The Internet is an Entertainment Machine

Sunday, January 12, 2025

Charlie Warzel and Mike Caulfield argue that the internet is a justification machine. Dan Williams responds that justifications have always been readily available, and thinks that Warzel and Caulfield romanticize past media environments, for their own self-justificatory purposes.¹ I think Williams is mostly right, but that both pieces are missing an important component of how changing media environments change how people engage with politics.²

That component is choice: how many entertainment options people have readily accessible to them. I don’t mean people ignore choice when they write about the internet. Choice is always part of how people address the effects of the internet. But I do think how people talk about choice in popular media tends to miss the most fundamental ways in which choice structures our politics.

Sometimes, you’ll see choice discussed in terms of “filter bubbles” — the idea that people self-select into media that affirm their opinions, or that might gradually change their attitudes to become more extreme.³ Sometimes it’s “echo chambers,” or worries about mis- or disinformation. Warzel and Caulfield are worried about a similar but arguably distinct problem: “misinformation is powerful, not because it changes minds, but because it allows people to maintain their beliefs in light of growing evidence to the contrary.”⁴ In their understanding, the danger of the internet is not necessarily that it changes your mind but that it gives you resources for resisting evidence that might otherwise lead you to change your beliefs.

Williams responds that motivated reasoning isn’t new, and is something everyone engages in — and always have when consuming media. He’s right. He also suggests that liberals are particularly attracted to explanations that conservatives are dupes, and if they only had the correct information they’d be good liberals. I think he’s right about that too.

Nonetheless, I think Williams’ piece suffers from a similar bias, one shared with the Warzel and Caulfield piece, a bias that is potentially consequential. Most people who write about politics — in the broadest sense — are knowledgeable about politics, care about politics, and think politics are consequential. Because of this, they wind up mostly being interested in explaining partisan behavior — the behavior of other people who are knowledgeable about politics, care about politics, and think politics are consequential.⁵

But many people aren’t knowledgeable about politics, don’t care about politics, and don’t think politics are that consequential. Many people do not have coherent “beliefs” for the internet to provide justification for. What are they up to, and what effect might the internet have on them? I want to prime some intuitions by looking at what should be a simpler example: the transition from broadcast TV to cable.

Low Choice vs. More Choice

It’s 6 pm on a Monday night in 1966, and you want to unwind after work so you turn on the TV. If you care about politics, you have some options: you can watch Peter Jennings with the News on ABC, or local news on NBC or CBS — their news programs, The CBS News and The Hunter-Brinkley Report, come on at 6:30. What about if you don’t care about politics? You have the same options. If you want to watch TV, you’re watching the news.

The reverse is also true. Let’s say you want to watch the news, but you work late. You don’t get to watch it. No TIVO, no internet, no time-shifting in 1966.

By 1970, the average household watched six hours of TV a day.⁶ What does this mean? It means that people who otherwise would probably not watch the news, instead did watch the news.⁷ The broadcast networks captured 80% of all television viewing during this period, and three-quarters of households watched one of the network news programs.⁸

In another piece, Williams writes that “what many think of as a mid-twentieth century “golden age of an objective press” was in fact a highly insular, elitist media ecosystem that disproportionately represented the interests and voices of a very narrow segment of society.” I want to be clear that my claim is simply that during the period during which broadcast TV was dominant, more people consumed more of the same content. That may well have been a bad thing. Nonetheless, it’s worth trying to tease apart the effect of the structure of that media environment from the content of that media environment.

The years in which broadcast TV was omnipresent and dominant represented a period in which there was relatively little media choice.⁹ The first thing that changed that wasn’t the advent of the internet. It was the advent of cable.

Not everything about the Internet is new. Before Internet users could access vast amounts of political information with the click of a mouse, cable subscribers could flip between different 24-hour news channels with the push of a remote. Before Internet users could download music or order movie tickets online, cable viewers could pick between different music channels and hundreds of movies and entertainment shows everyday. In short, long before the Internet offered an alternative to mainstream media, cable television satisfied many niche interests.¹⁰

So one question we might ask is: what did cable do? Markus Prior has argued that more choice led to two main consequences:

1. More choice means that individual preferences become more important in determining who consumes news content and who does not. That is, people who were not interested in news consumed less news media, and people who were interested in news consumed more. Suddenly if you wanted to watch TV at 6 pm you didn’t need Peter Jennings on. You could turn on TBS and watch MGM or UA movies. If you did want more news, starting in 1980 you could turn on CNN and get all the news you wanted.
2. These differences in media consumption can partially explain political polarization, without any underlying change in political attitudes. That is because it changes the composition of the electorate. As people who are less interested in politics consume less news content, they become less involved in politics. As people who are more interested in politics consume more news content, they feed their interest and become more involved in politics.

Media choice, on this understanding, can drive changes in the composition of the electorate that might increase polarization without changing any underlying political attitudes.¹¹ And that’s what Prior argues: that cable helps drive a decline in turnout among the least informed but most volatile voters, who might otherwise have had a moderating effect on election outcomes.

News is Entertainment

One thing that should be clear from the above is that the news is an entertainment product, and just one among many. People who consume a lot of news do so because they like it. People who don’t, don’t, because they’d rather spend their time doing something else — playing fantasy football, the latest Call of Duty, watching reality TV, or following Beyonce fan accounts.

That the news is an entertainment product becomes more clear if you ask what most people use the news for: nothing. Usually the news is not actionable. People follow the news because they are interested in it. Even the action most available to people on the basis of the news — how they vote in an upcoming election — should not be that affected by any individual news story. After all, which party you vote for is likely the product of a lifetime of socialization, consideration of values, and many policy positions. It doesn’t make much sense for most people to change their mind on the basis of an individual news story.

The people who are exceptions to this are mostly committed partisans. They follow news events doggedly, vote in primaries, and mobilize as activists. They already have their minds made up!

Our High Choice Environment

Media studies is a challenging field. People are unreliable narrators of their own media consumption, and immensely inflate, as an example, how much they pay attention to the news. This makes it difficult to just ask people what they read, watch, and listen to, which makes it hard to study.

I think if we were naively extrapolating from the shift from broadcast TV to cable news, we might expect that the internet has ushered in an even higher choice media environment, and that people are self-sorting even more efficiently. Not necessarily into partisan filter bubbles, but away from politics at all.

Politics, after all, is often depressing. Little political information is actionable, and people have their own problems. It shouldn’t be surprising, then, when studies find that most people consume relatively little political information.

We find that citizens prefer online entertainment, shopping sites, and celebrity gossip over news and public affairs. That may be nothing new (Prior, 2005). Yet, our data also show that citizens’ visits to non-news sites are the dominant source of political information. Even though politics in these kinds of sites comprised only 1.6% of all visits, the aggregate popularity of webmail, entertainment, shopping sites, or celebrity gossip means that an average citizen encounters most political content outside news. People, especially Americans and especially those with low political interest, encounter politics more frequently outside news outlets than within.

There has long been a tension in political science between naive theories of democracy that seem to rely on an informed populace and democracy as it’s actually practiced. The focus of Warzel and Caulfield’s piece is the January 6th, 2021 riot at the capital. They marvel that despite the “mountain” of evidence about an event that “played out on social media” in real time, that many republican voters have nevertheless been able to ignore it.

This doesn’t seem so odd to me. January 6th is one event, and the people they are talking about are committed partisans, most of whom are presumably committed Republicans for many reasons.¹²

Partisans behaving as partisans may be regrettable, insofar as partisanship can distort good judgment, but the bigger question to me is what the internet does to the many more people who are not that partisan, and to the media institutions we rely on.

Insofar as a high choice environment incentivizes pursuing niche audiences, one might worry about the future of news gathering organizations writ large. News gathering in old media institutions was supported by cross-subsidies that came from bundling the news with other products, like sports, arts, comics, classifieds, etc. The internet has, of course, led to the unbundling of all that, and with it the withering away of the local and regional “news bundle” — the many newspapers and media organizations that used to serve particular areas of the country, that have gradually gone out of business or been acquired as consumer attention has moved online. Now, news consumers consume news nationally, from a few big winners like the New York Times, and ideological outlets that serve their partisan needs, and everyone else finds their entertainment in products that appeal more directly to their interests.

To gesture at a broader argument, I think this is one part of a story where polarization is real and has been largely driven by party activists and political elites, as a product of the democratization of the primary process in the 1970s and increasing media choice that comes almost contemporaneously. The most politically committed activists become more and more engaged, driving partisan primaries that produce more ideological and more ideologically disciplined candidates. Most voters take their policy positions from elite cues, and those policy positions are driven by intra-party interest group competition. And all of the above can happen without very much attitude change at all on the part of the mass electorate.

#Media #Politics #Social Science #Political Behavior #Polarization