Tag Archives: Content moderation

Information and the public sphere

Freedom of speech and the US political crisis

Thom Dunn on Medium really hits it on the head in describing the current crisis surrounding the ejection of the President from major social media platforms. Many have been laboring under the illusion that social media dialogue is akin to public exchange in a town square; in fact, the correct operative analogy is to a private club, and this misunderstanding was decisively cleared up for those thus deluded when the bouncers at their own discretion kicked them out.

Indeed, rather than an assault on unfettered free speech, which was never really on offer in the first place, the move of the social media titans signals a realignment of business interests, which have decided to comprehensively ditch the MAGA movement. This development is wholly compatible with models of delegitimization crises, such as the classic one by Michel Dobry.

Information and the public sphere

Babies and bathwater

Just attended an EFF-run ‘Fireside Chat’ with US Senator Ron Wyden (D-OR) on Section 230. As one of the original drafters of the legislation, the Senator was eager to point out the core values it was meant to shield from legal challenge, permitting the full deployment of constitutionally-protected speech online without imposing an undue burden of liability on those hosting such speech.

The Electronic Frontier Foundation and other digital rights organizations find themselves in a complicated political position, for, having spoken out against the risks and abuses originating from Big Tech long before there was widespread public consciousness of any problem, they now have to push against a bipartisan current that has crystallized in opposition to Section 230. Even some generalist news outlets have seized on the matter, giving scant play to the values and legitimate interests the law was originally intended to safeguard.

It seems fairly clear that mainstream political discourse has been extremely superficial in considering key aspects of the problem: Section 230 has become a symbol rather than a mere tool of governance. It may also be the case that the wide bipartisan consensus on its ills is in fact illusory, simply being a placemarker for incompatible views on how to move beyond the status quo, with the most likely outcome being paralysis of any reform effort. However, the risk that the imperative to do something cause the passage of hasty measures with lasting damage is real.

In a way, the present situation is the poisoned fruit of a narrative that linked the unfettered expansion of the Big Tech giants over the past decade to the libertarian and countercultural ideals of the early internet: when the former came to be perceived as intolerable, the latter were seen at best as acceptable collateral damage. Most of the popular animus against Section 230 that politicians are attempting to channel stems from resentment at the power of social media platforms and digital gatekeepers. Therefore (and although there may well be a case for the need to curb mindless amplification of certain types of speech online), perhaps antitrust action (in Congress or in the courts) is more suitable for obtaining the results the public seeks. Comparative policymaking will also be extremely relevant, as the European Union pursues its own aggressive agenda on content moderation, permissible speech, and monopoly power.

Information and the public sphere

Disinformation at the weakest link

There was an interesting article recently in Quartz about 2020 electoral disinformation in Spanish-language social media. While the major platforms have taken credit for the fact that the election did not feature a repeat of the coordinated foreign influence operations of 2016, arguably the victory lap came somewhat too soon, as the problem cases in the information sphere this cycle are only gradually coming to light. Post-electoral myth-building about a rigged process and a stolen victory, for one, while of little practical import for the present, has the potential to plant a deep, lasting sense of grievance in conservative political culture in the US over the long term. Similarly, the fact that less public attention, less civil-society scrutiny, less community-based new-media literacy, and less algorithmic refinement were available for Spanish-speaking electoral discourse meant that disinformation was allowed to expand much more broadly and deeply than in English. The mainstream liberal narrative that such a fact per se helps explain the disappointing electoral results of the Democratic Party with this demographic (especially in States like Florida or Texas) is itself fairly insensitive and stereotyped. The Latinx electorate in the US is not a monolith, and segments within it have distinct political preferences, which are no more the product of disinformation than anyone else’s. Yet, it seems clear that in this electoral campaign factually false political statements received less pushback, both from above and from below, when they were uttered in Spanish.

Two general facts are illustrated by this example. On the one hand, because of the production and distribution dynamics of disinformation, it is clear that its spread follows a path of least resistance: minority languages, like peripheral countries or minor social media platforms, while unlikely to be on the cutting edge of new disinformation, tend to be more permeable to stock disinformation that has been rebutted elsewhere. On the other hand, where disinformation has the ability to do the most damage is in spaces where it is unexpected, in fields that are considered separate and independent, subject to different rules of engagement. In this sense, fake news does not simply provide partisans with ‘factual’ reasons to feel how they already felt about their adversaries: it can legitimately catch the unsuspecting unawares. One of the reasons for disinformation’s massive impact on American public discourse is that in a hyper-partisan era all manner of domains in everyday life once completely divorced from politics have been reached by political propaganda, and in these contexts a weary habituation with such wrangling has not yet set in, effectively tuning them out. This dynamic has been reflected in social media platforms: the ‘repurposing’ of LinkedIn and NextDoor in connection with the BLM protests is telling.

So, if disinformation at its most effective is the insertion of narratives where they are least expected, and if its spread follows a path of least resistance, seeking out the weakest link (while its containment follows an actuarial logic, the most effort being placed where the highest return is expected), what does this portend for the possibility of a unitary public sphere?

There is reason to believe that these are long-run concerns, and that the Presidential campaign may have been the easy part. As Ellen Goodman and Karen Kornbluh mention in their platform electoral performance round-up,

That there was clearly authoritative local information about voting and elections made the platforms’ task easier. It becomes harder in other areas of civic integrity where authority is more contested.

Foreign counterexamples such as that of Taiwan reinforce the conundrum: cohesive societies are capable of doing well against disinformation, but in already polarized ones a focus on such a fight is perceived as being a partisan stance itself.

Domestic politics and elections

Victory lap for the EIP

Today, I followed the webcast featuring the initial wrap-up of the Electoral Integrity Partnership (I have discussed their work before). All the principal institutions composing the partnership (Stanford, University of Washington, Graphika, and the Atlantic Council) were represented on the panel. It was, in many respects, a victory lap, given the consensus view that foreign disinformation played a marginal role in the election, compared to 2016, also due to proactive engagement on the part of the large internet companies facilitated by projects such as the EIP.

In describing the 2020 disinformation ecosystem, Alex Stamos (Stanford Internet Observatory) characterized it as mostly home-grown, non-covert, and reliant on major influencers, which in turn forced platforms to transform their content moderation activities into a sort of editorial policy (I have remarked on this trend before). Also, for all the focus on social media, TV was seen to play a very important role, especially for the purpose of building background narratives in the long run, day to day.

Camille François (Graphika) remarked on the importance of alternative platforms, and indeed the trend has been for an expansion of political discourse to all manner of fora previously insulated from it (on this, more soon).

Of the disinformation memes that made it into the mainstream conversation (Stamos mentioned the example of Sharpie-gate), certain characteristics stand out: they tended to appeal to rival technical authority, so as to project an expert-vs-expert dissonance; they were made sticky by official endorsement, which turned debunking into a partisan enterprise. However, their very predictability rendered the task of limiting their spread easier for the platforms. Kate Starbird (UW) summed it up: if the story in 2016 was coordinated inauthentic foreign action, in 2020 it was authentic, loosely-coordinated domestic action, and the major institutional players (at least on the tech infrastructure side) were ready to handle it.

It makes sense for the EIP to celebrate how the disinformation environment was stymied in 2020 (as Graham Brookie of the Atlantic Council put it, it was a win for defence in the ongoing arms race on foreign interference), but it is not hard to see how such an assessment masks a bigger crisis, which would have been evident had there been a different victor. Overall trust in the informational ecosystem has been dealt a further, massive blow by the election, and hyper-polarized post-truth politics are nowhere near over. Indeed, attempts currently underway to fashion a Dolchstoßlegende are liable to have very significant systemic effects going forward. The very narrow focus on disinformation the EIP pursued may have paid off for now, but it is the larger picture of entrenched public distrust that guarantees that these problems will persist into the future.

Organizational behavior and reputation

Censorship about censorship

In further news on a story I posted about in late September, it has now surfaced that Zoom has cancelled academic events scheduled to discuss its previous cancellation. Beyond the political merits of the issue, from a business standpoint I suspect that the company’s position will quickly become untenable and that if it persists in its current interpretation of its ToS its competitors will crowd it out of the academic market (already, one of the cancelled events was able to migrate to Google Meets). Telling universities to refrain from discussion is like telling a rivet factory to refrain from metalworking; the fact that in this situation it has become impossible to draw a frame around an issue, so as to modify the sociolinguistic norms that preside over it, has produced a surreal outcome: this is not an equilibrium, and is destined to change.