– Kevin B Korb
21 February 2021
When Tim Berners-Lee invented the worldwide web in 1990 (Not the internet! The internet was effectively invented in the 1960s and first given form as the “ARPANET” well before Berners-Lee or Al Gore became involved.), a kind of starry-eyed idea that the internet would spread a love of knowledge and freedom around the world, if it were simply left alone by politicians, was very prominent. Most of us, having experienced the rise of social media on the back of the web and the internet, have since then been disabused of such notions, if we ever had them. While the webnet has made science, journalism and entertainment very much more widely available than ever before, it has notoriously also made available huge amounts of misinformation and disinformation, as well as private and semi-private places in which correspondents from around the world can cooperate in burnishing stories embodying them and so spread misunderstanding like a dark cloud over the world. Also notoriously, well-financed state organizations, such as St Petersburg’s IRA, can and do orchestrate disinformation campaigns using unsuspecting useful idiots. In short, much of the internet now operates as a kind of intellectual cesspool, one which no one is yet cleaning up.
In keeping with this spirit of an unregulated wild west, social media have thus far escaped much of the burden of direct regulation. Google, Amazon, Facebook, Twitter, Netflix and others have captured huge amounts of personal data from their users and converted that information into huge revenue streams, in large part through capturing much of the worldwide advertising market. The regulatory environment that used to keep broadcasters and news organizations in check no longer applies. Of course, social media companies are corporations and come under existing regulations that apply to most corporations, such as tax and anti-trust laws. But their basis in new technology means that laws and regulations have, for the most part, yet to catch up with their behavior, influence and evasions. More than most traditional companies, for example, they have been highly adept at minimizing taxes, by having related companies provide services from low-tax countries and paying for them in high-tax countries, thus reducing profits where they hurt and maximizing them where they don’t. Of course, that’s an old game which manufacturing companies have played since well before Google or Twitter existed. However, moving manufacturing plants to low-tax districts is a good deal harder than moving around the nominal location of a web-based service, which can be provided from anywhere connected to the internet.
Not only do social media companies live in a low-regulatory environment, there are only poor prospects for that changing in an economic world largely dominated by an excess of a neoliberal ideology which views regulation as tantamount to corporate murder. However, it has never been more necessary to oppose this view: the threats to individual liberties and privacy posed by technology, both the communications technology of the internet and the emergence of applied AI, have never been greater and will not be controllable without proper regulation.
The Fairness Doctrine
The Fairness Doctrine was a part of the US Federal Communications Commission’s (FCC) regulatory framework from 1949 until 1987. The FCC did (and does) have regulatory authority over broadcast licences and used to enforce the Fairness Doctrine, which was:
The doctrine that imposes affirmative responsibilities on a broadcaster to provide coverage of issues of public importance that is adequate and fairly reflects differing viewpoints. In fulfilling its fairness doctrine obligations, a broadcaster must provide free time for the presentation of opposing views if a paid sponsor is unavailable and must initiate programming on public issues if no one else seeks to do so (The Fairness Doctrine, 2008).
In a shorter form, the Fairness Doctrine required broadcasters to cover issues of public interest in a manner that was fair and balanced. This was not interpreted as providing equal time for all points of view, but some coverage for important issues, plus some coverage for legitimate alternative points of view to what broadcasters had already presented. The doctrine had teeth and led to the cancellation of multiple licences (e.g., Watson, 2015; Parker, 2008). In fact, its effectiveness in supporting fair and balanced debate is arguably the reason that Ronald Reagan and his Republican supporters scrapped the rule in 1987.
USA Today has done a "Fact Check" on whether the scrapping of the Fairness Doctrine gave rise to the polarization in the US media most clearly exemplified by Fox News. They conclude that this is untrue, since the FCC's jurisdiction was limited to broadcasters, and cable news was not considered a broadcaster. Their argument is defective, however. USA Today acknowledges that the Fairness Doctrine was effective in getting individual licensees to provide balanced coverage of issues. But they ignore the fact that the scope of jurisdiction of the FCC was in dispute in the 1980s. Already in 1968 the Supreme Court acknowledged their jurisdiction over cable, despite cable not technically being a broadcast medium, on the grounds that otherwise the FCC would be unable to fulfill its intended role. Then in 1972 the FCC explicitly imposed the Fairness Doctrine on cable operators. During the 70s and 80s these rules were slowly wound back, until, under Reagan-appointed commissioners, the FCC scrapped the rule, with Reagan vetoing a Congressional attempt to retain the Fairness Doctrine. In other words, before 1987 the Fairness Doctrine was successfully applied to cable, and Reagan terminated that, not just for cable, but also for broadcasting. The result was that a cultural acceptance of news programs being balanced dissipated, in both cable and broadcasting. Fox News would never have been possible without these actions, despite their 100% phony slogan of being "Fair and Balanced" themselves. The USA Today "Fact Check" is well worthy of Three Pinocchios.
However, what I want to target here is Oreskes and Conway’s (2010) argument, in their otherwise excellent Merchants of Doubt, that the Fairness Doctrine did a great deal of damage to public discourse by making false equivalencing (“what aboutism”) a kind of norm, in counterpoint to the criticism that its scrapping has done damage by fostering polarization (see box above). They provide a detailed and well-argued account of how false equivalencing has undermined the public discussion, and so the public decision making, surrounding the harms of tobacco use, acid rain, pesticides, ozone degradation through CFCs and anthropogenic global warming. These issues are all importantly linked. They all have spawned devoted groups of deniers who fervently oppose regulatory measures for minimizing the harm caused by related industries — and these groups are largely overlapping, fueled by a common set of rightwing think tanks and common pools of money. (About the money, an especially revealing read is Nancy MacLeans’ Democracy in Chains.) While it’s clear that the scrapping of the Fairness Doctrine has encouraged voices of extremism, especially those backed by Rupert Murdoch, it’s also arguable that the Fairness Doctrine itself gave cover to extremists demanding to be heard on these and other topics — because it’s only fair! — when by rights they would have had a much smaller voice, should volume be in any way proportional to the merits of the cases being advanced. In a nutshell, that is Oreskes and Conway’s argument.
For those who look to the potential value of regulation returning to the role of promoting effective and useful public discourse, and to the Fairness Doctrine specifically as a model for that, this is an argument that must be addressed. The primary weakness in it is Oreskes and Conway’s elision of two key features of the Fairness Doctrine (at any rate, on my interpretation and that of Wikipedia, 2021). It explicitly does not provide for equal time for differing viewpoints, but some reasonable, if lesser, amount of time for legitimate alternative viewpoints (see box below). It provides for no (mandatory) time for illegitimate points of view. The legitimacy of differing points of view is up for debate in many cases, of course, and, when the Fairness Doctrine was in existence, legitimacy was ultimately settled by the courts, which have always been a rational backstop for deciding the limits of public discourse. Where the claims of a faction have been thoroughly discredited by science — as they have been in all the cases discussed in Oreskes and Conway’s book, and indeed already were at the times of the debate over their regulation — there is no need under the Fairness Doctrine to give any time to those points of view, nor would the courts force the presentation of illegitimate nonsense, regardless of the funds behind it. The push for false equivalency is indeed a prominent tactic of deniers of science, but, if it drew upon the Fairness Doctrine before 1987, then it did so without justification and under false pretences.
I am not an expert in the law, let alone in FCC law, but there are clear indications in US Supreme Court findings supporting my lesser point that legitimacy to some (unspecified) standard was required of a thesis or point of view before the Fairness Doctrine could be invoked, which I quote below. Regardless, even if my interpretation is mistaken, the more important point is that it could be true. If we are to adopt some version of a Fairness Doctrine for use in regulating social media, it needs to be one which supports legitimacy and rules out the discredited. Here are some quotes pertinent to the lesser issue (note that allowing disproven, illegitimate points of view a significant voice is clearly not in the public interest): Referring to legislation supporting the Fairness Doctrine, the US Supreme Court observed: `Senator Scott, another Senate manager [of the legislation], added that: "It is intended to encompass all legitimate areas of public importance which are controversial," not just politics.' (US Supreme Court, 1969) `The statutory authority of the FCC to promulgate these regulations derives from the mandate to the "Commission from time to time, as public convenience, interest, or necessity requires" to promulgate "such rules and regulations and prescribe such restrictions and conditions . . . as may be necessary to carry out the provisions of this chapter . . . ." 47 U.S.C. 303 and 303 (r).[note 7] The Commission is specifically directed to consider the demands of the public interest... This mandate to the FCC to assure that broadcasters operate in the public interest is a broad one, a power "not niggardly but expansive."' (US Supreme Court, 1969) The Fairness Doctrine is repeatedly described as supporting broadcasting on important public issues, which would rule out, for example, giving time to flat-earthers. For example, "[licensees have] assumed the obligation of presenting important public questions fairly and without bias." (US Supreme Court, 1969) On the restrictions imposed by the Fairness Doctrine on broadcasters' freedom of choice: "Such restrictions have been upheld by this Court only when they were narrowly tailored to further a substantial governmental interest, such as ensuring adequate and balanced coverage of public issues." (US Supreme Court, 1984)
Given the widespread and growing flow of misinformation and disinformation on social media, the Fairness Doctrine, or rather some descendant of it also incorporating protection of the public from promulgation of the illegitimate, could provide the justification and means of choking that flow and so allowing social media to serve the truly useful purpose of supporting a “marketplace of ideas” instead of being a poisonous wetmarket spawning misinformation pandemics.
In short, regulations sharing purpose with the Fairness Doctrine are fair game for nations wanting to foster valuable public debate, which is part of the foundation of any democracy. Such regulation is needed for traditional broadcasters. The US Supreme Court extended the doctrine to cable networks on the grounds that the FCC could not fulfill its function if cable were excluded. On the very same grounds, but with even stronger force, such regulation needs to be applied to internet- and web-based social media, which have collectively outgrown both broadcasting and cable in their reach and importance for public debate.
The EU’s General Data Protection Regulation (GDPR) was introduced in 2016, establishing EU-wide principles for the protection of personal data, including rights to informed consent to the collection of data and the restriction of its use to the purposes for which consent was given. The GDPR also provides for enforcement powers, with each member country having a Data Protection Authority (DPA) to investigate and prosecute violations. Of course, those US tech companies which have so successfully “monetized” your data objected long and loud to the GDPR. Once it became operational, however, they went quiet, since, while there are compliance costs, compliance is in fact feasible and doesn’t stop them earning money in Europe. The rest of the world benefits from EU regulation in a minimal way, when companies are either obliged to obey the GDPR because of doing business with the EU or where they simply prefer a uniform way of doing business across jurisdictions.
The social media goals for data acquisition are largely to do with (mis)using the data for better targeting advertising, because that’s largely where the revenue comes from. If users voluntarily agree to such use, knowing the scope of the usage in advance, that’s fair enough. And that’s exactly what the GDPR allows, as well as what it limits data usage to. But the threats involving data acquisition are now hugely greater than simply making money. Facial recognition software is now routinely used by police. With much of the world playing catch up with Chinese-level camera surveillance, the potential for abuse of such information is enormous. Deep fake technology has the potential to weaponize personal data, directing much more effectively manipulative advertising at you, as well as using your data to spread more effective and manipulative misinformation about you and groups you belong to. Identity theft using deep fake videos will be much easier than that using earlier technology, for example. As another example, blackmail and extortion based on compromising information have long been lucrative activities for criminals; blackmail and extortion based on compromising deep fake misinformation will be orders of magnitude easier. Deep Fakes will not for long be limited to passive videos and audios; they will soon be extended to real-time interactive simulations of a targeted victim, providing even more persuasive power for fakery (Wang, 2019). With the near-term development of the “Internet of Things” — wiring all of our refrigerators, cars, air conditioning systems, etc. into the internet — the raw data on which Surveillance Capitalism operates will expand exponentially for the foreseeable future. The rise, and combination, of Big Data and Machine Learning using Big Data (e.g., Deep Fakery) portends parlous times on the net. Berners-Lee style enthusiasm for a “free range” on an internet wild west is no longer so much quaint as simply dangerous.
There is still news reporting and journalism in the world. There are both private and public organizations which put a good deal of effort and money into tracking what’s happening of interest around the world and presenting it to their audiences. This is so despite, for example, US newspaper advertising revenue having declined about 55% since the invention of the worldwide web to 2018 (per a Pew Center Report), whereas in the same period US social media ad revenue grew from nothing to 3,571 times that of the newspapers (i.e., 357,100% more). Since news organizations originate and curate their news and opinion reports, it is reasonable to hold them accountable for the content, for example by allowing some defamation actions against them. Social media, on the other hand, simply offer platforms for others to write or speak upon. Especially given the size of their memberships, it is both impossible and unreasonable to expect them to police the content of posts in the same way as news media. Or, at least, that is the common view.
Indeed, this is the rationale behind the now famous Section 230 of the US Telecommunications Act of 1996 (“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.”) Making social media responsible for content, when that content is put out by millions or even billions of people, would make social media unviable. Not even the AI of the foreseeable future would be able to police their posts effectively enough to avoid huge legal costs. It’s possible, of course, that the courts would find a balance between the financial and operational health of social media organizations and their legal opponents in such a new environment, with Section 230 removed, but there is no guarantee. The nominal reason that Donald Trump wanted Section 230 deleted was that social media were censoring rightwing voices. But the reality is, of course, that without the protection of Section 230 social media would be forced to censor even more heavily or else just be shut down.
What I am calling for here is an even heavier forced censorship, in addition to new privacy protections. Illegitimate claims pushed by Russia’s IRA, Q, climate deniers, big monied interests must lose their voices. They are diminishing, not enriching, public debate. Illegitimacy, exposed by science and confirmed by courts, must not be heard. How exactly to make such a mandate operational is an open question. There need to be independent authorities for initial judgments of fairness and legitimacy in analogy to GDPR’s Data Protection Authorities, where independence refers to a lack of dependency upon both the social media organizations and a nation’s politics. In view of the latter, unlike the DPAs, it would be best if the new authority were explicitly international. There are plenty of precedents in international law for such organizations. Successful examples of international regulatory bodies include the UN’s Universal Postal Union, which coordinates worldwide postal systems, the UN’s International Maritime Organization, which regulates international shipping, or the World Trade Organization, which regulates international trade.
While forcing social media to report matters fairly, including intervening in their users’ mis/disinformation, would be a new burden on them, it is nothing like the threat revocation of Section 230 would raise. If social media are judged directly responsible for misinformation, through perhaps negligence, then penalties might be in order. But if a UN authority points out that some accounts are spreading disinformation, the existing practice of deleting those accounts would likely suffice to contain the matter. There is no need to threaten social media with punitive monetary damages. What we need is for public discourse to converge on civility, not suppression.
Open platforms resemble the public square, and the free discussion of politics that takes place on these platforms resembles an open marketplace of ideas. (Schweppe, 2019).
What about free speech? If social media organizations are to be made and held responsible for providing something akin to a digital public square — a forum where any public issue may be discussed within the bounds of public decency and fairness — then won’t our right to free speech be infringed? On any reasonable understanding of these terms, the answer is “No”. The requirement of public decency has always been maintained for public squares. Fairness was introduced in the US in the mid-twentieth century, but appropriately. It was always at least implicitly a requirement of real public squares in any case: any citizen who pulled out a bullhorn and spoke over everyone else would have been hauled off for disturbing the peace.
Democracy depends upon free speech. And it is fitting that it is included in the very first amendment in the US Bill of Rights. But that right has never been absolute, nor can it be. The community decides what constraints to put upon it, but there is no community which allows unfettered a freedom to abuse, incite hatred, or endanger people. Somewhat older style libertarianism asserted individual rights, including speech rights, up to, but not beyond, the boundaries of others’ rights (i.e., there is an obligation to “refrain from violating the rights of others”, van der Fossen, 2019). Since libertarianism recently married neoliberal fanaticism, however, it seems like all constraints are off: individual rights, for example, now extend to refusing to wear masks during a pandemic, that is, to a newly invented right to infect and kill other people. The logical extension of such libertarianism to all varieties of behavior would turn libertarian moral philosophy into Thrasymachus’s “might makes right” — that is, a full-throated cry to be evil.
Oreskes and Conway meticulously trace much of this neoliberal-libertarian fusion back to the monied interests fighting against regulation in the public interest of the lucrative businesses of fossil fuel extraction, agriculture, manufacturing and tobacco. They maximize profits by putting all the burden of their “externalities” — pollution — on the public. Neoliberal libertarianism is a con.
Social media tech companies are playing an extension of that con. They adopt internal policing practices to monitor and control content exactly and only insofar as it is necessary to stave off the kind of regulation I’m calling for here. To the extent that regulation can be forestalled or avoided, the burdens of social media’s externalities can be foisted onto the public. These externalities include the polarization of public debate, the domination of monied interests of that debate through targeted advertising and the Murdoch press, the creation and magnification of extremely damaging conspiracy theories, the promotion of hate over cooperation. We cannot wait another generation to protect the public interest from these con artists.
Social media have grown from nothing to dominating public discussions around the world. They have evaded regulation so far very successfully in most cases. The growth in data collection, the rapid advance of AI technologies, the imminent flourishing of Deep Fake technology, the proven ability of interested parties to initiate and promote disinformation campaigns all point to an urgent and growing need for proper regulation of social media. The goals of such regulation should include at least the protection of personal data, the shackling of disinformation and the curbing of misinformation. The GDPR and the Fairness Doctrine provide some successful models — starting points — for considering such regulations. But the social media themselves are far richer and more far-reaching than the media of the past, spanning the worldwide web, so the regulations required must likewise be worldwide, preferably operating across borders as a neutral international body under international laws.
I thank anonymous reviewers for their helpful criticisms.
Fairness Doctrine (2008). West’s Encyclopedia of American Law, edition 2. Accessed February 7 2021 from https://legal-dictionary.thefreedictionary.com/Fairness+Doctrine
Parker, Everett (2008). The FCC & Censorship. Democracy Now. Accessed 7 February, 2021. https://www.democracynow.org/2008/3/6/the_fcc_censorship_legendary_media_activist
United States Supreme Court (1969). RED LION BROADCASTING CO. v. FCC(1969) No. 717 Argued: Decided: June 9, 1969 395 U.S. 367, 89 S. Ct. 1794, 23 L. Ed. 2d 371, 1 Med. L. Rptr. 2053 (1969).
United States Supreme Court (1984). 468 U.S. 364 104 S.Ct. 3106 82 L.Ed.2d 278 FEDERAL COMMUNICATIONS COMMISSION v. LEAGUE OF WOMEN VOTERS OF CALIFORNIA et al. No. 82-912. Supreme Court of the United States Argued Jan. 16, 1984. Decided July 2, 1984.
Schweppe, J (2019). Hawley Defends the Public Space. First Things https://www.firstthings.com/web-exclusives/2019/06/hawley-defends-the-public-square. Accessed 16 Feb 2021.
van der Vossen, Bas (2019). “Libertarianism”, The Stanford Encyclopedia of Philosophy (Spring 2019 Edition), Edward N. Zalta (ed.), URL = <https://plato.stanford.edu/archives/spr2019/entries/libertarianism/>. Accessed 16 February, 2021.
Wang, G.E. (2019). Humans in the Loop: The Design of Interactive AI Systems, Stanford University Human-Centered AI.
Watson, Roxanne. “Red Lion Broadcasting Co. v. FCC”. Encyclopedia Britannica, 11 Sep. 2014, https://www.britannica.com/event/Red-Lion-Broadcasting-Co-v-FCC. Accessed 7 February, 2021.
Wikipedia (2021). “FCC Fairness Doctrine“. Wikipedia. Accessed 21 February 2021.