I have several times criticized Western observers who naively assume that pro-government statements by people who live under repressive regimes necessarily reflect their true opinions (see here and here). Even if they actually hate the government, such people have obvious incentives to pretend otherwise in order to avoid punishment. At the same time, it is also a mistake to conclude that all citizens of repressive regimes are closet dissidents. Often, many of them genuinely support the regime and its ideology, if only because indoctrination can be effective, especially in an environment where opposing views are suppressed.
How can we find out the true beliefs of people living under repressive governments? There are several strategies, each of which has its flaws. For example, interviewing emigrants can be very useful; but they are unlikely to be a fully representative sample of the society they come from.
In Foreign Policy, political scientists Angela Hawken and Matt Leighty describe an interesting recent effort to overcome this problem: “guerrilla polling”:
We’ve been intimately involved in the effort to conduct public-opinion surveys in countries controlled by authoritarian regimes.In January, we completed the analysis of an in-person survey of 1,046 adults living in Syria. The poll, conducted by the Democracy Council, a California-based NGO, was the first face-to-face survey collected by an unsanctioned organization on the ground in Syria.
Democracy Council had to overcome several hurdles to pull off the survey. First, it had to find 60 qualified interviewers in a country where such data collection is illegal and then train them from scratch. The interviewers were recruited by word of mouth, and each was put through an extensive background check to make sure that he or she had no association with the Syrian government.
New technology greatly assisted in the training process. Democracy Council prepared its field staff using Skype, the well-known Internet calling service, which now allows videoconferencing. Skype provided several advantages: The calls are encrypted, so any messages intercepted by Syrian security services would be unintelligible, and videoconferencing avoided the need for any in-person gathering, which might have attracted the attention of the authorities.
The resulting survey “reflected poorly on the Syrian government”:
The poll found that a majority of Syrians believe that their political and economic situation is poor and worse than it was five years ago. They consider the government to be corrupt and have little faith in its ability to confront the country’s problems. A substantial majority believes the state of emergency, which has been in place since 1963 and used to justify violations of civil liberties, should be lifted, and a majority reported that it would leave Syria if it had the opportunity to do so.
Hawken and Leighty discuss some methodological limitations of this kind of research. I would add that I do not share their confidence that it can be easily replicated in societies that are more repressive than Syria, such as North Korea (a country they discuss as a site for future research). Syria has a repressive government. But enough private individuals have access to the internet and Skype to make Democracy Council’s project feasible. Moreover, the punishment for dissent in Syria is probably less severe.
Hawken and Leighty also point out that the Syrian government is now aware of the study and will surely try to prevent a recurrence. Other authoritarian regimes will likely do the same. Future efforts at guerrilla polling will no longer enjoy the advantage of surprise that Hawken and Leighty admit played a key role in the success of their Syrian study.
It is also important not to overstate the significance of their findings. The fact that a repressive regime is unpopular doesn’t necessarily mean that it is ripe for collapse. Repression can often keep a despised government in power for long periods of time, because of the difficulty of organizing effective resistance. At the same time, such regimes do risk collapse if they start to liberalize their economic or political systems, as happened in the case of the USSR and Eastern Europe in 1989-91. If the threat of repression is lifted or even significantly eased, an unpopular regime may fall apart faster than anyone expected.
Despite these caveats, “guerrilla polling” is a useful methodological innovation, one that effectively takes advantage of modern technology that is difficult for government to fully control. Hopefully, future guerrilla pollsters will find ways to stay a step ahead of the secret police.