July 26, 2012

Sequels and adaptations in popular movies, 1936-2011

After some off-the-cuff thinking about spin-off movies from today and the mid-century, I decided to look into the data in greater quantitative depth. To re-cap, the goal is to see whether there was another period like today when so many movies were either adaptations of an existing work or a sequel to an earlier movie.

Based on a hunch, it looked like the '50s were another such period. The strongest influence on the zeitgeist is the trend in the violence rate, so not even knowing very much about all the popular movies of that period, let alone before or just after, that was a good guess -- today and the '50s were both well into a falling-crime time. And based on a quick look at the '80s, a rising-crime period, it looked like there weren't so many sequels and adaptations.

Granted, movies are more than just narratives, and this approach totally leaves aside the question of visual originality. But it's a feasible way to take a crack at the central question. Also bear in mind that originality and greatness aren't the same thing. I'm not judging how great or enjoyable all these movies are because I haven't seen most of them. The focus is entirely on how original the process was that gave us the plot, characters, themes, verbal tone, and so on.

I went through every year from 1936 to 2011 and tallied how many movies in the box office top 10 were a sequel or adaptation. I didn't go back any further because there aren't easily accessible data on the top 10. Wikipedia only lists 5 to 7 for many years, and doesn't always have a link to information about the source of the story that far back. This will do pretty well, though, because that still gives us just about the entire falling-crime period of 1934-1958, the whole rising-crime period through 1992, and the whole falling-crime period since then.

Rather than stall the presentation and discussion of the data, I've put the methodology part at the very end, in case you want to know exactly how I nitpicked what did and did not count as an adaptation and a sequel. They weren't mutually exclusive, though: for example, I coded the new Twilight movie as both a sequel and an adaptation.

The first graph below shows the total number of movies with less original narratives (whether a sequel only, adaptation only, or both). Then just to probe whether the type of unoriginality matters, the second graph splits them apart to show how many were adaptations (regardless of whether or not they were also sequels), and how many were sequels (regardless of their status as adaptations).

Score another one for my theory. There's visible year-to-year variation, micro-periods of a few years that go against a trend, etc., but there are three clear movements -- up, down, and up. The first one climbs toward a peak around 1956-'58, then it starts to fall off toward an end-point around 1988-'91, and then rises again through the present. That just about perfectly tracks the trend in the homicide rate through three different movements up, down, and up again. So it's not a coincidence.

Also, it looks like most of this link to trend in the violence rate is due to how common adaptations are. Sequels show a more steady one-way growth over the entire period, and usually aren't as common as adaptations anyway. That makes the case even stronger -- at least if it were mostly due to sequels, you could say they're still pretty original, just carrying over the characters, themes, and maybe settings from one movie to another. Instead it's more due to borrowing all that stuff, plus a good amount of the plot, dialogue (or at least its verbal style), and even songs if it's a musical.

What are some potentially surprising things to notice in the main graph?

First, today's culture where so much is adapted or a sequel milking a popular series is not that old. Even as recently as the 1980s, movie narratives were a lot more original.

Second, contra the standard story in the world of film geeks, the '80s were a high point for originality -- more so than the New Hollywood period of the later '60s and early '70s, few of whose movies brought in many viewers. It was still a somewhat out-there thing to be making original movies in the '60s, and they struggled to find an audience. By the '80s, everybody was in the mood for exciting new stories, so there was no incompatibility between being mainstream and being original. The '60s were just the beginning; the culmination arrived in the '80s.

(That pattern shows up in popular music, too -- in the '60s the Velvet Underground was, well, underground, but David Bowie and Peter Gabriel topped the charts in the '80s. And the Velvets were just pointing the way back then.)

Third, it is true that the '50s were a period a lower creativity -- not zero, but just compared to the reversal that would begin in the '60s and peak in the '80s.

Fourth, even more importantly, the world was not static before 1960. Movie narratives weren't that unoriginal in the mid-'30s -- they gradually moved from there toward the peak of adaptations in the late '50s. Nobody gets more confused about history than when they think about the '50s and '60s. Yes, the '60s were a break from the '50s, but the conditions of the '50s didn't stretch back forever, or even several decades. The '20s were just about the opposite of the '50s; they pre-figured the '80s. These changes in the zeitgeist are more like cycles than a one-time disruption of a formerly static equilibrium.

(Similarly, popular music from the mid-'30s wasn't that bland -- it had only begun to come down from the height of the Jazz Age, when the focus was on fun, danceable melodies. It took decades of gradual erosion to reach the chart-topping hit "How Much Is That Doggie In The Window?" Rock 'n' roll was still off in the distance during the mid-'50s. And in architecture, the mid-'30s had only begun to wipe off the ornamentation and boil the elegance out of Art Deco. The Streamline Moderne of the later '30s and early '40s was just the first step toward the soulless mid-century minimalism.)

Fifth, economics tells us little or nothing about long-term changes in the creative part of movie-making. When we look at the first graph, we don't see the business cycle, or whatever, staring back at us.

It's too bad I can't apply these methods back to the rising-crime period of ca. 1890 or 1900 through 1933. I bet you'd see something like the 1960-1990 period, an initially higher level of adaptations that fell off to a bottom sometime in the '20s or early '30s. Yeah, I know they adapted Frankenstein and All Quiet on the Western Front, but lots of hits were not adaptations of specific works but of common legends (Intolerance, Robin Hood, The Thief of Bagdad). Not to mention all of the blockbuster classics that had original narratives, like Metropolis and King Kong.

Finally, let me emphasize yet again that this is only about how original the narratives are. Obviously a lot else contributes to overall originality, like the equally important visual elements. And not all adaptations have the same snore-inducing effect on a normal person. At least those adaptations from the '50s weren't based on children's cartoons and toy lines, although each year might have had several campy-looking musicals back then, so even they weren't as mature as we tend to think.


Data on the box office top 10 come from the various Years in film pages at Wikipedia. I then read through the Wikipedia entry on each movie to see what the source material was. Given how popular these movies are, and how easy it is for the editors to find out if a hit movie was based on a hit play or novel, this information wasn't lacking.

We all know what a sequel and prequel are, but just to clarify, I included both plot-based sequels, where the narratives are inter-related across the movies (like The Empire Strikes Back), and sequels based on the same character, tone, and themes, but with little or no continuity in plot (like most of the James Bond movies).

An adaptation is a different way of being less original in the storytelling, so I coded it separately. Just about the entire literary side of a movie can be grafted in from a novel, play, previous movie, TV series, etc. If the Wikipedia entry said it was only loosely based on the source, that the plot was extensively altered, etc., I didn't count it as an adaptation. In those cases, maybe they were trying to ride on the coat-tails of the source's success, but at least they made up an original story.

I counted anything that seemed long enough in form to be used in a movie -- not a single short story, not a news or magazine article, and so on. Serial stories, series of shorter children's books, etc., were counted. Biographies, autobiographies, memoirs, non-fiction books, and the like I judged case by case. If the real events described, and the subjective tone relating to them, were well known and almost legendary, I didn't count it as an adaptation of another's original work. If they were more stylized, bringing less known events to light, etc., I gave the original one credit for creating something, and counted the movie as an adaptation.

After awhile, even original works take on a legendary status, so that a modern movie-maker could still give their take on it, rather than borrow the plot entirely. There's no clear way to decide what's legendary and what isn't, so I just considered any work older than roughly 100 years at the time of the movie to be a legend, myth, fable, tale, etc., whether we know exactly who originated it (Romeo and Juliet) or not (Aladdin). If the movie was based on a newer work, which in turn was based on an older legendary work, I counted that as an adaptation, like the 1959 movie Ben-Hur that is based on a novel from 1880, although relating stories from the Bible.

Overall, though, these somewhat puzzling cases were not common in any single year: most adaptations are from earlier movies, stage plays, novels, and other long forms of fiction. So, quibbling wouldn't alter the 76-year pattern in the graphs.

Lastly, if a movie was a sequel to an adaptation, I counted it as both. So the first Star Trek movie was just an adaptation, and the rest after it were both.


  1. Any comment on this quantification of the declining complexity of pop music? There's a lot more data points, but it actually shows the 50s as having some relatively complex pop tunes. A Slate writer gave the example of "Unchained Melody", featuring seven chords compared to the three of "Call Me Maybe". They try to argue against the conclusion by saying that rhythm has become more important and they didn't measure that, but I don't know if I'd call modern drum machine nonsense rhythmically complex and Damon Che can't carry the world of music on his shoulders.

  2. So why are Millenials/Silent Generation so obsessed with sequels and adaptations? Fear of risk?

  3. So why are Millenials/Silent Generation so obsessed with sequels and adaptations? Fear of risk?

    Sequels are a linear trend beginning in the 70s. So no real need for a milennial specific explanation.

    My guess is that it's all about the Youthquake again.

    Old Boomers and aging Generation xers want to see their old shit made new again, more than they wish to see new stuff.

    And the size of the Millenial population is pretty small in comparison, so they go along with it. The Millenials do have their own lame adaptations like Harry Potter and so forth (Hunger Games, Twilight, and what have you) but I still see the trend as primarily driven by old people who don't want new things (what possible nostalgia can a Millenial feel for Transformers? that's got to be some dorky Generation X manchild thing).

  4. I'd disagree with that. I don't think Gen-Xers are going to see those movies.

  5. Gen-xers also apparently don't care about global warming. Half Sigma should be pleased.

  6. I think it's cross-generational, part of the larger zeitgeist. I don't see it so much being driven by nostalgia, since the mid-century flood of adaptations weren't necessarily nostalgic. Often they were adaptations of plays and novels that were less than 10 years old.

    On the producers' side, they don't feel as strong of a push to make something original to last for the ages -- things are so safe and stable, what's the rush? The greater conformity of falling-crime times also makes them more hesitant to try out new material.

    I don't see too much on the consumers' side. They usually wouldn't know if a movie was an adaptation, so I don't see them expressing a strong preference for less original material.

    It's mostly a long case of writer's block on the producers' side.

  7. "Any comment on this quantification of the declining complexity of pop music?"

    The variety in pitch sequences showed an increase from the late '60s or early '70s through at least the mid '80s, then more or less fell off and stagnated after that, and also had declined in the early part of their series.

    They summarize this as no change over time, just because it didn't move steadily in one direction. But the actual picture shows a good deal of change. It's one of the figures in the Supplement.

    The stuff on timbre variety seems a little off. They're measuring variation by comparing songs to one another within a year. That's not my idea of timbral variety, which compares the "voices" to one another within a single song.

    So disco had a lot of timbral variety -- strings, horns, decent drum kit, piano, synthesized keyboards, electric guitars and basses, acoustic guitars... But if most disco songs featured a similar broad range of timbres, that year would get counted as "homogenous."

    One other thing that I worry about is that they used over 460,000 songs over 55 years, or roughly 8,000 songs per year. That's like studying 200 newly released songs from every week.

    That could be studying too broad of a sample, not necessarily the most representative songs of the time.

    And they didn't say if the number of songs is roughly the same from each year. That would pose a problem if the earlier years have much fewer songs, and the later years many more songs: the former are probably higher quality, and got sifted out of the whole mess, whereas the latter haven't been screened for quality.

  8. Great work, you're most likely right, but can I play devil's advocate a bit?

    Alternative hypothesis: could it be that in the 1950s Hollywood had the problem of HUAC, and 1990s the problem of the collapse of the soviets?

    Not that I think HUAC was a bad idea---if you're a member of a group whose stated goal is the overthrow of the U.S. government and the subversion of its way of life, I think the U.S. government has the right to investigate/put pressure on you to go away---but that, because by the mid 1950s many of the biggest writers and many of the big stars were commies or commie-sympathetic and Hollywood was under pressure to keep them out, that the blacklists and graylists meant that many of the big names weren't around for original ideas, hence adaptations/sequels became popular---its easier for bad writers to adapt well-written source material to the screen or continue an adventure for a second time.

    This also may explain the modern blandness----because, post-Cold War, many of the last remnants of Hollywood's anti-left have been pushed out. With the Soviets no longer an overt threat, a John Milius-type would never even get traction today, let alone rewrite for major directors. Heck, the biggest two non-lefties in film today are Clint Eastwood and Mel Gibson---the first a senior citizen and a neocon, the latter banished for daring to bitch about the Jews, despite making one of the highest-grossing movies of all time (Passion of the Christ), bringing back the Hollywood epic (Braveheart), and being a guaranteed box office draw.

    Without an opposing viewpoint around, Hollywood treats its blockbusters as pure money makers with no artistic value, just tent poles to make the films their studios actually want---politically friendly small films. It's only when the opposing viewpoints are around that Hollywood feels challenged to put original ideas out for the masses---because the those screenwriters with the non-popular views are shunted to the blockbusters, despite any talent, which improves the blockbusters.

    Anyway, like I said, the evidence is far more on your side here, but this idea does tickle me a bit.


You MUST enter a nickname with the "Name/URL" option if you're not signed in. We can't follow who is saying what if everyone is "Anonymous."