Is Justin Timberlake a Product of Cumulative Advantage?

April 21, 2007 4:59 PM

In a parallel universe Your Favorite Band Really Does Suck! Duncan Watts and others conducted a Web-based experiment [PDF] called Music Lab. Their findings: "while talent might distinguish good from bad, social pressure and pure dumb luck are also big influences on which bands gain the most fame." "Calling the [experiment] 'pathbreaking,' sociologist Michael Macy of Cornell University says the findings illustrate how a small advantage can snowball, making popularity hard to predict. Economist Robert Frank, also at Cornell, says the work shows 'we're all susceptible to the herd mentality.'" The effect of "cumulative advantage" has impact on the popularity of other aspects of contemporary culture: books, films, websites and more.
posted by ericb (42 comments total)

yeah it's a strange thing. I suspect we have all experienced seeing performers who are both more talented (in a measurable way), and more personable/charismatic then a random top 50 artist, yet they are stuck doing the bar circuit. We have a local music festival here coming up in a few weeks who's performers (in my opinion) would kick 5 kinds of ass against just about any mainstream music star.
posted by edgeways at 5:13 PM on April 21, 2007

"pure dumb luck are also big influences on which bands gain the most fame."

Duh.

"we're all susceptible to the herd mentality"

Double duh.
posted by flapjax at midnite at 5:18 PM on April 21, 2007

Amazing waste of research apparatus to generate meaningless quantitative results to characterize a qualitative phenomenon.
posted by spitbull at 5:19 PM on April 21, 2007 [3 favorites]

What our results suggest, however, is that because what people like depends on what they think other people like, what the market “wants” at any point in time can depend very sensitively on its own history: there is no sense in which it simply “reveals” what people wanted all along. In such a world, in fact, the question “Why did X succeed?” may not have any better answer than the one given by the publisher of Lynne Truss’s surprise best seller, “Eats, Shoots & Leaves,” who, when asked to explain its success, replied that “it sold well because lots of people bought it.”

This is not a "duh" statement, but contradicts conventional wisdom. I don't get the bitching.
posted by raysmj at 5:37 PM on April 21, 2007

"Pathbreaking" or "Duh."

I think "duh" wins on this one.
posted by kozad at 5:37 PM on April 21, 2007

When I read this study, I clapped my hands and said "Yeah."
posted by adipocere at 5:46 PM on April 21, 2007 [4 favorites]

I'm not reading this post until enough other people do.
posted by The Deej at 5:51 PM on April 21, 2007 [3 favorites]

This explains The Doors.
posted by scheptech at 5:53 PM on April 21, 2007

Yeah it's pretty obvious, but the fact that it's obvious doesn't mean it's not a valid kind of research. Most people would say it was "obvious" if the other result were found to be true as well.

It can be usefully to actually test commonly held beliefs to make sure that they're right. This is also something that people disagree about A lot of people think that the world really is a meritocracy, rather then mostly a randomocracy (which is what I think it is).

That said, the fact that so many people are musically talented is great for my political position, which is the utter annihilation of the RIAA!
posted by delmoi at 5:54 PM on April 21, 2007

In the words of the late great Howard Cosell, these researchers have "an uncanny grasp of the obvious". Now, wake me if they ever find a practical way to harness the 'dumb luck' factor.
posted by wendell at 5:54 PM on April 21, 2007 [1 favorite]

Uhhhh. Marketing & design & stuff have a hell of a lot to do with it. I mean, do you really think that one dishwashing detergent is more successful than another because it's actually BETTER? Please. People are so easily brainwashed into thinking they want stuff they don't. [By people like me. Sigh.]

What bugs me is that if Aretha Franklin or Ella Fitzgerald were starting out right now they would not have the careers they did back when people were judged mostly by how they sounded on the radio & how much musical talent they held. Now you can push a button in studio and snap someone's voice to be on pitch. As a jazz singer I have to be able to improvise something live and hit everything in one take... but that's really not expected of an awful lot of musicians now. Sometimes for one song they will do 20 takes and splice it all together... and so music ends up being more a creation of the engineers & ProTools than anything.

Damn lazy off pitch kids get off my musical lawn. ;)
posted by miss lynnster at 5:56 PM on April 21, 2007 [1 favorite]

P.G.O. !!!!

This is not a "duh" statement, but contradicts conventional wisdom.

The wisdom that people buy a book because it's good? That's not the only conventional "wisdom". There's also the wisdom that people buy a book because it's a bestseller. You really didn't know that one?
posted by Rich Smorgasbord at 5:56 PM on April 21, 2007

I see lots of confusion in the posts here.

Is the article a 'duh'? We all already know:
1. 'They' can't tell shit from shinola, and swallow any old crap that's currently popular, but
2. 'I' make rational decisions decisions by using my superior discernment to evaluate truly.

Yeah. Sure. Obvious. Not worth examining. Next you'll be trying to tell me I'm addicted to cigarettes, when I know I just like the taste and it's fun to do on breaks and it makes me look cooler.
posted by hexatron at 6:49 PM on April 21, 2007

This is totally pathbreaking, except for the fact that Pierre Bourdieu did this all back in the '70s and wrote several books on this theory.

Lazy academics, they be.
posted by sadmarvin at 6:49 PM on April 21, 2007 [1 favorite]

Thankfully there are social forces that counteract flocking behaviour as well. They come, for example, in the shape of the outspoken, the eccentric and the plain snobbish individuals among us who react against the mob, declare that the emperor has no clothes, and reintroduce some independence into the mix. [3rd link]

Yeah, this is why I didn't go see Nirvana in high school. I was like "anybody can play three chords"

[kicks self once again]
posted by pwedza at 6:53 PM on April 21, 2007

Nervermind "shit from shinola", that was more like picking between the turd with the peanuts or the corn in it. Can't wait to hear it again... in an ad campaign. Except the "rock meet reggae" geniuses. That shit's gonna be huge. Literally.
posted by tighttrousers at 7:27 PM on April 21, 2007

What, truth is political? D'oh!!!
posted by Rich Smorgasbord at 7:30 PM on April 21, 2007

Wow. I just gave up on the musiclab experiment. Holy garage-band wasteland.

I rated two of the songs at 4, none at 5, and the bulk at 2. I downloaded the 2 "4 star" songs, but they're unlikely to make my regularly rotation.

It's all birkenstock + sweater + acoustic + breathlessly-emotional OR cookie-cutter pop punk. Some "metal" influenced stuff. Nothing else really. No jazz, no rnb, no hip-hop, no electronica, no industrial, no classical, no country, no western, no no no no...

That being said...

It can be usefully to actually test commonly held beliefs to make sure that they're right. This is also something that people disagree about A lot of people think that the world really is a meritocracy, rather then mostly a randomocracy (which is what I think it is). -- delmoi

Wonderful comment.
posted by C.Batt at 7:46 PM on April 21, 2007

Hi. Gin Blossoms.
posted by trip and a half at 8:00 PM on April 21, 2007

Musiclab: Now I know what happened to all those bands that used to play the Bronze on Buffy the Vampire Slayer.

The findings strike me as a little on the "duh" side myself, but are certainly vindicating for anyone involved in the arts who toils in obscurity while the CHUDs who made "My Humps" roll around in huge, cokestained haystacks of money.
posted by kittens for breakfast at 8:13 PM on April 21, 2007 [1 favorite]

Your favourite band still sucks. I didn't need a Ph.D for that.
posted by Chocomog at 8:44 PM on April 21, 2007

Your favourite band still sucks. I didn't need a Ph.D for that.


But I suppose if you had one you might've been able to get some research money to say it.
posted by flapjax at midnite at 8:56 PM on April 21, 2007

This explains The Doers.
posted by anthill at 9:40 PM on April 21, 2007

Note to Galileo: Duh, its obvious that the heavier ball is going to fall faster, why go through all the trouble of climbing the leaning tower of Pisa? (I bet its still 'obvious' to many people that the heavier ball will fall faster, its so intuitive, right?)

I think that this was a pretty cool study for no other reason than the way in which the internet was used to create the alternate universes. It probably would have been very difficult to carry out a study on this scale previously.

One of the reasons why I like the recommendation system at Pandora, which is based on the intrinsic properties of the songs rather than upon "people who like this song also like song X", is that this system does more to reduce the bandwagon effect.
posted by mach at 11:17 PM on April 21, 2007

There's also the wisdom that people buy a book because it's a bestseller. You really didn't know that one?

But how does it become a bestseller in the first place? I also don't think this was what Bourdeiu wrote about, exactly--there's not much about class and taste here. I think I can can see Bourdeiu in everything from opera to ranking of barbecue places in the South and neighborhood gentrification via hipsters. But this is more about popularity across classes. I guess you could say social elites are involved in a gatekeeping role, but that's not addressed here.
posted by raysmj at 12:44 AM on April 22, 2007

You could probably apply Bourdieu more easily to Pandora.
posted by raysmj at 12:45 AM on April 22, 2007

hexatron
Is the article a 'duh'? We all already know:
1. 'They' can't tell shit from shinola, and swallow any old crap that's currently popular, but
2. 'I' make rational decisions decisions by using my superior discernment to evaluate truly.

Yeah. Sure. Obvious. Not worth examining. Next you'll be trying to tell me I'm addicted to cigarettes, when I know I just like the taste and it's fun to do on breaks and it makes me look cooler.
I think you're being a little unfair, and you're misrepresenting the findings. The study shows that there is a strong social influence on individual opinions or choices, and that this cumulative social advantage stems from small, random differing initial conditions. The key paragraph of the NYT article is this:

"Because the long-run success of a song depends so sensitively on the decisions of a few early-arriving individuals, whose choices are subsequently amplified and eventually locked in by the cumulative-advantage process, and because the particular individuals who play this important role are chosen randomly and may make different decisions from one moment to the next, the resulting unpredictability is inherent to the nature of the market."

There will always be individuals who buck the trends, stake out their own choices, and thus set in motion new snowball chains which will eventually, given the right conditions, change the crowd. I think the reason many people here are saying "duh" is because the people who visit MetaFilter are a fairly biased set, in that they are probably the types who already look to consciously buck prevailing musical trends (to varying degrees...). But there will be the Ryan Schreibers who will start their own Pitchfork Medias that will grow to have a strong hold on popular taste.

This should actually be heartening, because it dispells the idea of cultural control by the media companies, because those initial trendsetters are essentially random. Corporations have a huge advantage in terms of market reach, but there's still the hope that some guy somewhere will be the one to turn things around.

I agree with miss lynnster, though, that it would be interesting to rerun this test but add in marketing elements. For instance, perhaps before the people voted next time, some would be exposed to ads for Band X, some for Band Y, etc, and some for no bands, and see what that does.
posted by Sangermaine at 1:46 AM on April 22, 2007

Good point about Galileo, you gotta test everything at least once. The tone of "surprise, surprise!" is a hoot, though.
posted by Rich Smorgasbord at 1:47 AM on April 22, 2007

the impact of a listener’s own reactions is easily overwhelmed by his or her reactions to others

This is why successful music marketing should rely less on graphic image and more on the type of viral marketing the fashion industry uses to great success: getting the people perceived by their friends as hippest and best-connected to tout the product. If music marketing hasn't been successful, it's because music has been treated as something that's not a commodity like other products. But commercial music is a commodity; that's why the record companies are in business. They've just never known how to sell things.

Commercial success is no predictor how 'good' music is, barring the major technical and talent problems that brand a song subpar. There are a lot of talented bands and artists who are famous, and many also who are not. There's no problem at all until we conflate 'good' with 'famous.' They're two different things.
posted by Miko at 6:05 AM on April 22, 2007

I think that Bourdieu is still highly relevant, it's just that the conception of class in the modern world (especially North America) is different than it was in 1970s France. In France at the time, class was decided by money and culture; now adays, money can be (relatively) easily gained or lost and culture (in the form of "cool") has become more important. Bands consecrated as cool by the culture makers slowly make their way into the mainstream as members of the mainstream try to tailor their tastes to the 'higher class' culture makers in an attempt to appropriate their cultural capital. It's all in Bourdieu, just in slightly different terms, as one would expect after a 30-year gap in time.

The idea of "cool" as a class, however, was just explored a few years ago in Joseph Heath and Andrew Potter's book The Rebel Sell (Unfortunately namedNation of Rebels in the US). This book makes a great case for cool being the driving force in a new type of class struggle, in which Bourdieu's theories transfer neatly from the old conception of class to this new conception of cool.

Ultimately, I just take offence to the assertion that this work is "pathbreaking," especially when that assertion comes from a sociologist who should really know better. This study is indeed a valuable one, as it reinforces Bourdieu's theories, but to claim that it's new research and to try to apply new terms like "cumulative advantage" where we already have terms such as "cultural capital" and "social capital" seems like academic laziness and self-glorification. The cynical part of me even goes as far as to think that these guys just want to write the next Freakonomics and, to do so, have conveniently chosen to ignore Bourdieu and Heath & Potter.
posted by sadmarvin at 6:40 AM on April 22, 2007

it would be interesting to rerun this test but add in marketing elements.

This is why successful music marketing should rely less on graphic image and more on the type of viral marketing the fashion industry uses to great success: getting the people perceived by their friends as hippest and best-connected to tout the product.

Check out this old thread resurrected yesterday in MetaTalk in which a sub-par Australian musician, bankrolled by his wealthy father, attempted to use viral marketing to launch a successful musical career, but failed miserably.
posted by ericb at 7:48 AM on April 22, 2007

In France at the time, class was decided by money and culture; now adays, money can be (relatively) easily gained or lost and culture (in the form of "cool") has become more important. Bands consecrated as cool by the culture makers

Do you really think it works this way--that kids really dig rock critics and go for everything marketers push on them (the point of this, I thought, was that they very well might not)? And what is the background of these gatekeepers (who I'd agree do have some influence)? I highly doubt they all grew up poor, and that they didn't at least take college classes, even if they didn't graduate. So why do you discount class as a variable?
posted by raysmj at 9:48 AM on April 22, 2007

I wouldn't say that class should be discounted, just that it's grown less and less static over time. Major cultural gatekeepers like John Peel and Pitchfork show that class isn't always the primary factor. Sure, Paris Hilton's class will prove a strong motivator in someone's love of her 'music,' but this isn't the rule. In fact, her obvious exploitation of her class turns many people off, rather than encouraging them to buy her music, perfume, whatever she's selling right now.

If anything, class has become highly complicated, with money, education, and other variables all factoring into a general class ranking. Even then, these class markers are quite subjective, resulting in different groups reacting in totally different ways to the same class markers. Working in a record shop is a desirable occupation for a hipster (who, presumably, is primarily concerned with 'cool'), but the mere fact that you'd have to work--especially so vulgar an occupation--is a black mark against you in a rich, old-fashioned crowd.
posted by sadmarvin at 10:27 AM on April 22, 2007

sadmarvin writes 'Sure, Paris Hilton's class will prove a strong motivator in someone's love of her "music," but this isn't the rule. In fact, her obvious exploitation of her class turns many people off, rather than encouraging them to buy her music, perfume, whatever she's selling right now.'

I don't think her class has anything to do with it, sadmarvin. I think it's got much more to do with her appearance in a certain video.
posted by PeterMcDermott at 11:02 AM on April 22, 2007

My favorite band has sucked for 30 years.
posted by davy at 11:03 AM on April 22, 2007

I don't think her class has anything to do with it, sadmarvin. I think it's got much more to do with her appearance in a certain video.

Then why don't all pornstars have their own commercial empires? The video surely threw her into the pop-culture conscious, but if she hadn't been a Hilton (or some similarly well-thought of family/group) her video likely wouldn't have been noticed. I'm not saying that no one else would achieve such fame from a similar video, just that her class is an undeniable advantage in this instance.

Of course, other pornstars have done similar things, but this just underscores the complexity of the whole situation--one of the things that this recent study does well. I think that the emphasis on the unpredictability of exactly which songs became popular within the context of the trials is a fantastic point, but the researchers also downplay the whole body of research to which the study belongs.

As an aside, this almost makes me wonder if sociologists should take a hint from the quantum physicists and start talking about 'probability clouds' of popularity and success.
posted by sadmarvin at 11:14 AM on April 22, 2007

How greatly does Pitchfork affect mass popularity? It would seem to have more of an impact on subcultural popularity (where it seems as if obscurity is more highly valued--and now Pitchfork is getting into Rolling Stone territory with best-of list of oldies and whatnot, the fact that they make different choices than RS usually does being beside the point). I wouldn't compare its effect to that of BBC radio in Peel's heyday. For one thing, Pitchfork's still gaga over albums, which are not selling quite as well these days--the single is back among the great unwashed. Yes, this is a different sort of class, but it's still a way of setting groups apart and would appear to involve socioeconomic status to some degree (Why else does money seem to follow them in the gentrification dept., say?).
posted by raysmj at 12:43 PM on April 22, 2007

a sub-par Australian musician

It's not fair to say that the marketing didn't work by using a subpar musician. This experiment showed that you could control to identify music that was par and subpar, and didn't claim that marketing would help a bad (subpar) band, just that the approval of others would give an otherwise unavailable boost to a band that was at or above par.

How greatly does Pitchfork affect mass popularity? I'd say greatly. It's not that kids are listening to critics (though older people are); it's that the critics, people who have media outlets that let them listen to and judge new music before it's widespread, are the ones whose voices generate the impact for the band. It's alternative newsweeklies in local markets, then the next closest ears to the ground at Paste, Pitchfork, CMJ, All Songs Considered, and other early-impact reviewers who listen to nearly everything can get a band's work to the point where a music programmer on a college or independent station will hear of them. Without having that first level of influential listenership to either drop or boost a single artist, those artists wouldn't achieve the first buzz that gives them some national exposure. It's not fans that make artists big - culture brokers have to let fans know the artists exist first, using chains of influence to get them heard on radio, on Pandora, on sample CDs, etc.

That cumulative advantage does exist, and we do use the early judgements of others as a guide to what we should give our limited listening time to. Whether their judgements are a good guide is in question - it certainly depends on your tastes. And it's undeniably true that plenty of very good bands don't get the boost they'd need at that crucial moment when the audience is ready, but the ears in a position of judgement aren't aware of it.
posted by Miko at 1:33 PM on April 22, 2007

delmoi: "
That said, the fact that so many people are musically talented is great for my political position, which is the utter annihilation of the RIAA!"

Delmoi for president.
posted by Mr. Gunn at 5:24 PM on April 22, 2007

Didn't Malcolm Gladwell write a book about this?
posted by Mr. Gunn at 5:27 PM on April 22, 2007

He did touch on it, Mr. Gunn, and I've read similar studies regarding the way in which VHS beat out the superior BetaMax format, etc. But to me, the valuable part of this study is that it was a truly controlled experiment, not simply observations, anecdotes, or guesses.
posted by Miko at 7:42 PM on April 22, 2007

Hi. Gin Blossoms.

Hi. Radiohead.


(and your favorite research sucks.)
posted by ZenMasterThis at 7:46 PM on April 22, 2007