Journal of International Affairs Editorial Board
DISINFORMATION AND DEMOCRACY: THE INTERNET TRANSFORMED PROTEST BUT DID
NOT IMPROVE DEMOCRACY
Author(s): Anya Schiffrin
Source: Journal of International Affairs , Vol. 71, No. 1, THE DEMOCRACY ISSUE
(FALL/WINTER 2017), pp. 117-126
Published by: Journal of International Affairs Editorial Board
Stable URL: https://www.jstor.org/stable/10.2307/26494367
REFERENCES
Linked references are available on JSTOR for this article:
https://www.jstor.org/stable/10.2307/26494367?seq=1&cid=pdf-
reference#references_tab_contents
You may need to log in to JSTOR to access the linked references.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms
Journal of International Affairs Editorial Board is collaborating with JSTOR to digitize, preserve
and extend access to Journal of International Affairs
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
fall/winter 2017 | 117Journal of International Affairs, Fall/Winter 2017, Vol. 71, No. 1.
© The Trustees of Columbia University in the City of New York
Anya Schiffrin1
Professor at Columbia University School of International and
Public Affairs
DISINFORMATION AND DEMOCRACY: THE
INTERNET TRANSFORMED PROTEST BUT
DID NOT IMPROVE DEMOCRACY
Abstract: Recent years have seen a marked shift in global attitudes toward
social media platforms. In 2011, Facebook was hailed as a platform that
would bring democracy to the world, Google was breaking new ground in
convenience and access to information, and the protests taking place in Iran,
Egypt, Tunisia, Bahrain, and many other countries were spurred in part by
bloggers and social media commentators who used the platforms to galvanize
people and encourage them to take to the streets. But, by 2017, we had learned
that although the Internet transformed protest, it has not much improved
democracy. Moreover, we learned again the lesson that the post-Cold War
democracies had apparently forgotten: that misinformation and propaganda
are powerful, and that repeating “big lies” can persuade susceptible people of
nonsensical and dangerous ideas. This essay will examine the various sources
and forms of disinformation that are most prevalent in today’s political and
media environment, the implications of this new reality on democracy, and the
ways in which government can and must respond.
2016 was the year that public opinion turned against social media and big
tech companies. In 2011, Facebook was hailed as a platform that would bring
democracy to the world. We were grateful to Google. The protests in Iran, Egypt,
Tunisia, Bahrain, and many other countries were spurred in part by bloggers
and social media commentators who used social media to galvanize people and
encourage them to take to the streets.
By 2017, we had learned that although the Internet had transformed protest,
it has not much improved democracy. Moreover, we learned again a lesson that the
post-Cold War democracies had apparently forgotten: that misinformation and
propaganda are powerful and that repeating “big lies” can persuade susceptible
people of all kinds of nonsensical and dangerous ideas. This should not have been
a surprise, but critic Norah Ephron once said that “people have a shocking capacity
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Anya Schiffrin
118 | Journal of International Affairs
to be surprised by the same things over and over again.” The question now is what
to do. Regulation of social media platforms comes up repeatedly, but of what kind
is less clear.
Of course, it was not all boundless optimism in 2011. Even before the Arab
Spring, critics like Evgeny Morozov had warned that the Internet could be used as
a tool of surveillance, and Cass Sunstein and Markus Prior had warned that giving
everyone the right to select the news they wanted to read would compromise the
marketplace of ideas.2
What wasn’t clear at the time was the scale of the disinformation
that would flood the Internet and the effect this could have on voting. It
didn’t seem plausible that people would be so susceptible to lies on the Internet
and that they would resist reasoned attempts to explain facts, that truth would
seem not to matter. By 2017, it became clear that anger over social inequality
had turned into the conflation of privilege with expertise, and that many hated
experts. Global demagogues stoked the fires of this hatred with constant attacks
on the judiciary, the media, science, climate change scientists, and any institution
that could undermine their agendas.3
At the time of this writing, it does not seem
an exaggeration to say that disinformation spread by social media has undermined
the functioning of democracy globally. But if social media is undermining our
ideas of democracy, how can we solve the problem without also undermining the
processes of democracy?
Looking Back at the Optimistic Debates of 2010 and 2011
A few months before the Arab Spring, two books were published that discussed
the role of digital technology on society and democracy. One, The Net Delusion, by
Evgeny Morozov, got widespread attention for its robust attack on the “techno
optimists” who were foolish enough to believe that the likes of Facebook could
bring about social change and force governments to become more accountable and
democratic. “A dictator who answers his cell phone is still a dictator,” Morozov
wrote. Further, he argued, sophisticated authoritarian regimes would be able to
use the web not just for propaganda purposes but to track their opposition; so that
digital technology was actually helping authoritarian regimes survive—a point
that, with the passage of time, no longer seems novel.
But it was a book that got far less attention that turned out to be more immediately
prescient. Using a data set of Islamic countries from around the world,
political science professor Philip Howard argued that digital technology was
bringing communities together, providing vast amounts of information to closed
societies; and forcing governments to become more accountable. This in itself was
making the world more democratic. A few months after these two books appeared,
the Arab Spring revolutions cemented the idea that digital technology was a force
This content downloaded from
fff:ffff:ffff:ffff:ffff:ffff on Thu, 01 Jan 1976 12:34:56 UTC
All use subject to https://about.jstor.org/terms
Disinformation and Democracy
fall/winter 2017 | 119
for political change. The new conventional wisdom became that the Internet had
dispersed the power of international organizations and governments and emerging
communities online have undermined traditional state authority. From mobile
money to crowd-mapping crises and bringing citizens together to report on news,
distribute information, and organize politically, digital technology had the potential
to leave obsolete power structures behind. Scholars such as Zeynep Tufekci
and Jennifer Earl argued that the “affordances” of the web had transformed protest
in part by lowering the amount of time, effort, and money it required, and by
making it easier to gather large numbers of disparate people from around the world
into new communities.4
The recent scholarship makes it clear that the nature of
activism and protest had changed and that the web was not just recreating earlier
forms of protest.
2016, and the Values of Big Tech: Make Billions by Spreading Millions
of Dangerous Lies
By 2016, it was apparent that something had gone very wrong; many of the
optimists of 2010 and 2011 had changed their thinking, warning of the dangers
of digital technology. Wael Ghonim, whose Facebook pages are credited with galvanizing
the protests in Egypt, declared that the web had become a “mobocracy.”
Along with Emily Parker, Ghonim launched a site called Parlio that was meant to
encourage civilized and expert discourse
online about vital topics of the
day.5
The site never garnered a large
following but was bought by Quora
and eventually closed down. Philip
Howard began studying bot activities
and disinformation during the 2016
elections in Europe and the US and
came up with some startling numbers
about the amount of disinformation
shared over Twitter.6
Howard and
his colleagues at the Computational
Propaganda Research Project at the
Oxford Internet Institute looked at
seven million tweets that used hashtags related to the 2016 election between
November 1 and November 11 in 16 swing states. After developing a typology
based on the URLs included in these tweets, which sorted all tweets into six
categories including professional political content such as government and campaign
sources, professional news outlets, and polarizing and conspiracy content;
“How many Facebook
users saw what kinds of
disinformation, when they
saw it, and how often this
took place is unclear, in
part because Facebook
consistently refused to
provide information to
researchers about what
political advertisements
it displayed and who saw
them.”
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Anya Schiffrin
120 | Journal of International Affairs
Howard and his colleagues found overwhelming levels of news from Russian
outlets, Wikileaks, and “junk news” sources flooding Twitter just before the 2016
US presidential elections.7
Howard and his colleagues also noted that in these 16
swing states, levels of “junk” and polarizing news exceeded those of the United
States as a whole.
How many Facebook users saw what kinds of disinformation, when they saw
it, and how often this took place is unclear, in part because Facebook consistently
refused to provide information to researchers about the political advertisements it
displayed and who saw them. According to Howard:
At this point Facebook is the single most important platform for public life in
the vast majority of countries. Its advertising algorithms allow politically motivated
advertisers to reach a purposefully selected audience. Unfortunately, the
company provides no public record of the political advertisements it serves to
users, and there is no systematic way for analysts to measure the spread of junk
news. For other kinds of media, political candidates must declare their sponsorship
and file copies with the FEC. In the US election, for example, Trump
spent 70 million on Facebook ads we’ll never see.8
Without knowing what people saw, how many times, and for how long, it is
difficult to know whether or how much of an effect disinformation had on voting
patterns. An early study was released in early 2017 by economists Hunt Allcott and
Matthew Gentzkow and concluded that “fake news” had no effect on the US elec-
tions.9
The earlier draft of the study, however, was based on some assumptions that
seemed shaky at best, including the assumption that one piece of fake news was
comparable to 36 negative campaign advertisements.10
By the time the paper was
published, it had already been circulated widely and read closely by senior people
at Facebook. Another questionable part of the study includes the authors giving
their subjects “placebo” headlines so as to compare their judgement of real news
with their judgement of fake news. However, understanding media effects is far
more complicated than doing a scientific experiment in which randomized control
groups are necessary. It is possible that there was a backfire effect resulting from
the placebo headlines, or even what scholars call “misinformation persistence.”11
For all of these reasons, we don’t actually know whether or how disinformation
affected the 2016 elections. The use of social media to move public opinion is a
relatively new phenomenon, and the speed and volume of the incorrect messages
transmitted by social media may be unprecedented.
However, the example of Fox News is instructive. When Fox News began,
it was assumed that people watching it would be influenced to vote for the
Republican party, and early research suggested that this was indeed the case.12
In
the fall of 2017, a more definitive paper was published in the American Economic
Review that solved the question of causation.13
Consumption of Fox News pushed
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Disinformation and Democracy
fall/winter 2017 | 121
voters to vote for Republican candidates. The prevalence of junk news also suggests
that voters with low exposure to information participated in elections at high rates,
perhaps a different trendline from past elections, when the assumption was that
those who were uninformed and didn’t follow the news were also the people who
did not vote.
Propaganda, lies, and “truthiness” have been around for hundreds of years
and used by many political candidates, corporations, and religions to persuade
and mislead. What is different today is the speed and volume of disinformation.
We simply do not know what it means for the electorate when millions of Russian
propaganda messages are targeted at swing states. We can guess, but the research
has not yet been done, and the information is not available for us to know with
absolute certainty.
Even so, it is not too early to take action. When there is a strong possibility of
danger, society must act. Governments don’t wait for everyone to be in an automobile
accident before mandating that air bags should be put in every car. Now
is the time to consider low-hanging policy measures that may help the situation.
All policy involves tradeoffs, but we need to consider measures are acceptable to a
democratic society.
MIT researcher Yochai Benkler said at an October 2016 talk at Columbia
University’s School of International and Public Affairs, that there are five parties
circulating fake news:
1. Bodies thought to be close to the Russian government that circulate propaganda
and disinformation with the intent of sowing confusion and distrust;
2. Right-wing US groups such as Breitbart;
3. Groups that make money from circulating disinformation such as the notorious
Macedonians profiled by Buzzfeed in the fall of 2016;14
4. Formal campaigns using behavioral marketing tools, such as Cambridge
Analytica;15
5. Peer-to-peer distribution networks, including the far-right activists of 4Chan.
Benkler says:
The problem is potentially sufficiently serious enough that we should spend
a lot of money quickly to figure out what is happening so we know what
measures to take. At a minimum we should support transparency in political
advertising and that should include anyone paid to comment online on (or)
spread political information even if it’s by marketing companies as well as the
commercial equivalent of the 50-cent army.
While millions of dollars are being spent on research, there should be a focus
on policy prescriptions that can be put in place quickly. One example of policymaking
which involves an acceptable tradeoff was the bi-partisan bill introduced in
the US Senate in October 2016 that would have required social media companies
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Anya Schiffrin
122 | Journal of International Affairs
to tell the Federal Elections Commission the source of funding for online political
advertisements.16
Just as we require disclosure for political advertising on television,
so too should we disclose the sources of paid information online. Technology
expert Julia Angwin notes that such laws would not cover paid commenters, but
Benkler says that the law could be expanded to cover people paid to comment
online.
In many cases, countries with laws against hate speech and incitement will
need to find democratic ways to enforce them online so that the fight against disinformation
does not become an excuse for corporate and government censorship.
Asking big tech companies to deal with the problem on their own opens the way
to corporate censorship, free expression advocates have consistently warned.17
On
the other hand, it is important not to let technology companies use free speech as
an excuse not to take action.
In their comprehensive report on “Information Disorder” for the Council of
Europe, Claire Wardle and Hossein Derakhshan discuss the need to create cultures
of truth and provide recommendations for governments, journalists, technology
companies, and other parties.18
While many of the fixes being proposed, including
media literacy education and changes in ownership models, are long term, changing
norms and culture will be part of getting back to a culture of truth and evidence.
The topic is too important to leave to tech companies to handle alone and without
disclosure. Government, academia, and civil society need to lead the conversation
on how to address the problem of the millions of lies and propaganda mentions
that can spread so quickly on social media. E-Bay creator and philanthropist Pierre
Omidyar wrote in an October 2017 op-ed:
Just as new regulations and policies had to be established for the evolving
online commerce sector, social media companies must now help navigate the
serious threats posed by their platforms and help lead the development and
enforcement of clear industry safeguards. Change won’t happen overnight, and
these issues will require ongoing examination, collaboration and vigilance to
effectively turn the tide.19
In fact, more needs to be done. Columbia law professor Tim Wu believes that
Facebook should become a nonprofit or public benefit corporation, and Columbia
University professor Joseph E.Stiglitz argues that Facebook is similar to a public
utility and should be strongly regulated. Privacy, taxation and distribution of dis/
misinformation are all areas where there needs to be strong global regulation of the
tech and social media sectors. There are a number of options. It seems that there
is an emerging consensus around cracking down on tax avoidance and protecting
privacy. However, it is likely that some European countries will pass laws regulating
the dissemination of free speech, just as Germany has done. Regulations will
look different in different countries, as it will be hard to obtain a European-wide
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Disinformation and Democracy
fall/winter 2017 | 123
consensus. Facebook, of course, argues that ultimately these regulatory mechanisms
will be copied by authoritarian regimes.
Implications for Democracy
The implications of these developments for democracy are enormous. Is the
Internet killing our democracy and paving the way for uninformed mob rule?
Democracy rests upon the assumption of an educated populace; this is part of
why public education is so important. Understanding the important issues of the
day, as well as government representatives’ positions on these issues, is necessary
for citizens to participate actively in a democracy. Without this knowledge, voting
decisions may be arbitrary, and government can be based on capturing voters or
pandering, and can cease to be a truly functioning democracy.
The problem of misinformation on the Internet has come at a dangerous time,
when growing resentment over inequality and the worsening state of the American
middle class have stoked a deep mistrust in institutions of education, science, and
media that have traditionally served to keep “false facts” and demagoguery at bay.
Citizens are increasingly turning to the Internet, a forum for distributing information
that does not adhere to typical standards of truth, scientific inquiry, and
evidence-based news and information. At the same time, the institutions that have
typically distributed information to citizens are being usurped. Consequently, for
the average American citizen, distinguishing between true and false information
has only become more difficult.20
Polling data suggests that we live in a country where a large part of the populace
is either unable or unwilling to accurately educate themselves on the reality of
their country and leaders. An uninformed citizenry of this type is unable to act in
its own best interest when electing leaders and representatives. If misinformation
and fake news campaigns truly do frustrate citizens’ attempts to educate themselves—or,
even worse, actively manipulate citizens into believing false information—then
the very foundations of democracy are at risk.
Anya Schiffrin is the director of the Technology, Media, and Communications specialization
at Columbia University’s School of International and Public Affairs where she teaches courses
on media and development and innovation. Among other topics, she writes on journalism and
development, as well as the media in Africa and the extractive sector. Schiffrin spent 10 years
working overseas as a journalist in Europe and Asia and was a Knight-Bagehot Fellow at
Columbia University’s School of Journalism in 1999–2000. Schiffrin is on the Global Board
of the Open Society Foundations and the advistory board of the Natural Resource Governance
Institute and the American Assembly. Her most recent books are African Muckraking: 75
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Anya Schiffrin
124 | Journal of International Affairs
Years of African Investigative Journalism (Jacana 2017) and Global Muckraking:
100 Years of Investigative Reporting from Around the World (New Press, 2014).
NOTES
1
  Acknowledgments: Anamaria Lopez (CC 2018) assisted with research and writing the conclusion.
2
  Evgeny Morozov, The Net Delusion: The Dark Side of Internet Freedom (New York: Public Affairs,
2011); Cass Sunstein, #Republic: Divided Democracy in the Age of Social Media (Princeton: Princeton
University Press, 2017); Markus Prior, Post-Broadcast Democracy: How Media Choice Increases Inequality in
Political Involvement and Polarizes Elections (New York: Cambridge University Press, 2007).
3
  Heather Brookes in conversation, 2016.
4
  Jennifer Earl and Katrina Kimport, Digitally Enabled Social Change (Cambridge, MA: MIT Press,
2011); Zeynep Tufekci, Twitter and Tear Gas: The Power and Fragility of Networked Protest (New Haven:
Yale University Press, 2017).
5
  Nathan Gardles, “Wael Ghonim: We Have a Duty to Use Our Social Media Power to Speak the
Truth,” Huffington Post, 29 October 2016, https://www.huffingtonpost.com/entry/wael-ghonim-social-
media_us_580e364ae4b000d0b157b53a.
6
  Lisa-Maria Neudert, Bence Kollanyi, and Philip N. Howard, “Junk News and Bots during the
German Parliamentary Election: What are German Voters Sharing over Twitter?” (Oxford, UK:
Project on Computational Propaganda, 2017); Craig Timberg, “Propaganda Flowed Heavily Into
Battleground States Around Election, Study Says,” Washington Post, 28 September 2017, https://www.
washingtonpost.com/business/technology/2017/09/27/32855bba-a3a0-11e7-ade1-76d061d56efa_story.
html?utm_term=.03a84c04b6f5.
7
  Phillip M. Howard et al., “Social Media, News and Political Information during the US Election:
Was Polarizing Content Concentrated in Swing States?” (Oxford, UK: Project on Computational
Propaganda, 2017).
8
  Phillip Howard in conversation, 29 July 2017.
9
  Hunt Alcott and Matthew Gentzkow, “Social Media and Fake News in the 2016 Election,” Journal
of Economic Perspectives 31, no. 2 (2017), 211-236.
10
  The paper includes the placebo headlines but does not explicitly mention the 1:36 fake news
to campaign ad ratio, stating instead, “How much this affected the election results depends on the
effectiveness of fake news exposure in changing the way people vote. As one benchmark, Spenkuch
and Toniatti (2016) show that exposing voters to one additional television campaign ad changes vote
shares by approximately 0.02 percentage points. This suggests that if one fake news article were about
as persuasive as one TV campaign ad, the fake news in our database would have changed vote shares
by an amount on the order of hundredths of a percentage point. This is much smaller than Trump’s
margin of victory in the pivotal states on which the outcome depended.”
11
  Man-pui S. Chan, Christopher R. Jones, Kathleen Hall Jamieson, and Dolores Albaraccín,
“Debunking: A Meta-Analysis of the Psychological Efficacy of Messages Countering Misinformation,”
Psychological Science 28, no. 1 (2017), 1–16. This paper conducted a meta analysis of studies on how to
effectively debunk misinformation. Among other things it found that simply printing a correction or
warning does not bring people to change their minds but rather can result in “misinformation persistence.”
The authors point to Schwarz et al. (2007), who found that corrections often inadvertently
strengthen the misinformation they intend to contest when they merely ask people to “consider the
opposite” of stated facts. This risk is lowered only when a well argued, detailed debunking message is
offered (Jerit 2008). The authors concluded that detailed corrections produce a stronger debunking
effect than non-detailed ones. However, they can also inadvertently perpetuate misinformation.
12
  Stefano DellaVigna and Ethan Kaplan, “The Fox News Effect: Media Bias and Voting,” NBER
Working Paper 12169 (April 2006), http://www.nber.org/papers/w12169.
13
  Gregory J Martin and Ali Yurukoglu, “Bias in Cable News: Persuasion and Polarization,” American
Economic Review 107, no. 9 (2017), 2565-99.
14
  Craig Silverman and Lawrence Alexander, “How Teens in the Balkans are Duping Trump
Supporters with Fake News,” Buzzfeed News, 3 November 2016, https://www.buzzfeed.com/craigsil-
verman/how-macedonia-became-a-global-hub-for-pro-trump-misinfo?utm_term=.ygVV9NJpA#.jqbThis
content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
Disinformation and Democracy
fall/winter 2017 | 125
nQ58jk.
15
  Carole Cadwalladr, “The Great British Brexit Robbery: How Our Democracy Was Hijacked,” The
Guardian, 7 May 2017, https://www.theguardian.com/technology/2017/may/07/the-great-british-brexit-
robbery-hijacked-democracy.
16
  Kenneth P. Vogel and Cecilia Kang, “Senators Demand Online Ad Disclosures as Tech Lobby
Mobilizes,” New York Times, 19 October 2017, https://www.nytimes.com/2017/10/19/us/politics/face-
book-google-russia-meddling-disclosure.html.
17
  Courtney C. Radsch, “Deciding Who Decides Which News is Fake,” Committee to Protect
Journalists, 14 March 2017, https://cpj.org/blog/2017/03/deciding-who-decides-which-news-is-fake.
php.
18
  Claire Wardle and Hossein Derakhshan, “Information Disorder: Toward an Interdisciplinary
Framework for Research and Policy Making” (Strasbourg, France: Council of Europe, 2017).
19
  Pierre Omidyar, “Pierre Omidyar: 6 Ways Social Media Has Become a Direct Threat to
Democracy,” Washington Post, 9 October 2017, https://www.washingtonpost.com/news/theworldpost/
wp/2017/10/09/pierre-omidyar-6-ways-social-media-has-become-a-direct-threat-to-democracy/?utm_
term=.194674e5e885.20, Pharr, Putnam, and Dalton, “A quarter century of declining confidence,”
5-25.
20
  Camila Domonoske, “Students Have ‘Dismaying’ Inability to Tell Fake News From Real,
Study Finds,” National Public Radio, 23 November 2016, https://www.npr.org/sections/thetwo-
way/2016/11/23/503129818/study-finds-students-have-dismaying-inability-to-tell-fake-news-from-real.
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms
This content downloaded from
86.49.247.175 on Tue, 07 Sep 2021 20:57:00 UTC
All use subject to https://about.jstor.org/terms