The National  Memo Logo

Smart. Sharp. Funny. Fearless.

Monday, December 09, 2019 {{ new Date().getDay() }}

Reprinted with permission from AlterNet.


Next week, Congress will hold dueling hearings on social media’s ongoing tug-of-war with malevolent partisans, neither of which will likely shed any light on the elephant in the tech sector’s living room: how their platforms’ design empowers propaganda—and they’re not turning off the switch.

Seen narrowly, the relatively more mature Senate Intelligence Committee will hear from Facebook, Twitter and possibly Google executives on “social media companies’ responses to foreign influence campaigns.” You can expect Facebook to recount how it’s taken down nearly 700 pages, personas, events, etc.—put up by Russian- and Iranian-connected government intelligence services intended to push people’s prejudicial buttons domestically and abroad, to turn their activism into real conflict.

The Senate inquiry is at least based in reality. Those defrocked pages were real, were taken down, and apparently, have traceable ties to overseas spy agencies. Whether or not they had any real-world impacts is another question. You can bet this scripted affair, where tech execs will be alerted beforehand what they will be asked about, probably will not touch on a recent German study that looked at Facebook’s role in 3,335 hate crimes and attacks on refugees in that European country in the past two years.

The locales where Facebook users spent the most time on the site had the most anti-refugee violence, the academics explained—and they said why that was the case. Facebook, like other social media platforms, gives users a semi-public place to congregate around values or ideas or issues that might not be socially acceptable. Every now and then, someone who goes deep in this spinning vortex leaves the toxic bubble and lashes out in the real world—with threats or real violence.

Facebook’s “algorithm is built around a core mission: promote content that will maximize user engagement. Posts that tap into negative, primal emotions like anger or fear, studies have found, perform best and proliferate,” the New York Times explained in an investigative report based on the German study. “Even if a minority of users express vehemently anti-refugee views, once they dominate the newsfeed, this can have consequences for everyone else… Over time, they [the most persistent posters] appeared to lose sight of the line separating trolling from sincere hate.” Strikingly, the academics found, “whenever internet access went down in an area with high Facebook use, attacks on refugees dropped significantly.”

That German study highlights the real threat from malicious online actors. It isn’t Russian meddling in the mechanics of voting in America; it’s trolls and obsessed individuals who are empowered by social media platforms that are giant surveillance machines, tracking and analyzing everything users do in the service of bringing paying advertisers to micro-targeted audiences. That technology is one thing when it comes to selling towels or appliances; it’s another when malevolent actors or arch-partisans or foreign spies use that platform to incite polarization, fan societal conflict and sow violence.

But one doesn’t have to go to Europe for examples of where social media has normalized bigotry, conspiracy, and grievance politics. You can watch President Trump’s latest MAGA rally in Indiana this week, where his faithful relished the way Trump blames everyone else (but them) for their problems. You can look at Thursday’s FBI arrest of a California man who threatened to kill Boston Globe journalists for being “the enemy of the people”—language that tracked Trump’s remarks the same day in Indiana calling CNN, the New York Times, and the Washington Post “dishonest, terrible people.” Or you can look at what other congressional Republicans are doing, playing copycat to Trump by accusing Silicon Valley media of censoring Trumpians like them.

On Wednesday, the Senate will listen to Facebook and Twitter executives tell how they have tamped down on “foreign influence.” Meanwhile, the House Energy and Commerce Committee will hold a more juvenile and patently absurd hearing on how the biggest social media platforms purportedly are censoring conservative views.

As Rep. Stephen Scalise, the Louisiana Republican, insistently tweeted, he would demand answers as to why “supporters of @realDonaldTrump are consistently suppressed by Twitter.” Not to be outdone, Rep. Jim Jordan, the Ohio Republican “shadow-banned” by Twitter amid sex-abuse allegations, replied, “Twitter says just a glitch in their algorithm. Really??”

Is using Twitter to accuse Twitter of censorship an example of suppressed speech? But this silliness doesn’t stop there.

Not one to be left behind, Utah Sen. Orrin Hatch on Thursday released a letter urging the Federal Trade Commission to look at Google’s “anti-competitive conduct,” because, as Trump and other right-wingers have incorrectly stated, Google has “decided to remove from its platform legal businesses that the company does not agree with.” Perhaps Hatch is referring to right-wing provocateur and hate-monger Alex Jones, who keeps losing in a lawsuit brought by parents of murdered children at Sandy Hook Elementary School, which Jones repeatedly said was a government-planned mass shooting.

What’s most insidious about all these claims is the reality that social media helped elect Trump and legitimize the far right as much as any other factor in recent political history. Brad Parscale, who ran Trump’s 2016 digital campaign and will manage his 2020 reelection effort, credited Facebook micro-targeted advertising machine for the victory. They tested 5.9 million ads, compared to Hillary Clinton’s campaign testing 66,000 ads—where whatever drew responses was more widely used.

If social media wasn’t so valuable to America’s troll-in-chief, his various 2020 re-election political committees would not be the top nationwide spender on Facebook ads in July and top spender on political ads on YouTube (owned by Google) in August. As technology writer Kara Swisher noted this week, “Tech has, in fact, been very, very good for conservatives from the get-go and very much so for Mr. Trump and his minions.”

If you want to inhabit a world without Trump’s daily tirades (this week’s threats to sanction Google, the World Trade Organization, Canadian milk producers, European automakers, and on and on) or a political culture that elevates the visibility of those fanning incivility and conflict, the culprit is partly the messaging machinery—not just the messed up messengers.

Pay close attention to these upcoming congressional performances. On the House side, right-wingers will make all kinds of noise about the sacredness of free speech to defend their trolling and bullying. And they likely will be unsatisfied after hearing explanations of why Google Search hasn’t ranked PJ Media as high in its search results as the New York Times (which even Silicon Valley libertarians said was an idiotic line of attack).

And on the more somber Senate side, it is likely that Facebook, Twitter and Google may say the time has come for some regulation of online political content (no doubt based on what they are already doing). But will they acknowledge the role of social media’s architecture and business model has played in empowering extremists, social divides and conflict? Will they suggest doing anything meaningful about it—such as turning down the megaphones they built—or will they dodge responsibility?

This article was produced by Voting Booth, a project of the Independent Media Institute.

Steven Rosenfeld is a senior writing fellow of the Independent Media Institute, where he covers national political issues. He is the author of several books on elections, most recently Democracy Betrayed: How Superdelegates, Redistricting, Party Insiders, and the Electoral College Rigged the 2016 Election (March 2018, Hot Books).



Start your day with National Memo Newsletter

Know first.

The opinions that matter. Delivered to your inbox every morning

The Arizona 2020 election "audit" under way

Screenshot from

As ongoing threats by Trump loyalists to subvert elections have dominated the political news, other Republicans in two key states—Florida and Arizona—are taking what could be important steps to provide voters with unprecedented evidence of who won their most close and controversial elections.

In both battleground states, in differing contexts, Republicans are lifting the curtain on the data sets and procedures that accompany key stages of vetting voters, certifying their ballots, and counting votes. Whether 2020’s election-denying partisans will pay attention to the factual baselines is another matter. But the election records and explanations of their use offer a forward-looking road map for confronting the falsehoods that undermine election results, administrators, and technologies.

In Republican-run Florida, the state is finalizing rules to recount votes by incorporating digital images of every paper ballot. The images, together with the paper ballots, create a searchable library to quickly tally votes and identify sloppily marked ballots. Questionable ballots could then be retrieved and examined in public by counting boards to resolve the voter’s intent.

“The technology is so promising that it would provide the hard evidence to individuals who want to find the truth,” said Ion Sancho, former supervisor of elections in Leon County, where Tallahassee is located, who was among those on a January 4 conference call workshop led by the Division of Elections seeking comments on the draft rule and procedures manual revisions.

Under the new recount process, a voter’s paper ballot would be immediately rescanned by an independent second counting system—separate from what each county uses to tally votes. The first digital file produced in that tabulation process, an image of every side of every ballot card, would then be analyzed by software that identifies sloppy ink marks as it counts votes. Several Florida counties pioneered this image-based analysis, a version of which is used by the state of Maryland to double-check its results before certifying its election winners.

“The fact that it has overcome opposition from the supervisors of elections is telling because the number one problem with the [elected county] supervisors is [acquiring and learning to use] new technology; it’s more work to do,” Sancho said. “The new technology doesn’t cost much in this case. Everyone has scanners in their offices already because every voter registration form by law must be scanned and sent to the Division of Elections.”

The appeal of using ballot images, apart from the administrative efficiencies of a searchable library of ballots and votes, is that the images allow non-technical people to “see” voters’ intent, which builds trust in the process and results, said Larry Moore, the founder and former CEO of the Clear Ballot Group, whose federally certified technology would be used in Florida recounts.

But Florida’s likely incorporation of ballot images into its recount procedures, while a step forward for transparency, is unfolding in a fraught context. In 2021, its GOP-majority state legislature passed election laws that are seen as winnowing voters and rolling back voting options. In other words, it may be offering more transparency at the finish line but is also limiting participation upstream.

The new recount rule is expected to be in place by this spring, months before Florida’s 2022 primaries and midterm elections. Among the issues to be worked out are when campaign and political party officials and the public would observe the new process, because the election administrators do not want partisans to intentionally disrupt the rescanning process. These concerns were raised by participants and observers on the teleconference.

The Arizona Template

In Arizona, Maricopa County issued a report on January 5, “Correcting the Record: Maricopa County’s In-Depth Analysis of the Senate Inquiry.” The report is its most substantive refutation of virtually all of the stolen election accusations put forth by Trump loyalists who spent months investigating the state's presidential election.

Beyond the references to the dozens of stolen election accusations put forth by pro-Trump contractors hired by the Arizona Senate’s Republicans, the report offered an unprecedented road map to understanding how elections are run by explaining the procedures and data sets involved at key stages.

The report explained how Maricopa County, the nation’s second biggest election jurisdiction (after Los Angeles County) with 2.6 million registered voters, verified that its voters and ballots were legal. It also explained key cybersecurity features, such as the correct—and incorrect—way to read computer logs that prove that its central vote-counting system was never compromised online, as Trump supporters had claimed in Arizona (and Michigan).

“I’ve never seen a single report putting all of this in one place,” said John Brakey, an Arizona-based election transparency activist, who has sued Maricopa County in the past and routinely files public records requests of election data. “Usually, it takes years to understand all this.”

Taken together, Florida’s expansion of recounts to include using digital ballot images, and Maricopa County’s compilation of the data and procedures to vet voters, ballots, and vote counts, reveal that there is more evidence than ever available to confirm and legitimize election participants and results.

For example, Maricopa County’s investigation found that of the 2,089,563 ballots cast in its 2020 general election, one batch of 50 ballots was counted twice, and that there were “37 instances where a voter may have unlawfully cast multiple ballots”—most likely a spouse’s ballot after the voter had died. Neither lapse affected any election result.

“We found fewer than 100 potentially questionable ballots cast out of 2.1 million,” the report said. “This is the very definition of exceptionally rare.”

When Maricopa County explained how it had accounted for all but 37 out of 2.1 million voters, it noted that the same data sets used to account for virtually every voter were also used by the political parties to get out the vote. Thus, the report’s discussion of these data sets—voter rolls and the list of people who voted—offered a template to debunk voter fraud allegations. This accusation has been a pillar of Trump’s false claims and is a longtime cliché among the far right.

It is significant that this methodology, indeed the full report, was produced under Maricopa County Recorder Stephen Richer, a conservative Republican who has repeatedly said that he had voted for Trump, and was fully endorsed by Maricopa County’s Board of Supervisors, which has a GOP majority and held a special hearing on January 5 to review the findings.

In other words, the report is not just a rebuttal for the Arizona Senate Republican conspiracy-laced post-2020 review. It is a road map for anyone who wants to know how modern elections are run and how to debunk disinformation, including conspiracy theories involving alleged hacking in cyberspace.

“There is not a single accurate claim contained in [Arizona Senate cybersecurity subcontractor] CyFIR’s analysis of Maricopa County’s tabulation equipment and EMS [election management system],” the reportsaid, referring to accusations that counts were altered. “This includes the allegation that county staff intentionally deleted election files and logs, which is not true.”

When you add to Maricopa County’s template the introduction of a second independent scan of every paper ballot in future Florida recounts, what emerges are concrete steps for verifying results coming from Republicans who understand how elections work and can be held accountable.

Of course, these evidence trails only matter if voters or political parties want to know the truth, as opposed to following an ex-president whose political revival is based on lying about elections. However, more moderate Republicans seem to be recognizing that Trump’s stolen election rhetoric is likely to erode their base’s turnout in 2022, as Trump keeps saying that their votes don’t matter.

“You’ve got Republican buy-in,” said Florida’s Sancho, speaking of his GOP-ruled state’s embrace of more transparent and detailed recounts. “And Republicans, more than anyone else, should be concerned about whether their votes were counted as cast and as the voter intended.”

Steven Rosenfeld is the editor and chief correspondent of Voting Booth, a project of the Independent Media Institute. He has reported for National Public Radio, Marketplace, and Christian Science Monitor Radio, as well as a wide range of progressive publications including Salon, AlterNet, The American Prospect, and many others.

Keep reading... Show less

Michael Carvajal

Photo by Tom Williams via Reuters

The search is on for a new director of the federal Bureau of Prisons after Michael Carvajal announced on January 5 that he’s retiring from his appointed post and will leave when the Department of Justice finds his replacement.

The Biden Administration needs to replace Carvajal with a person who knows prisons inside and out: someone who’s been incarcerated before.

When President Joe Biden announced his first round of cabinet picks just weeks after being elected in 2020, then Vice President-elect Kamala Harris said: “When Joe asked me to be his running mate, he told me about his commitment to making sure we selected a cabinet that looks like America – that reflects the very best of our nation.

Keep reading... Show less
{{ }}