TV Party Tonight!

| Gabriel |

A month or so ago bloggingheads had Alyssa Rosenberg and Peter Suderman (mp3 only), my two favorite politically-informed-but-not-hacks culture bloggers. In the course of their conversation they talked about “recapping” culture, which is where a blogger reacts in about 1000 words to each episode of a tv show, usually the day after it airs. I’m sure there were earlier precedents on Usenet forums, but I associate the development of this genre of criticism with Television Without Pity. TWOP recaps are almost Talmudic exegesis that take as long to read as the show itself takes to watch. There are currently many other recaps sites, most notably The Onion’s tv club, and other bloggers do just one or two shows, as Alyssa is currently doing with Breaking Bad and True Blood. It’s a very interesting genre of writing and helps illuminate some theoretical issues with the superstar effect and the demand structure for entertainment.

The superstar effect is of course Sherwin Rosen’s observation that cultural products and cultural workers have a truly ridiculous level of inequality. Rosen first noted that a scope condition is technology for infinite reproducibility and this has held up. However his theoretical mechanism was ordinal selection that was hyper-sensitive to infinitesimmal quality differences and later research has pretty definitively discarded that mechanism. Rather, most everybody now agrees that the superstar effect reflects some kind of cumulative advantage mechanism and the only question is exactly how it works. We know for a fact from Salganik’s music lab work that information cascades are a part of this, but that doesn’t mean that there aren’t also other cumulative advantage mechanisms at work.

Probably the first article to propose a cumulative advantage mechanism for the superstar effect was Moshe Adler’s “Stardom and Talent.” Adler is often cited as synthesizing network externalities and the superstar effect, that is, people read him as articulating a model of “watercooler entertainment” where entertainment is a mixed coordination game (aka, “battle of the sexes“) consumed mostly or entirely for its utility in providing topics of conversation. When you see people citing Adler they are usually arguing that cultural consumption is a means to an end of socializing. For example, imagine that (like any sane human being) you find watching golf on tv to be incredibly tedious but you force yourself to watch it so that you have something to chit chat about with your boss, who is a big golfer.

This is a compelling model, but it’s not actually the model Adler proposed, in part because he’s coming from a theoretical background that emphasizes demand (i.e., micro-economics) rather than a tradition that emphasizes homophily (i.e., sociology). What Adler actually wrote is that chit chat is a means to cultivating taste in entertainment addiction goods. Adler starts from the premise that many art forms function as addiction goods (aka, acquired tastes). However it is often difficult to consume enough of the art to get you into a place where the addiction good has positive expected value and so we use discourse about the art in order to heighten the addiction and thereby increase the utility of arts consumption. That is, I discuss a tv show with you because it helps me develop my relationship with the tv show, not because it helps me develop my relationship with you. We can see this in a formal setting when people take “[wine / opera / painting] appreciation” classes, where (in price theory terms) the class increases your addiction to the good even more so than simply consuming the good.

Adler’s model seems a bit on the aspy side and, like I said, people often get it backwards when they cite it, perhaps because they are forgetting how weird it is and one’s memory’s reconstructs the article’s argument to be more intuitive. Nonetheless, I think that Adler’s original model is also pretty compelling. Notably, there’s no reason why the causation has to go one way. It could be endogenous or it might even be contingent, with “watercooler” for some types of art and “addiction good” for others.

These are subtly different models and provide theoretical implications that are in theory distinguishable (though may be hard to disentangle in practice). In particular, I’m thinking that we can use Omar Lizardo’s argument about the different types of network ties supported by high culture versus pop culture. Omar argues that since pop culture forms a more universal social lubricant it should be (and in fact is) associated more with weak ties whereas high culture is tricky enough that it requires more strong ties.

If we extrapolate this out, we can interpret it as meaning that the “watercooler” network externality effect (ie, the common misreading of Adler) is a mechanism that supports cumulative advantage for shows that are very accessible and not terribly nuanced. That is, you might watch American Idol in order to have a bunch of 2 minute long conversations with acquaintances and strangers whom you normally come into contact with anyway. An important corollary is that you wouldn’t normally seek out fellow fans of crap but just make sure that you’re sufficiently familiar with crap to hold your own in a conversation with random people.

In contrast, we can use the “addiction goods” model (ie, Adler’s actual argument) to explain consumption of less accessible cultural objects of the sort that might sustain an entire dinner’s worth of conversation. The objects might even be so inscrutable that they are difficult to consume without having an interlocutor to help you make sense of them and so you might either seek out strangers who already consume the object or try to convince a close friend to consume the object as well so you can discuss it together. For instance if you read the first paragraph of this post and said “I don’t know or care about this Alyssa person but I’m going to click the link because I’m hoping somebody can help me understand what’s the deal with Hank’s mineral collection” then that would be an illustration of the addiction good model at work. Now if it’s just people who already consume a show finding each other that’s not cumulative advantage but homophily. However there is cumulative advantage if you start watching a show because your favorite blogger is recapping it or if you read a book to participate in a book club or if you buy your best friend a box set of the first season of Battlestar Galactica so you have someone with whom to discuss the downward spiral of Gaius Baltar. In this sense recapping is a complement to the increasing narrative complexity of popular entertainment and one way to see this is that people tend to recap shows with a serial rather than episodic structure.