Longstanding concerns about the supposed ubiquity of online filter bubbles are not supported

There is an obsession among the punditry and academia with the need for censorship of speech.  Part of this is justified on the basis that speech is dangerous ("hate speech") and part of this is justified on the basis of the need to minimize misinformation or disinformation.

Both justifications are facially absurd and incompatible with the natural right of free speech (and the attendant freedoms of religion, assembly, etc.).  

There are four broad platforms of speech, all intertwined - 1) Print media, 2) TV media, 3) Radio, and 4) Internet and Social Media.  The authoritarian initiatives to control speech are often focused on the fourth category of the Internet and Social Media.  

Which makes this recent research interesting.  From Quantifying partisan news diets in Web and TV audiences by Daniel Muise, Homa Hosseinmardi, Baird Howland, Markus Mobius, David Rothschild, and Duncan J. Watts.  From the Abstract:

Partisan segregation within the news audience buffers many Americans from countervailing political views, posing a risk to democracy. Empirical studies of the online media ecosystem suggest that only a small minority of Americans, driven by a mix of demand and algorithms, are siloed according to their political ideology. However, such research omits the comparatively larger television audience and often ignores temporal dynamics underlying news consumption. By analyzing billions of browsing and viewing events between 2016 and 2019, with a novel framework for measuring partisan audiences, we first estimate that 17% of Americans are partisan-segregated through television versus roughly 4% online. Second, television news consumers are several times more likely to maintain their partisan news diets month-over-month. Third, TV viewers’ news diets are far more concentrated on preferred sources. Last, partisan news channels’ audiences are growing even as the TV news audience is shrinking. Our results suggest that television is the top driver of partisan audience segregation among Americans. 

Online partisan-segregation is only 4% online.  That could be a meaningful problem but at first blush does not appear so.  

And while partisanship-segregation is 17% on television, that is the media platform that is declining.

Click to enlarge.

The researchers properly discuss the various very real limitations to their investigation (it is a tangled info-ecosystem out there).  Interestingly:

These unresolved issues notwithstanding, we close by noting that any resulting inferential errors would have to be very large to alter our main conclusions regarding the relative state of partisan-segregated news consumption online and on TV. With respect to online, partisan-segregated news consumption affects at most a few percent of Americans (Fig. 1), only a few percent of those people remain partisan-segregated for more than a few months (Fig. 2), and even those few percent consume a nonnegligible fraction of nonpartisan news. Although highly concentrated consumption of highly partisan material can be a cause for concern even if it affects only a small number of people—if, say, it facilitates extremist or violent behavior— longstanding concerns about the supposed ubiquity of online filter bubbles are not supported. With respect to TV, the picture is more complicated. On the one hand, our results make clear that partisansegregated consumption is far more prevalent on TV than it is online, affecting as much as 17% of the population (Fig. 1). It is also considerably stickier and more concentrated: After 6 months, the fraction who remain partisan-segregated is several times larger than it is online (Fig. 2), and the inhabitants of the most partisan archetypes consume partisan content almost exclusively (Fig. 3). On the other hand, whether even these much larger numbers should be considered large depends very much on what they are being compared with. If one were under the impression that the entire country was living in echo chambers, for example, 17% might sound reassuringly small. Likewise, if one had assumed that an echo chamber, once inhabited, was a permanent state of affairs, then it might be reassuring to learn that three quarters of inhabitants had left after 6 months. However, viewed from another perspective—say the percentage of voters needed to sway an election—17% may seem like a very large number indeed. Between these differing interpretations, our own is that partisan segregation in TV audiences—whether it is large enough to be considered alarming—is large enough to justify TV news receiving at least the same level of scrutiny as its online counterpart.

Nothing in these results changes my default assumption. 

I have seen no data in recent years to support that there is a general danger or even issue with American media consumption and speech.  It appears to me that claims of potential danger are speculative and almost always associated with those with predicate intentions to regulate speech.  

Further, that the proliferation of platforms, channels, and mechanisms for exercising speech has been a boon for increased connectivity for all citizens with even the most remote or specialized interests.  While at the fringe there are always criminal elements who should be detected and prosecuted for that criminal activity, the volume, nature, veracity, and variety of American speech is now, probably at greatest flower since the era of broadsheets and penny newspapers.  

Until evidence emerges, it appears to me that the wish to control speech owing to a fear of Hate Speech or a fear of Misinformation is mere authoritarianism dressed up under the guise of Emotional Caring.