I'm very sorry, Mrs. Chomsky, but we have to abort the pregnancy
A
realized some time ago that I usually immensely like films that don't
tell their story in linear (chronological) order, but tell it out of
sequence so that they always move on to a part that is presently the
most interesting for the storytelling purposes. Off the top of my head,
such movies include Reservoir Dogs, Pulp Fiction, Go, 21 Grams, Memento, Batman Begins and Retroactive.
Quality films one and all, and I can't even think of a single bad movie
in which the storytelling was done in a similar nonlinear fashion.
(There must be some, once the technique became fashionable.) Even
Retroactive, a cheap B-action flick, is greatly improved by its witty
nonlinear flow of events, as the heroine keeps taking the time machine
back to the past to fix it, but each time causes things to go even
worse than before. "Nonlinear" seems like such a silly pomo relativist
word, so what's going on here?
My present hypothesis is that the human brain simply likes the nonlinear storytelling and the extra puzzle-solving activity associated with it, as long as these puzzles are light enough so that they don't exert too much strain on the brain but can be solved without any conscious thought. The subconscious brain activity that takes place in deducing the meaning of and the connections between events told out of sequence feels pleasant, if it is not too confusing. This is probably at least slightly analogous to the way that watching cartoons and reading comics somehow feels more pleasant than watching live actors or reading written text. The linear storytelling just doesn't tickle certain parts of the brain that nonlinear storytelling reaches.
I have read some people criticize "Pulp Fiction" and other similarly nonlinear movies for being confusing. If the nonlinear storytelling is done right, there doesn't need to be any confusion, and of course shouldn't be. For example, when I watched each of the movies listed above, at no point did I feel any confusion what was going on. Well all right, the first half hour of the movie "21 Grams" was pretty confusing, since this movie is by far the most nonlinear mainstream movie ever made that I know of. After almost every single scene, the storytelling makes a long jump ahead or back in time. For this reason, the first half an hour in the movie is spent learning who these characters are and what they want. But when we gradually learn what is going on, it's fun to put pieces together and watch the rest of the movie to fill in the remaining gaps.
Of course, you can't just take any linear movie and recut its scenes into a nonlinear order to improve it. The storytelling must from the start be designed so that at each moment, the viewer has enough information to understand and enjoy the events so that the whole thing is not just a confusing mess of seemingly unrelated events. Nonlinearity could also be used to improve the storytelling itself, because it might allow leaving out parts of the story that are otherwise useless but that are needed establish something that is necessary for understanding some later parts. For example, if the parts of the story listed in chronological order were A-B-C-D, so that the viewer must see B to be able to understand C, linear storytelling constraint requires that B is included. However, if we could depict the events in order D-A-C, and seeing D also gives the viewer enough information so that he can understand or pleasantly deduce what is going on in C, we get to drop the useless part B. Instead of wasting time on B, we can now fill the rest of the 90 to 120 minutes with something else, which will hopefully be more interesting.
In computer programming, the grand master of the field Donald Knuth came up with the concept of literate programming, which is amusingly analogous to ideas that I presented above. Normally, when you write program code, it is presented to computer in very different sequence than how you would explain it to another programmer. If you had a program of 10,000 lines of code which you had to go through and explain to another competent programmer, it is highly unlikely that you would go through the code one module at the time, and each module one line at the time strictly from beginning to end.
The best order of explanation and understanding would be totally different from the order in which the code is given to the computer. First you would explain the general idea and structure, then start presenting selected highlights, leaving out parts and details that are necessary to give to the computer but that a competent programmer would automatically see what they are, based on the information given so far. In fact, in practice you would end up going through only a small portion of the code, because after that the rest should be obvious.
Removing the useless restriction of strict linearity improves the explanation by liberating it from the strict linearity of the natural world. And as I argued earlier in "In praise of the artificial", artificial pretty much always trumps natural.
My present hypothesis is that the human brain simply likes the nonlinear storytelling and the extra puzzle-solving activity associated with it, as long as these puzzles are light enough so that they don't exert too much strain on the brain but can be solved without any conscious thought. The subconscious brain activity that takes place in deducing the meaning of and the connections between events told out of sequence feels pleasant, if it is not too confusing. This is probably at least slightly analogous to the way that watching cartoons and reading comics somehow feels more pleasant than watching live actors or reading written text. The linear storytelling just doesn't tickle certain parts of the brain that nonlinear storytelling reaches.
I have read some people criticize "Pulp Fiction" and other similarly nonlinear movies for being confusing. If the nonlinear storytelling is done right, there doesn't need to be any confusion, and of course shouldn't be. For example, when I watched each of the movies listed above, at no point did I feel any confusion what was going on. Well all right, the first half hour of the movie "21 Grams" was pretty confusing, since this movie is by far the most nonlinear mainstream movie ever made that I know of. After almost every single scene, the storytelling makes a long jump ahead or back in time. For this reason, the first half an hour in the movie is spent learning who these characters are and what they want. But when we gradually learn what is going on, it's fun to put pieces together and watch the rest of the movie to fill in the remaining gaps.
Of course, you can't just take any linear movie and recut its scenes into a nonlinear order to improve it. The storytelling must from the start be designed so that at each moment, the viewer has enough information to understand and enjoy the events so that the whole thing is not just a confusing mess of seemingly unrelated events. Nonlinearity could also be used to improve the storytelling itself, because it might allow leaving out parts of the story that are otherwise useless but that are needed establish something that is necessary for understanding some later parts. For example, if the parts of the story listed in chronological order were A-B-C-D, so that the viewer must see B to be able to understand C, linear storytelling constraint requires that B is included. However, if we could depict the events in order D-A-C, and seeing D also gives the viewer enough information so that he can understand or pleasantly deduce what is going on in C, we get to drop the useless part B. Instead of wasting time on B, we can now fill the rest of the 90 to 120 minutes with something else, which will hopefully be more interesting.
In computer programming, the grand master of the field Donald Knuth came up with the concept of literate programming, which is amusingly analogous to ideas that I presented above. Normally, when you write program code, it is presented to computer in very different sequence than how you would explain it to another programmer. If you had a program of 10,000 lines of code which you had to go through and explain to another competent programmer, it is highly unlikely that you would go through the code one module at the time, and each module one line at the time strictly from beginning to end.
The best order of explanation and understanding would be totally different from the order in which the code is given to the computer. First you would explain the general idea and structure, then start presenting selected highlights, leaving out parts and details that are necessary to give to the computer but that a competent programmer would automatically see what they are, based on the information given so far. In fact, in practice you would end up going through only a small portion of the code, because after that the rest should be obvious.
Removing the useless restriction of strict linearity improves the explanation by liberating it from the strict linearity of the natural world. And as I argued earlier in "In praise of the artificial", artificial pretty much always trumps natural.
Comments