Class 17. Election Meddling 101

Understanding the theory behind foreign election interference and why whataboutism doesn't really work.

Now that we’ve covered the history of Russian active measures and the strategic goals behind them, it’s worth turning to one of the history-altering ways they have been deployed recently in the U.S.: election interference. It’s hard to believe that just ten years ago the idea that a foreign adversary would be brazen enough to try to undermine a U.S. presidential election would not have been top of mind for your average American; it would have been even more unthinkable that any such effort could possibly be effective. And yet, here we are.

It would be wrong, however, to approach this tactic as a “new” phenomenon. On this point, I had hoped to get as a speaker for this class David Shimer, author of Rigged: America, Russia, and One Hundred Years of Covert Electoral Interference; I’ve had the pleasure of both moderating a panel that included David and former Secretary of Homeland Security Jeh Johnson for the Yale Jackson School of Global Affairs and co-authoring a piece with him before the 2020 election, and his historical insight is illuminating. Unfortunately, David currently serves on the National Security Council and was not able to do a talk, so I’ve included on the syllabus for this class the next best thing: A C-Span video where he discusses the major themes of his book, which I will also explore in further detail below. Understanding how election interference works in theory — and illustrating these tactics through two historical case studies — provides some context for what, exactly, makes us so vulnerable to this particular form of active measures today.

The Theory

For me, the biggest takeaway from Shimer’s book — which simplified the entire universe of election meddling operations — was understanding that election interference basically takes two main forms:

1. Changing hearts and minds

2. Changing actual votes

Seems basic enough, right? You would think so, but since the 2016 election journalists and commentators have routinely conflated these tactics, which confuses the issue in the public mind and also obscures the needed policy solutions to address each of them. I remember writing an op-ed for The Washington Post in June 2017, right after James Comey had testified to Congress following his firing by President Trump: The headline described Russia’s election interference as “hacking,” which to me a) created the misleading impression that the foreign threat was mainly about technical intrusions and b) underemphasized the insidiousness of Russia’s social media disinformation operations (which were not well understood by the public at the time).

To be sure, changing votes and changing hearts and minds are both “hacks” in a sense: One is a technical hack and the other is a cognitive one. In last week’s guest lecture, former FBI Deputy Director Andrew McCabe described how, in 2016, the FBI was observing activity on both fronts. On the one hand, there was the hack of the DNC server and signs that Russia was probing election infrastructure in various states (the Senate Intelligence Committee ultimately concluded that Russia had targeted infrastructure in all 50 states) — this was something that, for the most part, the existing protocol and capabilities of the FBI could detect and address. On the other hand, the influence operation on social media — the cognitive “hack” designed to change hearts and minds of the voting public — was not only not anticipated or well understood, but was not something that the FBI’s toolbox could really do much about. Even worse, McCabe noted that the FBI failed to appreciate the link between technical “hacks” and cognitive ones: Namely, that stolen information, like the emails taken from the DNC, could be weaponized to shape public perception.

The second part of Shimer’s theory, which is a little more nuanced, is that election interference can be evaluated along two “planes”:

1. Individual change: “whether the intention of an operation is to promote a friendly candidate, defeat an unfriendly candidate, or exhibits no preference”

2. Systemic Change: “whether the intention of an operation is to strengthen, weaken, or not at all impact the internal functions of a democracy”

This analysis is critical to analyzing the threat posed by election interference efforts and comparing them across contexts. For instance, while discussion of Russia’s 2016 election interference has focused on its goals of individual change (helping Trump/hurting Clinton), there hasn’t been enough emphasis on its objective of systemic change, i.e., eroding public faith in the electoral process as a whole. In fact, we tend to forget that at the time, Putin was looking at the same polls we were — and he expected Clinton to win! The fact that Russia had these twin objectives meant that from a systemic change perspective, a Clinton victory could have still been a “win” for Russia, as I wrote with former CIA officers John Sipher and

Alex Finley

for Just Security in 2018.

Similarly, “whataboutism” regarding U.S. covert electoral interference typically fails to acknowledge that these actions were undertaken with a (perhaps misguided) systemic goal of trying to promote democracy abroad, not to destabilize its target countries. This doesn’t mean they were wise or aren’t debatable/controversial as a policy matter, but it does undercut the seeming moral equivalency between the two countries’ operations. More on that in a bit.

For now, let’s unpack exactly how Russia’s efforts in 2016 are an evolution of tried and true tactics. To do that, we’ll turn to two historical examples of electoral interference — one by the U.S., the other by the Soviet Union — that each utilized one of the two basic forms of interference.

Changing Hearts and Minds: The 1948 Italian Election

Shimer writes that in the immediate aftermath of World War II, Italy’s Communist party — the largest in Europe with around two million members — was gaining traction in the country, particularly over the more centrist Christian Democratic Party. The Communist Party’s growing influence was significant, because Italy had only just transitioned from a monarchical system to a democracy — the 1948 election was therefore going to be an important ideological proxy battle for America and the Soviet Union. A Communist and Socialist Front victory, the CIA warned in a classified memorandum, could have dire consequences:

In the Cold War’s emerging battle of ideas, a Front victory would mark ‘the first instance in history of a communist accession to power by popular suffrage and legal procedure’….Were the Front to win, the memorandum continued, civil war might ensue. Matters would only worsen once the Communists took office, because the Soviet Union would establish influence over Italy’s foreign policy, economy, and military bases, from which Moscow could then threaten shipping across the Mediterranean. Ultimately, the CIA portended, Italian democracy would collapse ‘by processes made familiar in Eastern Europe,’ replaced by ‘a fully developed police state under open and exclusive Communist control.’

Believing that the Soviets were fully backing the Communist Party covertly as well as overtly, President Truman authorized his agencies to engage in electoral interference operations in favor of the Christian Democrats. The State and Justice Departments gave official policy warnings to Italians about the future under Communism, like the noncontinuation of economic aid or immigration restrictions, but they “had a common weakness: None targeted voters personally, based on their individual views and biases.”

Enter the Italian-American immigrant community, which, with the overt encouragement of the State Department, began a letter-writing campaign to their relatives and contacts in Italy, encouraging them to vote against the Communists and for the Christian Democrats. The CIA covertly facilitated this efforts by helping to spread the campaign to ethnic enclaves across the U.S., making it hugely successful:

In all, Italian Americans sent an estimated ten million messages to Italy. In one town in New York, more than 40 percent of Italian American residents took part in the campaign. John Ellis, a Republican politician who traveled to Italy in April, found that the letter writing was “heard all over” the country, in what can best be understood as a primitive form of micro-targeting. In 1948, loved ones could influence voters on a personal basis.

The newly-created CIA supplemented the letter writing campaign by also amplifying anti-Communist propaganda being published by the Roman Catholic Church, which had its own reasons for opposing the godless Communists. And, of course, the CIA funneled tons of cash to the Christian Democrats (about $107 million, Shimer estimates, in 2020 dollars).

It worked: Christian Democrats son with 48.5% of the vote, while the Front received only 30%. As Shimer observes, the Italian letter writing campaign was merely an analog form of the same targeting Russia now engages in on social media — the difference is that nowadays the “trusted communities” aren’t necessarily friends and family people know, but other social media accounts they are connected to through curated preferences, shared identities, and partisan loyalties.

The letter writing campaign also highlights the downsides of “hearts and minds” operations: It is difficult to measure outcomes or isolate the precise factors which made the most difference. Was it the letters? The propaganda? The money? “After the election,” Shimer writes, “CIA officers, while triumphant, had no evidence that their tactics, which were part of a much broader operation, had made the difference.” For the same reason, when the U.S. is a target of a psychological operation affecting voter preferences and behavior, it’s hard to prove, quantitatively, that it impacted the vote. That opacity and ambiguity is one way in which Trump and his allies have been able to downplay, and even negate the existence of, Russian interference in 2016:

Changing Votes: The 1972 German Parliamentary Election

So if good old fashioned letter writing was the old “micro targeting,” what did vote tampering look like before the ability to hack into a computer system? Shimer’s description of the 1972 vote of no confidence called by the conservative opposition against chancellor Willy Brandt in the Bundestag, West Germany’s parliament, is one example.

The basic background is thus: Brandt became chancellor with a policy to depart from his predecessors’ adherence to the Hallstein Doctrine, “which called for the nonrecognition of East Germany and any state that recognized East Germany (excluding the Soviet Union).” Instead, Brandt wanted to build relationships with the Soviet Union and its satellite countries. Meanwhile, back at the ranch, the Soviet leader at the time, Leonid Brezhnev, had faced a series of popular revolts across Eastern Europe. While outwardly taking a hard line, Brezhnev privately feared that the Soviet Union sphere of influence was eroding; ‘[w]hat Brezhnev needed was stability, which he hoped to achieve through detente, or deepened cooperation, with the United States and its allies.” Brandt’s leadership offered that chance. If, however, the vote of no confidence succeeded and the conservative opposition, led by Rainer Barzel, won the vote, West Germany would return to the Hallstein Doctrine.

Spoiler alert: The Soviet Union was able to rig the parliamentary vote in its favor. How it did so highlights two important features that are salient today. The first is that the Soviet Union was not savvy in the ways of electoral politics because, well, it was a totalitarian state which didn’t hold elections. Shimer writes:

Covert electoral interference operations often reflect the character of the states that execute them. In America, presidential elections occur every four years, so CIA officers instinctively knew how to export traditional campaign techniques to faraway places. William Colby’s work in Italy, and the 1964 operation in Chile, benefited from this familiarity with competitive campaigning.

The Soviet Union and its Eastern European satellites had a far different domestic experience. Competitive elections were not held, and secret police terrorized citizens. The KGB and its peer agencies like the Stasi, East Germany’s security service, proved most adept at exporting the tool kit of a surveillance state: bribery, blackmail, and psychological warfare. At opportune moments, Eastern operatives could use these weapons to manipulate foreign electoral processes. Such was the case in April 1972.

The second important feature is that affecting actual vote tallies is tricky because it’s difficult to do at scale, which is often necessary to affect an electoral outcome. However, it’s easier where margins are thin, and just a few votes can make a big difference.

Both of these features came together in the 1972 Bundestag vote because the Soviet Union was able to figure out that flipping just two votes could keep the conservative opposition from taking control. To get those two votes, the Stasi targeted two men who were vulnerable for recruitment using both blackmail and bribery:

Both were conservative members of the Bundestag who drank heavily, chased women, and amassed debts. Both were thus ripe for recruitment. ‘Irresponsible habits are all classic signals that someone can be recruited, and money was always the go-to tool for people in debt,’ said Donald Gregg, a CIA operations officer for almost the entirety of the Cold War.

Did I hear someone mention a $15,000 ostrich jacket? You know, like the the one owned by Paul Manafort, the former political consultant for the pro-Russia Ukrainian political party who was $10 million in debt to Russian oligarch Oleg Deripaska and who randomly showed up on the Trump campaign’s doorstep offering to work for…[checks notes]…free?

I’m mentioning this because the 1972 parallel to Russia’s efforts in 2016 comes through Manafort’s role in the Trump campaign. Specifically, as my colleague John Sipher has highlighted when speaking to my classes, Russia is no savvier about the ins and outs of electoral campaigns than the KGB was during the Cold War. (OK, maybe it’s a little savvier…but not much.) That is, there is a lot that Russia can figure out from afar, but to get very granular about electoral margins and where interference might make a difference, it needs human sources. On the ground. Someone like, say, Paul Manafort. And it also needs information it wouldn’t know or be able to get on its own. Like, say, polling data.

Sadly, the significance of Manafort in Russia’s 2016 electoral interference operation has largely been overlooked or forgotten. But like the West German operation in 1972, Manafort was a perfect target for recruitment (I mean he had likely long been recruited by the time he came on to the Trump campaign), and he clearly was loyal till the end — my dude was still ghost writing op-eds with Russian intelligence while he was awaiting trial!

Most importantly, though, the bipartisan Senate Intelligence Committee report found that Manafort was passing polling data to Russian intelligence operative Konstantin Klimnik. The salience of the polling data becomes evidence if you view it against the backdrop of operations like the 1972 vote in West Germany: Russia wanted to know where small efforts could make a big difference.

As McCabe and I discussed last week, the U.S. has a huge buffer against vote tampering by virtue of the fact that voting in this country is decentralized: To try to change votes at scale, hackers would need to get into 50 different state voting systems, and into local districts beyond that. It would be almost impossible. HOWEVER. If you had data showing the handful of places where differences of a few thousand, or even just hundreds, of votes could swing the district, and thereby swing the entire state…well, then you’ve got yourself a more doable plan, don’t you. The Senate Intelligence Committee determined that there is no evidence that Russia actually changed any votes. But it certainly seems like they’d have known where to do it if they wanted to.

Post Script: Manafort was pardoned before President Trump left office in 2020. He currently lives as a free man in Florida.

---

So looking back on both the U.S. and Soviet electoral interference operations during the Cold War, how do they fare along Shimer’s second framework, regarding individual versus systemic change? Well, it’s clear that in both operations, each country favored a particular party/candidate. They seem pretty equal there.

And, interestingly, on the systemic front, I would not say these examples are that far apart, either. Of course for the U.S., the 1948 Italian election would have been one of the first instances of employing the strategy of containment, i.e., preventing the spread of Communism. That in itself doesn’t make it right — the policy of containment, after all, also justified the CIA’s instigation of violent coups in Latin America, and the U.S. getting into bed with dictators in that region and in Africa. But as far as election interference goes, the letter writing campaign seems about as benign as an overt-covert government operation could be. And the U.S., whether for moral or self-interested reasons, was motivated by a desire to help preserve Italy’s nascent democracy.

Likewise, it’s hard to really fault the Soviet Union for their interference in 1972. After all, Brandt had been awarded the Nobel Peace Prize in 1971 for his efforts to achieve reconciliation between West Germany and Eastern Europe. Shimer observes that “in the long term, the reunification of Germany, precipitated by deepened ties between East and West Germany, would have unfolded entirely differently, as would the trajectory of Soviet history….The very arc of the Cold War…might have been transformed — if not for covert action.”

What’s different now, and what makes these more dangerous, is that not only is the U.S. not engaging in these types of operations anymore, Russia has accelerated them… and for less lofty purposes than Brezhnev in 1972. From the American perspective, covert electoral interference is more easily detected in the digital age — and the blowback on the U.S. much greater than it is for rogue countries who engage in it. Instead of covert operations, Shimer observes, the U.S. now engages in overt influence, through public endorsements and “democracy promotion” through economic and expert assistance.

At the same time that the U.S. has passed on election interference as a policy tool, Russia has doubled down. The demise of an ideological competition between superpowers has given way, in the post Cold War era, to a Hunger Games mentality where weakening adversaries and fracturing alliances is Russia’s best bet in the world order. “As Washington moved away from covert electoral interference,” Shimer writes, “Moscow rediscovered and enhanced this weapon. It has done so under the leadership of one man: Vladimir Putin.”

We’ll drill down on his efforts in 2016 and beyond in the coming weeks.

Audio Here:

Discussion Questions:

1. U.S. electoral interference in the 1948 Italian election was largely overt — to the point where, Shimer notes, “Italy’s Communist leaders swiftly denounced the letter writing as foreign interference.” What are the costs/benefits of covert vs. overt interference? Does transparency make a difference in its acceptability as a foreign policy tactic? Why or why not? (see also Class 2 for more on this the overt/covert distinction)

2. How does Russia’s lack of ideological constraints (compared to the Soviet Union during the Cold War) give it more flexibility to affect both individual and systemic change through electoral interference?

3. Hacks of technical infrastructure can be fixed by things like software patches, or other security measures. Is it even possible to create a “patch” for the cognitive hacks which target voters’ hearts and minds?