (4.15). A pistol is tossed into the frame (Fig. 4.16).
4. 13 sec: (ms):
his head, opens his eyes, and says: "Mildred" (Fig. 4.8).
Monte wobbles his head, eyes open, and says: "Mildred" (Fig.
Pan up to mirror; sound of door slamming (Fig
Pan with Mildred coming
in; sound of door slamming (Fig. 4.18).
She meets Veda in the parlor (Fig. 4.19).
In the parlor, Veda
tells Mildred lies about shooting Monte. Near the doorway, she begs Mildred to
protect her (Fig. 4.20).
Empty parlor, with Monte's
corpse in the firelight. Doorway empty (Fig. 4.10).
Car outside pulls off