This came up in two books I read recently: “Elements of Surprise,” by Vera Tobin (who has a Ph.D. in English and teaches cognitive science) and “How Music Works” by David Byrne (Psycho Killer, etc.). Tobin’s book is about plot and suspense in stories—she talks mostly about books and movies. Byrne’s book is about music, live and recorded.
Here’s Byrne, describing one of his stage shows which he derived in part from Kabuki and other traditional modes of Asian theater:
There is another way in which pop-music shows resemble both Western and Eastern classical theater: the audience knows the story already. In classical theater, the director’s interpretation holds a mirror up to the oft-told tale in a way that allows us to see it in a new light. Well, same with pop concerts. The audience loves to hear songs they’ve heard before, and though they are most familiar with the recorded versions, they appreciate hearing what they already know in a new context. . . .
As a performing artist, this can be frustrating. We don’t want to be stuck playing our hits forever, but playing only new, unfamiliar stuff can alienate a crowd—I know, I’ve done it. This situation seems unfair. You would never go to a movie longing to spend half the evening watching familiar scenes featuring the actors replayed, with only a few new ones interspersed. And you’d grow tired of a visual artist or a writer who merely replicated work they’ve done before with little variation. . . .
So here’s the puzzle:
With stories we value suspense and surprise. Lots of interesting stuff on this topic from Tobin. Even for books and movies that are not “thrillers,” we appreciate a bit of uncertainty and surprise, both in the overall plot and in the details of what people are going to say and what comes next. But in music we value familiarity. A song or piece of music typically sounds better if we’ve heard it before—even many times before. The familiarity is part of what makes it satisfying.
In that way, music is like food. There’s nothing like “comfort food.” And, yes, we like to explore new tastes, but then if you find something new that you like, you’ll want to eat it again in future meals (at least until you get sick of it).
Yes, literature has its “comfort food” as well. When I first read George Orwell, many years ago, I liked it, and I read lots of other things by him. I like Meg Wolitzer’s books so I keep reading them. They’re all kinda similar but I like them all. OK, I have no interest in reading her “young adult”-style books, but that fits the story too, of wanting to stick with the familiar. And, of course, when it comes to movies and TV, people love sequels.
But there’s a difference. When I read one more book by Meg Wolitzer or Ross Macdonald or whoever, yes, it’s comfort food, yes, it’s similar to what came before, but there’s plot and suspense and a new story with each book. I’m not rereading or rewatching the same story, in the same way that I’m rehearing the same song (and, yes, it makes me happy when I hear a familiar REM song pop up on the radio). And, yes, we will reread books and rewatch movies, but that’s just an occasional thing, not the norm (setting aside the experience of small children who want to hear the same story over and over), in the way that listening to a familiar album is the norm in music listening, or in the way that when we go to a concert, we like to hear some of the hits we’ve heard so many times before.
So. With stories we like suspense, with music and food we like familiarity. Why is that? Can someone please explain?
One explanation I came up with is that when we listen to music, we’re usually doing other things, like jogging or biking or driving or working or just living our life, but when we read, our attention is fully on the book—indeed, it’s hard to imagine how to read without giving it your full attention. But that can’t be the full story: as Byrne points out, we also want familiarity when seeing a live concert, and when attending a concert we give it as much attention as we would give a movie, for example.
Another twist is that surprise is said to be essential to much of music. There’s the cliche that each measure should be a surprise when it comes but it should seem just right in retrospect. There are some sorts of songs where the interest comes entirely from the words, and the music is just there to set the mood—consider, for example, old-time ballads, story songs such as Alice’s Restaurant, or Gilbert and Sullivan—the music is super-important in these cases, and without the music the song would just fall apart, but there’s no need for surprise in the music itself. The music of Sullivan is a perfect example, because without Gilbert’s words, it sounds too symmetric and boring. For most songs and other pieces of music, we want some twists, and indeed this seems very similar to the role of plot and surprise in storytelling. I wrote about this before: “Much of storytelling involves expectations and surprise: building suspense, defusing non-suspense, and so on. Recall that saying that the best music is both expected and surprising in every measure. So, if you’re writing a novel and you introduce a character who seems like a bad person, you have to be aware that your reader is trying to figure it out: is this truly a bad person who just reveals badness right away, is this a good person who is misunderstood, will there be character development, etc.”
But this just brings us back to our puzzle. Surprise is important for much of the musical experience. But when we listen to music, unlike when we read or listen to or watch stories, we prefer familiarity, even great familiarity. You might say that this is because only with deep familiarity can we really appreciate the subtleties of the music, but (a) we often prefer familiarity for very simple music too, and (b) that same argument would apply to stories, but, again, when we receive stories we usually prefer surprise.
So the puzzle still remains for me. I guess that something has been written (or sung?) about this, so maybe youall can help me out.
Derek Thompson explores this in depth in The Hitmakers. Also worth examining Archer and Jockers, The Bestseller Code, which uses NLP to analyze successful fiction.
“One explanation I came up with is that when we listen to music, we’re usually doing other things, like jogging or biking or driving or working or just living our life, but when we read, our attention is fully on the book—indeed, it’s hard to imagine how to read without giving it your full attention.”
I wonder if this can be tested in some way. Do people prefer listening to audiobooks (while jogging or biking or driving or working or just living their lives) that are less suspenseful and more familiar than their printed counterparts?
On the subject of attention, when I used to try to run fast (fast for myself, not in the absolute sense), I found podcasts/audiobooks annoying. Like it felt like a big distraction, I remember it like gnat in the ear.
But then just chilling and whatever running, podcasts/audiobooks rule.
I asked someone who was a better runner than me and they were like oh yeah definitely no audio. Don’t know how widespread that is (commercials seem to suggest people like to go hard with their earbuds), but N = 2 says otherwise.
I have to push back on this. Many people use music as their primary activity: as per attending concerts, or listening in critical environments. This principle still holds even if attention is not distracted. However, music produces a more holistic excitation of our brain across all areas in ways that natural language does not. Music is also parallel processing: we listen to many different parts at the same time (melody, harmony, rhythm, color, interplay), and so each time we travel a piece of music we can pick different “paths” of attention. However each path confirms the overall structure of the piece, and we arrive satisfyingly at our ultimate destination, but have had a different experience due to where our attention took us. A story, on the other hand, is sequential: we read words, and each word has its place in a sentence.
I may predict that if a story had alternate paths (“what if the DON’T take that action and instead to THIS”), that the experience may be similar, because the story can change. But this is merely speculation.
There’s books I like to know about before reading. Like old stuff can just be hard to read, but there’s sorta context that makes it amusing or interesting if you know about it going in (about the author, about the reception, what’s new and cool about this specific book, etc.). Actually, on that last point, maybe the point is that if a book is first widely known book to deploy some literary device, then that book won’t be special if you don’t know that. So you need to know the surprise going in, or it won’t be a surprise.
New music rocks too. Last time I listened to Pandora it was doing these Featured Track advertisements where I’d get random songs in the genre of my channel that rocked. Woulda preferred Pandora kept playing me random Featured Tracks but I guess those cost someone money to put there, and maybe the selection wasn’t actually that large. I looked up a couple of the songs and they seemed like small indie things. In a very hipster way it was fun to think like, damn! I’m pretty early on enjoying these bands! (here’s one I still had a link too: https://www.youtube.com/watch?v=812Wv87ebr8 [Bus by PANDAMIC if link fails], like brand new Blink-182, so is that new cause it’s new or familiar cause it’s like Blink-182?)
Re: re-reading the same books, the Hobbit is pretty solid comfort food for me, and you can know the story without knowing it all (was the troll encounter Hobbit? Or was it LOTR? I forget — I think Hobbit — what a good excuse to read the story again).
Okay that’s enough disagreeing for this topic. I really like familiarity of Christmas music. I like singing it and I mess up the words and forget them and then remember them — it’s a good time.
I don’t like being dismissive, but they’re completely different things. Even music that tries to tell a story (The Wall, Imaginos, Joe’s Garage…) is very different from just straight up stories you get from books, movies or videogames, etc. It’d be like asking why we expect sweetness from ice cream, when we expect spice from curry. I suspect that our brains just processes music and (conventional) stories in different ways. There might be some established neuroscience literature that talks about this.
“The music of Sullivan is a perfect example, because without Gilbert’s words, it sounds too symmetric and boring.” Disagree completely. Sullivan’s music is wonderful on its own.
One hypothesis is that it is due to the distinction of aesthetics of form and those of content. Music moves us by its form whereas literature by its content. One thought experiment is how much of the joy of listening to a familiar song can be derived if _only_ the tune is played or if _only_ the lyrics are read to you. I reread poetry quite a bit more than novels because the form adds to what is pleasing in poetry.
Then again it doesn’t apply as cleanly to comfort food vs haute cuisine, so the explanation is probably bogus.
Having been involved with three types of music (classical, folk (fingerpicking), and now jazz guitar) (on and off, but mostly on) over the last 65 years, much of the commentary on music makes no sense to me. For starters, I largely can’t listen to music while doing something else. I’m an extreme non-multitasker. If it’s good enough to listen to, it’s good enough to deserve a careful listen.
One point is that music is complicated, and for music that’s reasonably decent (i.e. has musical and/or intellectual content) there’s a lot going on, so it has more to offer on each listening. The structure and details of a good bebop solo, or pretty much anything from the classical war horse repertory, would be examples. Lots of pop songs have really kewl arranging going on, even the musak in your dentist’s office often have wild arrangements.
Still, I think you are right that music and prose fiction are radically different. When I read a novel, even a schlock one, I want to be told something new. When I listen to music critically, I’m listening for ways to speak to a listener through music, how the lines are shaped dynamically, how the rhythm works. But just enjoying the melody, rhythm, how the solos work against the harmony can be done over and over again.
Mis-nested this comment. Belongs here:
https://statmodeling.stat.columbia.edu/2022/01/11/suspense-in-music-suspense-in-stories-how-do-they-differ/#comment-2042308
RE: “comfort food.” I have argued that part of this is that different sensory modalities are differently connected to pleasure and pain: we are picky about taste and smell but less picky about sight. Hearing is in the middle. So we can enjoy random splashes of paint on a canvas, but not random smells or tastes (drawn from the universe of possible smells and tastes, not just food tastes). Random notes are in the middle; John Cage made a career out of them but he’s very much a niche taste.
So one stab at an answer is that familiarity is more important when we are more picky about the stimulus.
Avante garden jazz and atonal or modern classical music, in my experience, require rather focused concentration. They just don’t work as background music. Kind of funny to even think of them that way. There’s a likely reason why muzak amhas the features it has.
Dmitri:
You can look away from an ugly sight but you can’t look away from a sound or smell in the same way. Taste is different if you get to choose what you eat.
It might be worth breaking out both stories and music into written stories and recorded music, versus stories spoken aloud and music performed live. (If anyone else takes this up I will provide anecdotal personal data from a real outlier, me, someone who has reread several hundred mystery stories in the last year.)
I’ve never thought of this as a puzzle at all!
With music in a way it’s even worse than David Byrne suggests, because even his new songs are old songs! Virtually all popular music is built on the 1-4-5 chord progression, with the occasional 2 or 6 chord to liven things up. Jazz is more complex and correspondingly less popular.
But OTOH, stories are also built on standard progressions. I’ve heard it said that Shakespear was the last original author.
Maybe the main difference between music and stories is the length of the pattern. In music, the pattern is usually only a minute or two long and then repeats. In stories, the pattern is more detailed and takes much longer to develop, but all stories follow a similar pattern. Stories have patterns and cadence that are analogous to the patterns and cadence in music. In both music and stories, tension builds through the pattern, climaxes and is resolved at the end of the pattern. In music, for example, often the dominant 7th resolves into the major at the end of the pattern (the blues turnaround for example). In stories the tension between the characters also builds throughout the story, with new twists and minor resolutions, then resolves completely at the end of the story. And at least for my money, stories by a given author and songs by a given musician aren’t much different. Read one novel by Author X and you’ve read them all. Heard one Taylor Swift or ACDC song and you’ve heard them all.
Also, people do watch shows and movies over and over. They don’t usually go to the theatre to see a movie twice, but that’s because it will be available much cheaper soon enough. They also listen to the same music over and over. And people do pay to see **live** stories – theatre – over and over, just as they pay to see live music over and over. How many times have I been to the Nutcracker or seen a performance of A Christmas Carol (four or five of each)?
You are right that popular music reuses a relatively small number of progressions. However, its harmonic vocabulary is pretty different from the I, IV, V (and some ii and vi) vocabulary of earlier music. Starting in the 1960s there’s a turn to modality (importantly drive by the Beatles, I believe) and a real change in language. So a song like “Psycho Killer” is based on VI-VII-i, which is not a common jazz or classical-era progression.
So that’s interesting too: you have a dramatic shift in harmonic vocabulary during the 1960-1970 period and then the harmonic language stabilizes again.
《Jazz is more complex and correspondingly less popular.》
When Swing was the most popular music, was there a lot of improvisation? Playing along with once-popular swing bands today, am I not continually surprised by drum riffs, solo phrases, stop choruses, etc. even if I’ve heard the tune many times? And once I learn the tune, why do I want to improvise on it like David Byrne? Were people just generally more musical in the 1930s, so musicians were able to indulge their love of variety more than today?
Lately I’ve been riffing off a few jazzy Brian Setzer patterns. I love finding an interesting pattern and exploring the “riff space” in it!! But that seems to me like variations on the main “plot”. Another way of saying that is that “riff space” constrained by the “plot” or larger structure.
Maybe extending the analogy a bit we can think of the progression and key as the overall “plot” and the verses, choruses, bridges, movements and various subcomponents as chapters that have some resolution, while the riffs that comprise those chapters and close them are the specific actions of the characters. Musically swing jazz is more complex than, say, ACDC. But it still has a very prominent structure (as do stories).
> And people do pay to see **live** stories
Oh I forgot about the live angle — this is tangential but live sports vs. tape delay vs. watching a recording later vs. re-watching. Big differences for different people! With competitive things I like getting them live and I like the surprise a lot.
With Starcraft 2 tournaments I liked the live surprise of watching, but then it was also really fun to rewatch the replays a lot and try to learn how to mimic what looked interesting. So the surprise was good, but then the repetition was good too.
Jim:
I thought of some of these issues when writing the above post. In particular, see my discussion of Meg Wolitzer etc. Or if you want an even more extreme example, consider a writer of series mysteries such as Erle Stanley Gardner. The Perry Mason books are kinda the same story over and over again, but people would buy new ones. Each new story has its own plot and own surprises. Even though you know the general form, there’s pleasure in the novelty.
Similarly, as an REM fan I like hearing their familiar songs on the radio, and I like to play their old records. But if I could hear a record of new REM songs, that would be great in a different way. That’s an easy one for me to think of, because I can just put myself back a bunch of years to a time when I hadn’t heard all their songs, and I remember how fun it was to hear ones that I hadn’t heard before.
Regarding your last point: sure, when I was a kid we watched The Wizard of Oz every year when it showed up on TV—it was a national event each year (I think it was in February). And, sure, there are some movies or theater performances you’ve seen many times, and some books you like to reread. But I think these are the exceptions: usually people prefer to read new books and see new movies. Theater is a bit of an exception—people will keep seeing Shakespeare plays; in that way they’re more like music, as David Byrne notes in the quote in the above post.
Interesting thoughts Andrew. I dig old REM, but I’m not sure I would be cranked to hear new REM. For me most bands hit a stride I get into for three or four albums, then either become repetitive or drift into new directions that I find less appealing. I’ve found the same with writers, fiction and non-fiction alike.
To some extent that’s probably part of what’s frustrating for David Byrne. It’s not that people aren’t into “new”. It’s that they don’t like his “new” as much as his “old”. Ha, if you think about that he is experiencing a cognitive bias!! Hilarious. Very few songs or musical acts (or artists of any kind in general) become popular – the real anomaly is that any of his music was popular at all. The appropriate comparison of his new music is to some other new unknown song by an unknown artist, not to his previous songs. In fact his new music might be much less likable and the only reason anyone starts to like it at all is because it’s played a lot based on his reputation.
That’s an interesting angle too: how an artist’s previous popularity creates an exceptionally positive reception of newer work, skewing it toward popularity.
Jim:
I don’t like the last two REM albums, and I didn’t like the one they did after Automatic for the People, but I love all their other albums, all the way from 1983-2004. (I didn’t really start listening to them until the 90s, but I did go to an REM concert in 1989, I think it was. I didn’t like the concert so much, because I wasn’t familiar with their music!) Anyway, I don’t know if I’d enjoy a hypothetical new REM album, but if a “lost” album from the 1983-2004 period were discovered with entirely new songs, I’d probably love it—after I’d heard it a few times, of course!
I guess it depends upon what you mean by “original”, but I think it’s odd to call Shakespeare the last original author. The stories he told certainly weren’t original, nor did people expect artists to be “original” in any modern sense of the term. Originality in the modern sense doesn’t come until the romantics.
I find this view to be pretty reductionist. A usual 4-chord loop is indeed the building block of modern pop music. But the idea that the songs are actually the same, or that this represents a net loss in musical complexity is just ignoring the role of production entirely. If all pop music were transcribed to just what can be represented through musical notation, then played back as pure sine waves, then sure yeah you’d be right. But what’s actually happened is that older musicians had a standard set of instruments and played with arrangements or improvisation, while most new musicians today use machines that can produce effectively any sound they can imagine and so play more with production and lyricism. The idea that this:
https://www.youtube.com/watch?v=yydNF8tuVmU
is musically identical to this
https://www.youtube.com/watch?v=1J26YBEN9ys
At an even more extreme end, stuff like this:
https://youtu.be/EV95Yu6gZSY?t=92
or this
https://youtu.be/fU6qDeJPT-w
would look utterly boring judging by transcription, but can have profound effects on actual listeners. You can say that’s due to things outside of the core musical content, but I’m just not interested in that kind of a semantic debate. You could also say that, in most cases then, the “REAL musicians” of modern pop are the producers and engineers that nobody knows, and the names we recognize are just faces–and you might have a point, but that’s a different discussion.
Many people like to sing along or hum along with the music – hard to do if it is not predictable. Most of us don’t memorize stories and try to repeat them verbatim. Perhaps audiobooks change that somewhat, but I doubt it.
On repeat consumption, this paper is relevant and includes distribution of number of times items are consumed across a few different settings, including music vs. non-music video. https://www.cs.toronto.edu/~ashton/pubs/repeat-consumption-www2014.pdf
Dean:
Scientific papers are another example! If a paper is worth reading, it’s almost worth reading more than once. Upon repeated reading, some papers become more and more interesting. But bad papers upon repeated reading just fall apart. Sometimes, though, this repeated reading has to be done at a societal level, not just at an individual level.
For example, that Bem (2011) paper on ESP looked kinda reasonable at first, and lots of informed observers thought it was a solid bit of research, even if it had iffy conclusions. When I first looked at that paper, I was sure there was something wrong with it, but I couldn’t quite put my finger on what it was. A few years later I looked at the paper again, and its flaws were obvious. My best analogy here is how Conan Doyle could get fooled by those photos of fairies, photos which, to a modern eye, are obvious fakes—it’s not even close.
> Scientific papers are another example!
In other genres, comics? Far Side at least is good for the surprise and for the rereads.
Ben:
Jokes! That’s a great example. Jokes involve surprise, and it’s fun to tell jokes over and over. But not quite like music, because it’s not so fun to hear a joke over and over. For example, I can pleasurably listen to a family member sing the same song many times, but it’s not so much fun to hear her keep repeating the same joke. Even though the song is 3 minutes long and the joke just takes 10 seconds.
Actually, another new genre in the surprise vs. familiarity, people!
Talking to the same people all the time vs. talking to new people, for instance.
Maybe there are a bunch of similar generalizations that make generalizations in general less interesting. Places, food, etc.
First, I agree with the people above that this just isn’t a universal observation. I think there are plenty of people who like novelty in music and familiarity in literature/visual entertainment.
But to the extent it is generalizable, I would just guess that there are completely different parts of the brain processing one versus the other. Taste and smell and hearing are really old parts of the brain and like (pardon the anthropomorphizing here) repetition. Vision is really old, too, but the language comprehension parts of the brain are much newer and thrive on variety.
Again, I wouldn’t make too much of this, because all parts of the brain are involved in everything, and different people process things with different parts of the brain… but that might help explain why this isn’t universal. People who process music like language crave variety and people who process words like smell want comfort…. or something like that.
Expanding on this slightly, this might help explain why small children exhibit such different behavior in language. Until the language functions have matured (and other senses as well) variety is too scary to be enjoyed.
A craving for variety may be a sign of maturity. The old endorphins just stop kicking in some part of the brain after enough repetition and new stuff is no longer scary. But the stimuli for which that happens vary across individuals.
I was about to comment that if we’re going to compare reading to music, we should also make a comparison to taste / smell; each is very different from the other, and I’m not sure we should expect similar patterns.
For taste, one could argue that we like familiar things (comfort food) or that we like novelty (see the ridiculous proliferation of flavors in breakfast cereals or potato chips). It’s hard to even state what’s familiar and what’s new. A few weeks ago at “Bombay Burger” in Seattle, I had “masala fries.” It was excellent — because it combined familiar tastes (spices, french fries) or because it was novel (spices + french fries)?
I think the difference is between listening to music and reading (or listening to music and watching a movie) rather than literature in general. In fact, a lot of live performances probably involves stories the audience already knows, whether theatre (and not just Shakespeare) and poetry readings. I suspect it’s reading and watching movies that’s the odd one out.
I believe this is because listening to music, even music with lyrics, is much more a directly sensory experience than reading or following a visual story. Comparing music to other more sensory experiences, even visual ones, this phenomenon isn’t so surprising.
Consider getting a massage or going for a run. One does not mind the absence of novelty in those cases either. For a better visual comparison than film, consider strolling in a garden or watching a sunset. Obviously novelty and surprise play a role in all these experiences, enough that we can compare them and notice new things. But generally speaking they rely much, much less on novelty or surprise than stories do. Music maybe straddles the boundary to an interesting degree, but really I think a huge amount of our enjoyment of music is just because it does something sensory to our brains that just feels good. And familiarity, up to a fairly high point, only enhances the effect.
This also fits with your observations about food. So it seems that stories, whether read, seen or heard, may be cognitively a different type of thing.
Matt:
The “going for a run” example is interesting, because when I go for a run, or a swim, I’m not expecting novelty at all. But when I play a sport that involves running or swimming, I do get joy from novelty. As the man said, “The symphony’s the same every time. Football, every game is different.”
Perhaps that’s related to the fact that those sports have more of a narrative feel to them and the accompanying tension/resolution relates more to the story of the match than the sensory experiences of physical exertion?
I don’t think any of these distinctions are absolute, of course…
I think you’re starting off from a false dichotomy:
“With stories we like suspense, with music we like familiarity. Why is that? Can someone please explain?”
We also like familiarity in our stories. Most stories follow a familiar structure. This paper suggests the emotional arcs of stories are dominated by six basic shapes https://epjdatascience.springeropen.com/articles/10.1140/epjds/s13688-016-0093-1 (e.g., rise and fall). The musical counterpart to this would be song structure. Most songs follow familiar structures (e.g., verse-chorus-verse-chorus-bridge-chorus, 12 bar blues, ii-V-I, etc.).
Both in stories and music, these structures create anticipation; there’s an emotional payoff when a major plot point is revealed, or when you get to the chorus. If you read the same story or listen to the same song a bunch of times, you lose the surprise part of this anticipation making structure, but there’s still emotional payoff. For example, if I rewatch Game of Thrones I know King Joffrey will die, but it’s still going to be a satisfying part of the experience when it happens. And I’ll still be anticipating it because that anticipation is baked into the structure of the story. Same thing with music, you’ll still anticipate the chorus of your favourite song on repeat listens because great songs are designed to do that.
There are other comparisons I could make between stories and songs here to demonstrate the false dichotomy of suspense and familiarity, but this is good enough. I would suggest watching Rick Beato’s “What makes this song great” series on YouTube though—he breaks down a lot of these elements of music, and it’s really interesting diving into the finer elements of songs you love. As you noted in your post Andrew, surprise is essential to much of this.
Now your next point:
“But that can’t be the full story: as Byrne points out, we also want familiarity when seeing a live concert, and when attending a concert we give it as much attention as we would give a movie, for example.”
If I go to see the Talking Heads live, I expect them to play Talking Heads songs, not Dead Kennedys songs. If I go to a theatre to watch the new Batman movie I expect it to be about Batman, not Superman. And as discussed above, I expect Batman to go through a familiar character arc before the movie ends. A villain will do something villainous, Batman will try and fail to catch them, he’ll brood for a bit, then finally catch them in the end. I also expect the Talking Heads to play familiar music. In both cases I’m going to the movie/show to experience a specific genre that elicits specific emotions.
Now, maybe the director decides to mix things up: This time Batman dies. If the story is well crafted and the actors do a good job, this surprise might be exciting. But crucially, it will still be about Batman. Similarly, if the Talking Heads decide to play new songs at their show, and the songs and performances are good, that would be exciting (and surprising) too. But crucially, the new songs still need to be recognizable as Talking Heads songs.
So I think we want familiarity in both these cases (and you could apply this to whatever genre of film or music you wanted). If your main goal is to be surprised and experience something novel, you would probably choose to go to a film festival or music festival, where there’s a variety of things to experience; many of them unfamiliar to you.
Moving away from the false dichotomy, I think the bigger question underlying your post is something like “Why do we relisten to the exact same songs more often than we rewatch the exact same movies?”. Or phrased in a different way, “Why do we watch new movies, but listen to the same old songs?”. I think this has less to do with surprise and familiarity, and more to do with things like how music and movies are owned and distributed. I’m also not sure how true it is anymore now that music streaming is as widespread as movie streaming.
Michael:
1. I don’t think Byrne’s complaint was that a Talking Heads audience wanted to hear Dead Kennedys songs. I think his complaint was that a Talking Heads audiences wanted to hear old Talking Heads songs, not new Talking Heads songs (even though the new songs were future Talking Heads hits). In a concert, the audience might expect 50% old songs and 50% new, which is fine, sure, that’s what I’d want at a concert too, but, as Byrne points out, when people to the movies they’re not expecting 50% of it to be a shot-for-shot remake of an old movie they’ve already seen.
2. Yes, new stories have lots of familiar forms to them, just as does new music from the same artists. But, again, the point is that with books and stories we will typically want something new within that form, whereas with music we will often want to hear the exact stuff we’ve heard many times before. I guess I’ve listened to Radio Free Europe more than I’ve ever read any short story, for example, and it’s not like Radio Free Europe is so damn amazing, it’s just that it comes on the radio a lot and I like it every time, also sometimes I play it when I’m running (to connect to a different comment on that thread). There are short stories I love, and I could read them dozens of times, but I don’t. I read new stories instead.
3. I get your point in your last paragraph. Back in the day, I guess lots of people owned just 10 books or whatever, and they spent lots of their leisure time reading, so they’d read the same books over and over again. When I was a kid, I read some books many many times, not so much because I wanted to but rather because that’s what was there on the shelf. That said, there’s still the enjoyment factor. We watched zillions of Gilligan’s Island reruns and hated it every time, but we watched it because that was all that was on (or, I guess, whatever was on the other channels was worse).
As someone who gets a lot of aesthetic enjoyment from both writing and music, I’ve thought about this paradox for a long time. I have a loose sense, very proto-theoretic, that it has to do with the structure we do or don’t bring to the experience. Language is highly structured from the outset, of course, and writing has another layer on top: paragraphs, dialogue, etc. Music is not intrinsically structured in the same way; the structure is consciously created by the musician. (Yes, there are tunings and instrumental timbres, but isn’t that about it?)
With a background of so much structure, good writing strives to set it off against surprise, not only at the level of plot or characterization, but also the use of unexpected words, metaphors, etc. When I taught writing I always emphasized the importance of irregular cadences to go against the tendency toward monotony.
Music’s structure is applied. A songwriter chooses a chord progression. In classical music, the role of themes and motifs goes back to the beginning, and the history of the art form is a lot about the increasing complexity of their treatment. When you listen to Bach, you hear a tune and think to yourself, what will he do with this? The popularity of Beethoven’s #5 has a lot to do with how varied his use of da-da-da-dum is. Even more contemporary music mostly works on cells or themes, although to the point that it takes a pretty educated ear to pick them out. Music that does away with this structural basis has to supply something else to take its place, or it is just about unlistenable.
I’m sure there are philosophers and neurophysiologists who have gone into this much more deeply, and maybe I’m on the wrong track. I took a philosophy of aesthetics course as an undergrad ages ago, and that’s it. (Wrote a term paper on Emanual Lasker’s theory of aesthetics as patterned problem-solving.)
“A songwriter chooses a chord progression.”
Not always! A song can also start from a lyric (or a partial lyric), or a melody (a measure or two, up to the whole thing), or a rhythm. Maybe other ways too that are less obviously sensual (e.g., an emotional sequence or more complicated scheme of emotions),
If I’m going to spend $300, drive for 2 hours, wait in line for 1.5 hours, to stand for 2 hours pressed into a tin can with 60,000 other people, I need a near certainty that I’m going to enjoy myself. What’s more, me enjoying myself depends on sensing the enjoyment of everyone else, all jamming along. If I could pay $18 walk in to see Anderson.Paak at my local theater in a big comfy chair with 150 other people, I’d hope he’d play us an all-new album.
Hi Andrew,
I have exactly the book for you! It’s called “Sweet Anticipation: Music and the Psychology of Expectation” by David Huron. To quote the Goodreads blurb:
“Huron proposes that emotions evoked by expectation involve five functionally distinct response systems: reaction responses (which engage defensive reflexes); tension responses (where uncertainty leads to stress); prediction responses (which reward accurate prediction); imagination responses (which facilitate deferred gratification); and appraisal responses (which occur after conscious thought is engaged). For real-world events, these five response systems typically produce a complex mixture of feelings.”
Or in other words, the key to resolving your paradox is that there are different types of “expectation”.
I have read the book in its entirety and couldn’t recommend it enough. One of the best books that I’ve ever read.
The factor time works in a completely different way in music. Reading a story, we choose the timing. If we miss something we can go back and re-read it. We can do breaks. I read 20 pages today 20 pages tomorrow and so on. Timing is totally part of the essence of music though. Every element has to come at the time it comes. When it’s gone it’s gone. The only way to get some kind of control over it is to decide when to listen to the same piece again.
Also (as has been mentioned before) music works directly on the body and with the body. The music experience is tied to the presence, and it’s an experience for the whole body. It doesn’t go away with repeat listens (although it can change). Stories can be spoiled by knowledge. Music cannot be spoiled by knowledge, so there’s normally more pleasure in re-listening what we love than in re-reading. We can love a story we love having it or parts of it in mind. Good music will catch us when we listen to it and won’t catch us otherwise. (I a composer friend who enjoyed music, or so he said, just reading the scores… that’s not my kind of music experience. I’ve got to admit it exists though.)
Rhythm is repetition, repetition is rhythm.
But then, nothing beats listening to something great for the first time. Try this, it may be your first time.
https://www.youtube.com/watch?v=kaZzAGR0VNk
Christian:
I agree that nothing beats listening to something great for the first time. But listening to something great for the second or third time is great too! Right now the radio is playing a great song which I guess was just released, because the first time I heard it was two days ago and the second time I heard it was yesterday . . . they have a pretty good variety of music but when they put a song on heavy rotation I think they kind of overdo it . . . I think the reason is I’ve been sitting at home all day most of the time with the radio on in the background for hours on end. I guess the target audience would be listening for 45 minutes in the car, then 20 minutes while in the kitchen a few hours later, etc., so then the heavy rotation won’t seem so heavy. Also I wish they’d show a bit more variety on the oldies.
P.S. I can definitely believe that someone could get enjoyment from reading scores straight up. That’s not me—I can’t read music at all—but can get lots of enjoyment from reading book reviews, often (but not always!) more than from reading the books themselves.
“The only way to get some kind of control over it is to decide when to listen to the same piece again.” Or (depending on the nature of the piece and one’s own abilities/inclinations/available resources) play (or sing) it oneself (or with amenable and capable others; or, for that matter, if you’re Christian Ludwig, Margrave of Brandenburg-Schwedt, just request a command performance).
I’ll agree, and eve put $20 on just the duration of the experience. 2 hours is a long time to listen to a piece of music. And lots of people love watching the same movies over and over again (indeed, apparently making the same movie over and over again.) I even like rereading short stories. Poetry is totally for rereading.
They all involve only one sleep-wake cycle. Except Wagner, proving the rule.
But if I am investing days in a novel, I expect something seriously new, because I know I am going to have to pack the idea out with me. I confess that I don’t read a lot of them; too many disappointments.
> With stories we value suspense and surprise.
Was it always like this? I think “spoiler culture” is relatively modern. Old literature used to have no twists / declare its twists beforehand. You can imagine Ancient Greeks, gathered for the annual listen of Iliad, were not expecting any surprises in the story.
You’ve spoiled modern culture. Now I don’t have the opportunity to discover it on my own! Boooo
Could depend on what the audience is capable of. Most of us can write stories. Or at least we’re savvy enough to argue endlessly about how a story written by someone else could be improved. Not many can write music or even words to music. So most of (non musicians) can’t even articulate why we like a song never mind argue about how it could be made better. Your expertise with the medium probably matters when it comes to preference for familiarity.
Another factor that probably matters is duration of entertainment unit. Songs in minutes. TV shows more minutes. Movies/short-stories hours. Novels days. This factor might help to simplify how we think about the apparent puzzle. Maybe there’s a constant level of surprise/familiarity we want per unit time?
The first factor (expertise with the medium) might also explain jokes; but not the second factor (surprise rate). But there’s likely no single factor here.
Music is about a melodic (linear) and harmonic (vertical chord structures) connection with human emotions we feel, tied to a particular period we remember in our lives. Like a soundtrack to a movie, music enhances our emotional connection to events. Slight deviations or variations from musical allusions can be delightful. Going too far off script from familiarity (e.g. Schoenberg) weakens these connections because they become unrecognizable, no longer triggering memories from a time passed.
Another point is that new music might not resonate (become as liked) as well with audiences as already established hits.
An example of this is when the Met takes a popular opera from Verdi and reappropriates the context to a different setting in modern times.
Maybe it is a hit, maybe not.
A:
I get what you’re saying but I don’t think this tells the whole story. I prefer familiarity for new music too, even if it has no past emotional resonance for me. If I hear a song I like on the radio, I’ll typically like it better the second and third time as I become familiar with it. And even after hearing it 10 times over the period of a month, I’ll still enjoy hearing it again.
Right. So if new music at a David Byrne concert illicits groans where older hit tune is preferred, listeners might revise their appreciation of the new music after the concert once they have heard it a few times and understand it better. Or not ;). But that’s what stands the “test of time”.
Lots of comments, but none so far mentions what I always thought was the definitive work on expectation and surprise in music: “Emotion and meaning in music” by Leonard B. Meyer (1957). His account is a bit too complex for me to summarize here. All his examples are “classical” music. (The most amazing analysis is the Beethoven C# minor quartet 5th movement, p. 145.) But I have noticed some mechanisms he discusses in the “run on forever” themes of Brahms, Rachmaninoff (particularly the cello sonata), Hoagy Carmichael (“How little we know” in “To have and have not”), the Beatles (various), various current popular songs (excluding the chanting of rhyming poems over a rhythm section – I don’t count these as songs), and lots of film music (Wolf Hall in particular).
Andrew, I think this is an interesting insight and I expect to ponder it occasionally until I feel there’s nothing more to ponder.
One thing that comes to mind, though, is that I think there’s a sense in which people like to see things repeated in movies and TV shows that is similar to people wanting to hear their favorite songs at a concert. Sitcoms are a good example: if you like a sitcom it’s because you like the characters and the way they interact with each other. You’d be disappointed if they acted really differently, even if the result was still a good show, in the same way you’d be disappointed if you went to a concert and the music was all different from what you expected, even if it was still good. Movie sequels, too: people want a movie that is a lot like the first one. If you liked the first Jason Bourne movie (action-adventure-spy) and loved it, then went to the sequel and Bourne was working a desk job and stumbled across some puzzling stuff that turned out to indicate there was a mole in the department, and it turned into a cerebral cloak-and-dagger story like Tinker Tailor, it wouldn’t really matter how good the sequel was, a lot of people would be very disappointed while they were watching it.
Also, of course sometimes people do rewatch a movie or TV show, just like listening to an album over and over. I don’t think anyone watches their favorite movie even 1/10 as much as they listen to their favorite music, though.
I do think you’re onto something with music being different, but maybe the difference isn’t as big as it seems from the way you framed it.
Had very similar thoughts, Phil.
Also, isn’t novelty an large appeal of live jazz/improv? I feel like there are plenty of examples for which this particular model (literature=suspense/novelty, music=familiarity) doesn’t quite hold.
abc: I think Jazz and improvisational music do not fit the mold exactly, but I also think that fans of improvisational music also are not “normal” music fans. Similar things go for say Bluegrass where a big part of the appeal for the fans is to hear something they know that sounds *different* from what they know.
For example:
Bela Fleck and friends do The Grateful Dead with Banjo and Tablas
https://www.youtube.com/watch?v=LxQjMc7NsIU
Of course the Dead were pretty eclectic, but here’s the original:
https://www.youtube.com/watch?v=crHvWsFVeD4
Not everyone is into Jam Bands, but those who are… are really into them.
Until you can repeat the song without mistakes, isn’t there always something new to learn by listening to that song over and over?
Rsm:
Sure, but that would true of stories, and with stories, movies, TV shows, etc., people would generally rather see the next episode rather than seeing old familiar episodes over and over.
If I’m studying something, may I repeat looking, tasting, smelling, touching, hearing it over and over till I get it right?
Maybe people are repressed musicians?
one thing that’s amazing is that in this long and thoughtful discussion of what makes music different than prose or speech, words like “dance”, “groove” and “rock” are entirely absent! Lots of people like music – including musicians – because the energy in music makes them want to dance or “rock out”:
https://www.youtube.com/watch?v=_OqUYgiQmnY
https://www.youtube.com/watch?v=Vus8TT0DSww
Rocking out can come with pageantry:
https://www.youtube.com/watch?v=ponTbDDMYjw
And some music is to dance and groove:
https://www.youtube.com/watch?v=DPOeGWl6x4E
And some people feel the music:
https://www.youtube.com/watch?v=DPOeGWl6x4E
https://www.youtube.com/watch?v=Ihh_H5wHK-Q
So one thing music definitely does for many people is inspire them to move
I’d also add that I think Byrne needs to distinguish between listening to an album on your own versus going to a public concert. From my own experience, I’m much more open to novelty and surprise when listening independently. Heck, that’s how I find the songs from an artist that I most connect with. Going to a concert, however, is much more of a social experience. There’s an amazing feeling when everyone in the crowd is vibing, dancing, and singing with a song live. Playing a brand new song live can be a little jarring because the crowd is trying to get a feel or intuition for the new material.
emotional appeal vs. intellectual appeal
from infant times, repetition/structure gives humans emotional security, and that is mediated via sound (probably even in the womb) and food
intellectual appeal requires novelty in a setting of emotional stability
hence we like new books that follow a formula we enjoy
May be surprising – Karl Friston;
“The mathematics of mind-time
“The special trick of consciousness is being able to project action and time into a range of possible futures
by Karl Friston + BIO
“I have a confession. As a physicist and psychiatrist, I find it difficult to engage with conversations about consciousness. My biggest gripe is that the philosophers and cognitive scientists who tend to pose the questions often assume that the mind is a thing, whose existence can be identified by the attributes it has or the purposes it fulfils.
“But in physics, it’s dangerous to assume that things ‘exist’ in any conventional sense.”…
…
” To recap: we’ve seen that complex systems, including us, exist insofar as our Lyapunov function accurately describes our own processes. Furthermore, we know all our processes, all our thoughts and behaviours – if we exist – must be decreasing the output from our Lyapunov function, pushing us to more and more probable states. So what would this look like, in practice? The trick here is to understand the nature of the Lyapunov function. If we understand this function, then we know what drives us.
It turns out that the Lyapunov function has two revealing interpretations. The first comes from information theory, which says that the Lyapunov function is surprise – that is, the improbability of being in a particular state. The second comes from statistics, which says that the Lyapunov function is (negative) evidence – that is, marginal likelihood, or the probability that a given explanation or model accounting for that state is correct. Put simply, this means that if we exist, we must be increasing our model evidence or self-evidencing in virtue of minimising surprise. Equipped with these interpretations, we can now endow existential dynamics with a purpose and teleology.”
https://aeon.co/essays/consciousness-is-not-a-thing-but-a-process-of-inference
Some of the Aspects or re-listening are listed here: https://journals.sagepub.com/doi/full/10.1177/0305735617751050
My guess is that these components of the music linstening experience shape the differences in revisiting music vs. literature:
– The earworm phenomenon and wanting to scratch the itch, that an earworm never resolves
– The Emotional response to music and unique coupling with memory
– Embodiment of music –> needing a stable and predictable structure to dance to music and/ or follow its structure
.. and a few more reasons, which I can add later, but now I have to study for a math exam :(
There is a tone of research on this topic from the small community of music (neuro-)psychology research on the first two topics and the third is getting more and more explored. But i dont know of a study, that concentrated on this exact comparison.