Results of variations on "exquisite corpse"

What can I say? It doesn't fit above, put it here. Also the location of board rules/info.
Post Reply
User avatar
eldin raigmore
korean
korean
Posts: 6379
Joined: 14 Aug 2010 18:38
Location: SouthEast Michigan

Results of variations on "exquisite corpse"

Post by eldin raigmore » 08 May 2014 18:42

(Maybe this should go in the Linguistics&Natlangs subforum, but I took a guess and put it here.)

I am wondering about how certain variations on the surrealist auctorial game "exquisite corpse" (a.k.a. «cadavre exquis» or "rotating corpse") might change the results.

(My guess is that the majority of us aren't familiar with this game. But that's just a guess.)

In particular I'm interested in variants of this: "The technique was invented by Surrealists and is similar to an old parlour game called Consequences in which players write in turn on a sheet of paper, fold it to conceal part of the writing, and then pass it to the next player for a further contribution."

In the version with which I am most familiar (which is to say, am barely acquainted), each contributor adds one complete sentence (starting on the line below where the previous contribution ended), folds the paper so that all contributions before his/her own are concealed (the originator perforce skips that action), and passes the paper to the next contributor; who then does the same.
Eventually the end result is unfolded and read out for everyone.

I'm interested in three kinds of variations.
Edit: Has anyone on the CBB ever played anything similar to any of the following variations?

Firstly, how many turns does each player get before the end? And how does that change the result?
If the game circulates more than once, and unless the game is played by mail among a huge number of players, each player might (and I guess probably will) remember, more-or-less, what his/her own previous contribution(s) was/were (and probably to the same or a lesser degree what the contributions just before his/hers were). Depending on how many other players contributed in-between, that can make it a lot easier to "keep on track", or make "throwing it off track" require intention, rather than being accidental.

Second, how many previous contributions is the next contributor allowed to see? And how does that change the result?
My guesses are as follows.
If the answer is "zero", the "story" will usually not even be connected.
If the answer is "one", the story's "path" will be jagged; it can take an abrupt turn at each sentence.
If the answer is "two", the story's "path" will "turn" ("change direction") smoothly, though often not gently.
If the answer is "three", the turns will be smoother and gentler; if it's "four", even smoother and even gentler (maybe, even, close to a "straight line").

I also guess that the number of contributions one may see must be less than the number of other contributors to even make it a game; from each contributor at least one other contributor's contribution must be concealed.

This variation might interact with the "how many turns" variation.

Thirdly, what if, in addition to (or instead of) the last contribution (or the last two, or the last three, or the last four), the next contributor is also permitted to see the first contribution (or the first two, or the first three, or the first four)?
If each contributor is allowed to see the beginning of the story, but not the most recent contribution(s) before his or her own next contribution, my guess is the resulting "story" would read like excerpts from several different subplots.
But, if the story "goes around" only once, then (provided there are at least four contributors) if each contributor is allowed to see both the first sentence and the last contribution, the story's "arc" could still be pretty jagged.

I wonder if being allowed to see the first two sentences would make a significant difference versus being allowed to see only the first sentence? And how, if at all, would being allowed to see the first three or four sentences vary from being allowed to see only the first one or two?

For it to be a game, in which some contributors never see some previous contributions, the total number of sentences any contributor could see must be fewer than the total number of other contributors. So if, for instance, each contributor may see the first sentence and the last two previous sentences, or the first two sentences and the last previous sentence, there need to be at least five contributors playing the game; if each can see the first two sentences and the last four previous sentences, or the first four sentences and the last two previous sentences, there need to be at least eight contributors playing the game.

This variation probably would not interact significantly with the "how many turns" variation, I surmise. Shortly after the first complete turn, (for instance, if players are allowed to see the first four sentences, after the fourth player has had his/her second turn), every player would already have made a contribution while being allowed to see as much of the beginning as anyone is allowed to see.

But I wonder how it would interact with the "how many of the last few directly preceding contributions is one allowed to see" variation?


[hr][/hr]


I suppose that to make complete experiments on all of these questions one might need (relatively) many friends with (relatively) much time.

But one small experiment might need only four players and enough time to think of, compose, write, and read, eight sentences.
Let each player see only the first line and the last contribution before his/her own next contribution.
Circulate the paper twice.
Read the resulting eight-sentence "story".

A moderately ambitions experiment that would result in an eighteen-sentence story;
SIx players.
Each player is allowed to see the first two sentences and also the last two sentences before his/her own current contributiion.
Circulate the paper three times.

(See below* where I think a set of twenty experiments with six participants would probably give one all the information one "needs", to the extent that any of us really need any.)

For a complete set of one hundred experiments:
For each f running from 0 to 4, and each l also and independently running from 0 to 4, and each t independently running from 1 to 4;
Let the number p of players be at least f+l+2.
Circulate the paper t times.
Let each player see the first f sentences and also the last l sentences.

The story in each such experiment will contain p*t sentences. p can vary, down to a minimum that depends on f and l; but when f=l=4, p must be >=10.
If p=10 in each experiment, you'll get 100 stories, 25 of them being 10 lines long, 25 of them being 20 lines long, 25 of them being 30 lines long, and 25 of them being 40 lines long.
The amount of time each experiment takes will be proportional to the product of t times p.

Sustaining each participant's enthusiasm for the full 100 stories might (? almost definitely would) be difficult, moreso for more participants.
So it might be better to assemble a different group of subjects for each of the experiments where t*p is "highish".

But a basis or starting-point could be just the following four experiments (three four-liners and one eight-liner) with the same set of four participants: I'm guessing this should be easy.
i. t=1, l=1, f=0; circulate the paper just once, allowing each participant to see just the last previous contribution. Story length t*p=1*4=4 lines.
ii. t=2, l=1, f=0; circulate the paper twice, allowing each participant to see just the last previous contribution. Story length t*p=2*4=8 lines.
iii. t=1, l=2, f=0; circulate the paper just once, allowing each participant to see the two immediately preceding contributions. Story length t*p=1*4=4 lines.
iv. t=1, l=1, f=1; circulate the paper just once, allowing each participant to see the immediately preceding contribution, and also the first contribution. Story length t*p=1*4=4 lines.
The four experiments would involve writing twenty lines in all. It should probably be possible to do all four of them in one session.

A modestly more ambitious set; which IMO and IMG would give a good hint as to the direction the results will take, would be the following ten experiments, with a constant set of five (somewhat dedicated) participants;
i. t=1, l=1, f=0; circulate the paper just once, allowing each participant to see just the last previous contribution. Story length t*p=1*5=5 lines.
ii. t=2, l=1, f=0; circulate the paper twice, allowing each participant to see just the last previous contribution. Story length t*p=2*5=10 lines.
iii. t=1, l=2, f=0; circulate the paper just once, allowing each participant to see the two immediately preceding contributions. Story length t*p=1*5=5 lines.
iv. t=1, l=1, f=1; circulate the paper just once, allowing each participant to see the immediately preceding contribution, and also the first contribution.. Story length t*p=1*5=5 lines.
v. t=3, l=1, f=0; (circulate the paper three times). Story length t*p=3*5=15 lines.
vi. t=1, l=3, f=0; (each sees the last three lines contributed while writing his/her own current contribution); Story length t*p=1*5=5 lines.
vii. t=1, l=1, f=2; (each sees the first two lines as well as the last preceding line); Story length t*p=1*5=5 lines.
viii. t=2, l=2, f=0; (circulate the paper twice, each participant seeing the last two lines); Story length t*p=2*5=10 lines.
ix. t=2, l=1, f=1; (circulate the paper twice, each contributor seeing the first line as well as the last line); Story length t*p=2*5=10 lines.
x. t=1, l=2, f=1; (each participant sees the last two lines and the first line); Story length t*p=1*5=5 lines.

Result in toto; six five-lliners, three ten-liners, and one 15-liner. Total, 75 lines. At a minute a line that would take an hour and a quarter. It shouldn't be that hard to get five college students together for an hour and a half, I'm guessing, though I might be wrong -- again.
If it takes only forty seconds per line it could be done in forty-five minutes; or at forty-eight seconds a line it could be done in an hour.
But at two minutes per line it would take two and a half hours -- half a morning or half an afternoon -- and might better be split into two sessions.
At three minutes a line it would take three and three-quarters hours -- almost an entire morning or almost an entire afternoon. It would definitely be better split into at least two, and maybe three, sessions.

*One could probably get and confirm all the data one wants and needs with the following set of 20 experiments involving p=6 players each time.
i. t=1, l=1, f=0; circulate the paper just once, allowing each participant to see just the last previous contribution. Story length t*p=1*6=6 lines.
ii. t=2, l=1, f=0; circulate the paper twice, allowing each participant to see just the last previous contribution. Story length t*p=2*6=12 lines.
iii. t=1, l=2, f=0; circulate the paper just once, allowing each participant to see the two immediately preceding contributions. Story length t*p=1*6=6 lines.
iv. t=1, l=1, f=1; circulate the paper just once, allowing each participant to see the immediately preceding contribution, and also the first contribution.. Story length t*p=1*6=6 lines.
v. t=3, l=1, f=0; (circulate the paper three times). Story length t*p=3*6=18 lines.
vi. t=1, l=3, f=0; (each sees the last three lines contributed while writing his/her own current contribution); Story length t*p=1*6=6 lines.
vii. t=1, l=1, f=2; (each sees the first two lines as well as the last preceding line); Story length t*p=1*6=6 lines.
viii. t=2, l=2, f=0; (circulate the paper twice, each participant seeing the last two lines); Story length t*p=2*6=12 lines.
ix. t=2, l=1, f=1; (circulate the paper twice, each contributor seeing the first line as well as the last line); Story length t*p=2*6=12 lines.
x. t=1, l=2, f=1; (each participant sees the last two lines and the first line); Story length t*p=1*6=6 lines.
xi.t=4, l=1, f=0; (circulate the paper four times) Story length t*p=4*6=24 lines.
xii. t=1, l=4, f=0; (each sees the last four lines); Story length t*p=1*6=6 lines.
xiii. t=1, l=1, f=3; (each sees the first three lines as well as the last preceding line); Story length t*p=1*6=6 lines.
xiv. t=2, l=2, f=1; (circulate the paper twice, each contributor sees the last two lines and the first line) ; Story length t*p=2*6=12 lines.
xv. t=3, l=2, f=0; Story length t*p=3*6=18 lines.
xvi. t=3, l=1, f=1; Story length t*p=3*6=18 lines.
xvii. t=2, l=3, f=0; Story length t*p=2*6=12 lines.
xviii. t=1, l=3, f=1; Story length t*p=1*6=6 lines.
xix. t=2, l=1, f=2; Story length t*p=2*6=12 lines.
xx. t=1, l=2, f=2; Story length t*p=1*6=6 lines.

Summary of results: twenty stories, including ten six-liners, six twelve-liners, three eighteen-liners, and one 24-liner. A total of 210 lines.

To do all twenty of those experiments all with the same set of six subjects, I suspect that the subjects would have to be both dedicated and paid (perhaps by the line? would $2.00 per line be about right? or is that too much?), and available for some six longish (42-line) to ten shortish (24-line) sessions (or maybe four marathon (60-line) sessions). But of course, that's just my guess.

I also guess that once you get your subjects to start a session it will be easier to keep all of them for the entire session than it will to get the same set of subjects committed to multiple sessions.

But I'm not sure it would be necessary to have the same set of subjects for each experiment. In fact, if you have six subjects, you can run two of the four l=1 f=0 tests (i., ii., v., & xi.) simultaneously, each with a set of three players.

Edit: It's possible some CBBer already knows some answers to some similar questions, and/or has played something like some of these variations. If you have, please answer. Thank you.

Post Reply