If you want a history of cinema that switches from popular
cinema to 'art' cinema [i.e., from Ford to Resnais], then
one could certainly use Deleuze's distinction between the
action-image and the time-image as an historical
distinction.  But this is not what Deleuze does--i.e., take
the distinction as an historical one.
 
To take Deleuze's distinction between two types of images as
a distinction between two periods would necessitate leaving
out a great deal of post-WWII popular cinema--most of it, in
fact.  That's what Deleuze does:  he switches from talking
about a wide range of films to leaving out great popular
genres (as Damian Peter Sutton points out).  But this
demonstrates that the distinction he's making is not an
historical one.
 
Deleuze is not offering a film history.  Nor can the
concepts he offers be mistaken for descriptions of either
periods or movements.  A glance at the table of contents
supports this.  E.g., Hitchcock represents one turning point
from the action-image to the time-image.  While Deleuze does
not cite Hitchcock's early British films, it would not be
difficult to take what he says about the post-WWII Hollywood
films and apply them to the earlier films.  Thus the
conceptual distinction is not an historical one.
 
Two other clues to the fact that the distinction between the
time-image and the action-image is not historical are:  (1)
the way Renoir's pre-WWII work is used to explain the
time-image, and (2) the way Welles' 1941 _Citizen Kane_ also
presents an example of the time-image.  If the time-image
described a post-WWII period or movement, then neither
Renoir nor _Kane_ would fit.
 
FROM DAMIAN PETER SUTTON:
> I realise that Deleuze never professes that his work is an historical
> one
[....]
> Deleuze historicizes film into two distinct periods, separated in
> time by the Second World War, and in development by the achievement
> of a narrational movement-image of final complexity.
 
This seems to me a bit strange:  Deleuze never says his work
is historical; but he "historicizes film into two distinct
periods."  But isn't the distinction between action-image
and time-image conceptual and not historical?
 
> Historical problems with this approach are his starting and finishing
> points (1917-80ish), whilst genre problems exist in his not dealing
> with popular genres, and in particular actions films.
 
Well if the concepts Deleuze offers, which he never calls
historical, are in fact problematic when taken as historical
concepts, then maybe there's a problem with so doing.
 
If one *does* take Deleuze's work as a film history, then it
is a Hegelian one, in which concepts follow each other in
historical order because of their logical development.  I
think many people are not so sanguine as they once were
about reducing history to a logic of concepts.
 
Sincerely,
Edward R. O'Neill
UCLA
Sociology/General Education
 
----
Screen-L is sponsored by the Telecommunication & Film Dept., the
University of Alabama.