Unspoken Cinema 2012 banner

Saturday, June 13, 2009

Average Shot Length in CCC

This is only a selection of a few films surveyed in the database of Average Shot Length compiled at Cinemetrics (strict model in red, and other broad contemplative films in blue). So these data don't give a conclusive picture of the whole trend yet. Though with 50 entries, it begins to give interesting results.
One film was off the chart : Five (Kiarostami) with an ASL of 14 minutes 45 seconds!
15 seconds is considered a slow movie within the mainstream commercial editing. The usual ASL tends to be around 2 or 4 seconds these days. Some very contemplative films can go below this mark, at 7 or 8 seconds, while maintaining throughout a narrative deprived of wordy scenes. Though most of them are easily 10 times slower than the typical mainstream narrative editing rhythm.
Chronologically, from Jeanne Dielman (almost the start of this CCC trend as I define it) till 2009. There is no pattern emerging at this point (from this limited sample), no clear evolution to be noted that would show a tendency towards longer ASL or on the contrary towards faster ASL.
There is no common tempo that would naturally emerge from the practice of similar plotless/meandering/contemplative plan sequences. But there is a group of 18 films within the range of 50-80 seconds. On the next graph below, you can see that most of the films corresponding to this ASL range come from Asia (and Latin America), while the same range is blank in Europe (which has however a dense cluster below 30 seconds and over 100 seconds).
And then beyond that, really experimental extend of shot duration, with ASL going over 90 seconds, which the realm of 2 filmmakers : Angelopoulos and Tarr, not surprisingly (with isolated shots by Kiarostami and Encina).
At the extreme end, The Man From London reaches an Average Shot Length of 4 minutes 24 seconds, with the conventional plot of a heist story. The shots last 66 times longer, on average, than the commercial standard at 4 sec per shot. Quite a margin to tell similar stories!
I also note that the broad CCC (blue) don't necessarily have a shorter ASL than the the strict models (red)... So using more dialogue and more of a narrative plot doesn't induce more cuts in the editing style. But they (films in blue) don't go in the very top of the ASL spectrum (the extreme long takes) either.

And this one is the scattered plot of the distribution of all ASL, with a repartition by geographical area. Each dot represents a film's ASL.
Here we have a more decisive picture emerging. Asia (with 19 films) and Europe (with 14) are the most represented, so the others need more films in the database to be conclusive. I left Kiarostami's Five in this one, so you can see the scale difference with the rest of the films surveyed.
But already we can see that the range in Asia is dense and remains below 1 minutes and a half. While in Europe the ASL range extends much wider.

7 comments:

ADRIAN said...

Very interesting study!!! GOsh! i should blog about this!

HarryTuttle said...

Yes please. That's why this data is out there, to generate reflection and discussion.
Oh I see you already did blog about it now (read ADRIAN's post at his blog, here). That was fast. Thanks so much for sharing your thoughts on this.

celinejulie said...

This is very interesting. Thank you very much for including AFTERNOON TIMES in it. Did you time it by yourself?

Now I wonder what the average shot length of AUTOHYSTORIA is. Maybe someone who saw this film can give the answer. :-)

Edwin Mak said...

@celinejulie

I just watched Autohystoria, and from my notes the approx. ASL info would be as follows...

Duration: 5445s
Cuts: 16
ASL: 337.5

HarryTuttle said...

I think I asked Tossapol to calculate for me the ASL back when I wrote my review of his film.

celinejulie said...

Thank you very much, Edwin and Harrytuttle. :-)

Matthew Flanagan said...

Weird - Edwin, I got 14 for Autohystoria (not inc. the black screen + music at the beginning or upside-down end credits)...

Surely we can't have a discrepancy for a film with so few shots! Have you made one up?! ;-)