More data on scene rhythms
A couple of quick updates. First, my article that I published previously to the blog, “Playing for Plot in the Lost and Portal Franchises” has been published in a revised version in the journal Eludamus: Journal for Computer Game Culture.
Second, I’ve been playing with the metrics I discussed in my post earlier this week about “scenic rhythms,” and added a number of other episodes to my list. After much tinkering with various spreadsheets and finally finding success with Open Office, I figured out how to put those data into a chart to visualize the different patterns. Again, the two variables are Scenes per Hour and Foci (essentially meaning how many storylines are threaded throughout the episode). Here’s the chart:
My choice of programs was very much based on whatever was handy for me to tabulate. I added a few single camera sitcoms to see how they compare—not surprisingly, the shorter 20-25 minute length and quick comic pacing leads to high SpH. Cougar Town was highest, with a scene per minute, a number boosted by the comic device of the smash cut to another time and/or place for a joke. I also think it’s interesting what a gap there is in number of foci between The Wire / Game of Thrones at 12, and everything else that falls in the 3-7 range. That might be selection bias, so I need to think of other programs that are likely to have more than 7 foci (beyond Treme) – maybe Deadwood?
Do people have ideas for other programs or episodes to include here? I don’t know if I’m ready to launch a full-fledged database of SpH yet like Cinemetrics, but a bit of crowd-sourced brainstorming could be helpful, as I’m thinking of writing an essay on this topic. Some thoughts I have are: to look at some pilot episodes to see how their rhythms might differ from typical episodes; to look at more network dramas, including serialized shows (Revenge and Good Wife are soon to come) and procedurals; to consider multicamera sitcoms in the mix; and maybe spanning back into other eras. There are also some data variables that I need to think about, including defining foci a bit more clearly and also being able to account for the variability of the counts – in some instances, a “focal thread” might only consist of one scene (a serialized “runner”) while another might take up more than half the episode, and likewise a single scene of 10 minutes might skew an episode’s measurement quite a bit – thus Breaking Bad and Game of Thrones have similar SpH, but Breaking Bad features a wider range of scene lengths, while GoT is much more consistently paced throughout.
It feels like this is potentially an interesting tool to provide a clear basis for comparison across genres and production modes, but like any measure it is only useful for particular questions and parameters. So any feedback is much appreciated!
Filed under: Media Studies, Narrative, Television, TV Shows | 11 Comments
Tags: cougar town, quantitative analysis, Scenes per Hour
random thoughts from media scholar Jason Mittell
check out my books:
Narrative Theory and ADAPTATION.
Complex TV: The Poetics of Contemporary Television Storytelling
How To Watch Television
Television & American Culture
Genre and Television: From Cop Shows to Cartoons in American CultureRecent Comments
-
Just TV 
I think this could be quite useful in terms of mapping a certain level of comparison between shows within broad structural parameters. For instance, I think it’s really interesting how Game of Thrones is sort of over there by itself, with a striking number of foci played out in relatively few scenes, by comparison with the differing serial styles of The Wire on the one hand and Breaking Bad on the other.
I think Reality Series should find a place on the chart too, primarily because they are scripted/edited to create a narrative arc and it would be quite interesting to see where those place within this foci/SpH structure; especially in comparison to fully scripted television. Since I don’t watch much reality TV, I am not sure which ones would be good to sample, probably those where cast members interact such as The Real L Word or the new series Push Girls?
As far as series with more than seven foci I’d be hard pressed as well, maybe Boardwalk Empire, Shameless (just to move from HBO), True Blood, and on network and cable… I have nothing but The Office… maybe.
I think pilot versus mid-series-sample episode is a great idea, because the rhythm is often so clearly different (And when it isn’t, how viable are they to audience interest -I am thinking Flashforward here? So that is an interesting way to maybe use the idea of narrative rhythm).
When I read your first post on the subject and now this one I keep going back to Ryan McGee’s (very problematic but clearly thought provoking) blog post on the “decline of the episode” in my head and wonder if there may be a correlation here… mmmhhh
Agreed that the “episode” vs “installment” question plays into things potentially – I see that GoT is often “thematic installment” more than “discrete episode,” which may have a lot to do with the number of foci it juggles each week – when it cuts down (as on “Blackwater”), it feels more like a tight episode.
Reality examples would be great to show the fractured structure – of course, there are many subgenres that would play differently. Amazing Race or other multicontestant competitions would certainly be much more broken up than Real Housewives or other docusoaps (I assume).
I agree. Maybe including the docu-soaps, as more closely related narratively to series, first would be interesting. I wonder how much fragmentation would be visible… I’ll watch PushGirls
on Monday and shoot you the numbers according to your set-up after. 🙂
this is a great great great idea! loved it!
i suggest some Community and Arrested Development in there as well. (arrested will top 120 SpH!)
I was always amazed at how extreme The Simpsons was on this score.
This is great, I would love to use this idea to frame a study in an ongoing research program on complexity and media choice. I’ve been thinking of the best way to incorporate it since your earlier blog post. Would you be open to discussing more?
Another exciting line of inquiry being pursued here! As you say in the post on the scenic rhythms of ‘Game of Thrones’, “Rhythm is a hard thing to analyze and measure”. A similar comment in Bordwell & Thompson’s ‘Film Art’ lead me to spend my four years of PhD study on exactly this question. Perhaps my definition of rhythm in film editing distilled from the first four chapters of my book ‘Cutting Rhythms’ (Focal Press, 2009) may be useful to you? “Rhythm in film editing is time, energy and movement, shaped by timing, pacing and trajectory phrasing, for the purpose of creating cycles of tension and release.” (p.80)
The issue with quantification is not that it isn’t useful, it is, but that it only really describes pacing, which, according to my definition anyway, is a part of rhythm, but not all of it. I define pacing (ch 3) as the rate of movement or change, which may be set by the rate of cutting, but may also be set by rate of movement or change within a shot, or the rate of movement or change overall.
Movement (as the next four chapters go on to discuss) may be movement of plot or events, movement of emotion, or movement of image and sound. Your inquiry is, I think, dealing primarily with movement of plot. (Though a question concerning rhythm and tone will eventually lead you to consider all three).
By adding foci to your graph you make it a much more effective tool for measuring movement of plot or events than just scenes per hour – you bring in content or what I would call ‘open questions’ to the discussion of pace. This allows you to begin to account for the cycles of tension and release, not just the number of cuts or number of scenes. Acknowledging your work on closure in series and serials, my colleagues and I discuss the movement of plot or events in terms of the opening and closing of dramatic questions. This would be difficult and much more subjective than scenes per hour to graph, but as a third vector of rhythm (not just pace) it would ask: is there a dramatic question opened, complicated or closed in the scene? Is tension initiated, increased or resolved?
Would not seem to be correct protocol, in a comment on a blog post to unpack this too much further, but happy to correspond if you think it may be useful!
Karen – these are great comments, and thanks for pointing me toward your book! I just ordered it for our library and I look forward to reading it.
I agree the question of narrative events would be another key variable – and within TV writing, they call them “beats” in a clear rhythmic reference. But those are much harder to identify clearly, especially in a serial where plot points can retrospectively emerge. But as I dive in more to the literature on rhythm, pacing, and scene structure, I’ll definitely be in touch…
Amazing idea!