Here’s the bad news: The erosion of democracy continues to advance around the world, according to the 2023 report from V-Dem, the independent research institute that seeks to measure the qualities of supposed democratic nations. A record 42 countries experienced democratic setbacks or the process V-Dem calls “autocratization,” compared  to 22 last year and just 13 in 2022. “The level of democracy enjoyed by the average world citizen in 2022 is back to 1986 levels,” said Staffan Lindberg, director of the V-Dem Institute, in a press announcement. “This means that 72 percent of the world’s population, 5.7 billion people, live under authoritarian rule.” For the first time since 1995, Lindberg said, there are more “closed autocracies” (aka dictatorships) than genuine liberal democracies.

But that’s not the whole story: V-Dem also detects a current of “defiance” over the past year. For the first time since the “third wave” of autocratization began, a number of countries have significantly reversed the trend, suggesting a clearer picture of how democracy can bounce back — which has potential application in the United States. Only two such countries were listed last year. Now there are eight:  Bolivia, Moldova, Ecuador, the Maldives, North Macedonia, Slovenia, South Korea and Zambia.

“The fact that eight democracies that were in a period of autocratization have stopped that process and ‘bounced back’ is uplifting news for democracy,” Lindberg said. “The countries that have succeeded in doing so have brought about a pro-democracy mobilization, they have re-established an objective judicial system, deposed authoritarian leaders, introduced free and fair elections, worked to reduce corruption and rejuvenated civil society.”

Although all eight on the list are relatively small counties, their experience appears relevant to that of the U.S. and Brazil — also noted in the report — where the decay of democracy has at least been stalled, and efforts at democratic renewal are underway. The report identifies five elements that unite most of the eight cases. None is a silver bullet, but all have played an important role in helping to re-establish democracy. They include large-scale popular mobilization against an autocratic incumbent, judicial reverses of an attempted executive takeover, a unified opposition and civil society, critical elections or other key events and the support of international democratic institutions.

Much of that isn’t news. V-Dem’s 2020 report, after all, was titled “Autocratization Surges — Resistance Grows.” It reported significant mobilization in countries like Bolivia, Poland and Malawi, and mass protests in 34 autocracies, including Algeria, Hong Kong and Sudan. But it’s only now that we have enough solid examples to begin providing some guidance for the renewal of democracy.

With a clear pattern in eight smaller countries, and distinct echoes in two of the world’s largest democracies, V-Dem’s latest report may point the way forward — but only if you believe what it’s saying. I discussed right-wing criticism in my article about V-Dem’s 2021 report, but now there’s criticism from serious academics, who question whether there’s really been any significant backsliding at all. A working paper on that subject by Andrew Little and  Anne Meng, released in January, argues that the evidence simply isn’t there. 

When V-Dem’s report was issued and that paper’s findings were cited again, Lindberg retweeted two threads that stood out in response. I’ll return to those arguments below, and let Lindberg himself have the last word. But first let’s dive into what the report itself has to say, as Lindberg explained in the following interview, which has been edited for clarity and length. 

I saw three distinct processes at work in this report. First is the process of autocratization that you’ve documented for years. Last year your report [Salon story here] found the process becoming more old-fashioned with five military coups, for example, along with the interconnected rise of polarization and misinformation and the erosion of formal democracy. All that was a sign of things getting worse. This year, on the flipside, your report highlights widening defiance in the face of autocratization.  How do all of those fit together? 

What we see in this year’s report is both a continuation of last year and maybe something new, the defiance aspect. First of all, the wave of authorization is escalating, up from 33 countries last year — that was a record — to 42. So in that sense autocratization is deepening and broadening. In this year’s report we also go further into the analysis of the way polarization and the spread of disinformation is connected to this and furthering it. 

On the other hand, we wanted to highlight these eight cases, over the past 20 years, that have bounced back and defied this wave, in the sense that they were democracies at some point and then dipped down, were engulfed in autocratization and not only stalled it but also turned it around and became democracies again. Or in some cases democracies didn’t fully break down, but they turned that autocratization process around.

That’s pretty significant, because before we only knew about two cases in the last 20 years, South Korea and and Ecuador. We think that’s important to highlight, even if it’s only eight cases, we think we can learn something from them about what can be done. 

You’ve said there are five elements that unite most of the cases. Say a little bit about each of those, starting with large-scale popular mobilization against the incumbent.

It is an element that we see at least seven of these eight cases. In South Korea, as many as  2.3 million people were out in the streets at the same time. This also tallies with broader literature in political science. My colleague and friend Andreas Schedle, in one of his publications, called it “magic protest,” all the color revolutions in former Soviet republics and Eastern Europe. But also, if you look at earlier periods in Africa in the ’90s, where large-scale protests have played a role in bringing down authoritarian regimes or stopping autocratization. It’s not a silver bullet. If regimes want to clamp down on protest, put enough people in jail and are willing to kill enough people, then it’s not going to help. Look at Hong Kong and other cases where you have mass mobilization and yet it hasn’t been enough. But it is something that runs through seven of these eight cases. 

What about the second element: the judiciary reversing executive takeovers? 

That’s another element that we see is really significant in half of these cases. We think there are signs in some of the others as well, but in four cases it’s really prominent: The judiciary refused to back down, refused to bend over when the executive sought to expand its power and control the judiciary. They played critical roles in turning things around and, if necessary, such as in South Korea, to throw the previous incumbent in jail. 

Want a daily wrap-up of all the news and commentary Salon has to offer? Subscribe to our morning newsletter, Crash Course.

If we take a little bit of a broader perspective. I think we see this in other cases, like in your own country. Think about the last presidential election: It was really critical that the judicial system across the country stood up against these false claims of election fraud. You saw that in Brazil as well. Although Bolsonaro really did his best to try and intimidate and take over the judiciary, the Supreme Court and other parts of the justice system refused to back down. God knows what would have happened in Brazil if they had. 

What about the third element, the unified opposition coalescing civil society?

For the American audience, you have to remember that many of these cases are parliamentary systems with multi-party systems. So you often have the problem of an opposition that consist of several parties, and they all want to be the ones coming into power if it’s possible to unseat an incumbent. In a period of autocratization, we know the saying, “divide and rule.” But in seven of these cases, it really jumped out at us how the opposition didn’t necessarily form a coalition in a formal sense, but unified in their actions and pushed back against the incumbent, often then coalescing with civil society to help this large-scale mobilization, which was the first element. They also stayed peaceful. That strategy is also important. Sort of a unified, Gandhi-type opposition, coalescing with civil society. That seems to be important. 

And what about the fourth element: critical elections and other key events that bring an alteration in power?

Maybe this is not very surprising, but it’s still important — in all these cases, elections or events like the end of a term limit were used by a unified opposition, by civil society and the judiciary as a critical turning point, a watershed moment where the entire society can be mobilized. It’s also the point where you can use the formal opposition to unseat an autocratizing leader. So that seemed to us to be important to point out, that those those moments are critical.

Finally, what about the fifth element, international democracy support and protection?

There’s been much discussion in both the academic literature and also in the policy community about what international democracy support — or in times of autocratization, then maybe international democracy protection — can really do in the face of China and Russia and others internationally pushing for something else than democracy. But in at least five of the eight cases it seems that this international support actually made a difference. Maybe not decisive on its own, or definitely not. But in combination with the other pressures, they played an important role and helped to turn things around. I think that’s good news for the international community: There is actually a point with engaging, investing taxpayer money and putting your diplomatic corps to work. You can make a difference for democracy internationally. 

These eight were relatively small countries, but something similar, if not measurably the same has happened in the U.S. and Brazil, as you’ve mentioned. I’d like to ask you to focus on them a bit more — what similarities can we see with those patterns?

I mentioned the judiciary already. Both Brazil and the United States are countries where the process of autocratization has stalled and not turned around, but the curve has flattened out. I think that’s also been somewhat because of popular mobilization in the United States and definitely in Brazil. Although in Brazil it was — and maybe in this United States too — you can say it was popular mobilization on both sides, both the autocratizing set of leaders and parties and the opposition mobilized. Definitely the elections were critical, with throwing out Bolsonaro and throwing out Trump. Now the others I’m less certain about. In Brazil the the opposition was unified behind Lula to a large degree. In the United States there’s not really much of a choice, because you have Democrat or Republican and nothing else.

Yes, but the fact that extreme election deniers in 2022 did significantly worse than conventional Republicans seems to indicate that there was some significant shift there, conventional Republicans breaking with the MAGA base. So that is somewhat similar in spirit.  

Yes, similar in spirit, and there’s somewhat of a parallel at least to South Korea and Ecuador, where the autocratizing parties behind the leader started to see internal splits. Some of them were not willing to go along, and that was important in the downfall of both leaders and eventually losing elections and turning things around. What hasn’t really happened in the United States with the GOP that happened in both Ecuador and South Korea is that the [defeated] ruling parties have really reformed. 

Both in South Korea and Ecuador they shifted out the party leadership, got some conservatives who were still right-wing but committed democrats into the leadership and reformed the party program. We haven’t seen that with the GOP. Instead we see the democratic opposition within the GOP are being thrown out and marginalized, especially in the House, Maybe less so in the Senate, but we really don’t know what’s going to happen in the House. 

We haven’t seen what’s happening in Brazil, finally, but immediately after the election a lot of the leadership of Bolsanaro’s party turned against him, in the sense of immediately saying, “We’re going to accept the election, there’s going to be a peaceful turnover” and so on. There was a clear signaling that they were willing to make the shift back to being fully democratic. So I’m a little bit more hopeful about Brazil than about the United States.

Is autocratization real?

Lindberg outlines a number of hopeful signals, but there’s a challenge on another front: whether people will actually listen, and believe what V-Dem is saying?  More fundamentally, should they believe it? As noted above, when the report came out, the working paper by Little and Meng was cited online to question its basic foundations. Perhaps their most potent claim was that there had been no significant change in electoral turnover, arguably the most salient significant indicator of democracy. 

Lindberg retweeted two threads that pushed back. The first, from Michael Miller, shared his more nuanced measure of what he called “clean” turnovers: “ignoring those where the incumbent was ousted for reasons other than the vote (protests, coups) or where there were mass protests alleging election fraud or w/ violence.” His data suggest that the decline is real. The second, from Carl Henrik Knutsen, took on the broader issue of complications in the division between supposedly “subjective” and “objective” data, which is central to Little and Meng’s argument.

It’s important to note that Knutsen is one of five principal investigators on the V-Dem team — but is not part of V-Dem Institute, which produces the annual report, and was not among its authors. He told me that “there is no consensus within the wider V-Dem team on exactly which measures (of the many V-Dem indicators and indices) are best to use … for the specific purpose of discussing global trends in democracy, and there is no consensus in the wider V-Dem team on exactly how to best describe current trajectories in global democracy levels.”

Knutsen said that measures of democracy that claim to be totally objective, such as those used by Little and Meng, “are affiliated with considerable validity problems,” which is a polite way of saying he thinks they’re useless:

People who think that “objective” measures of democracy are always much better than measures using some kind of expert evaluation/subjective judgments are, at best, presenting a very simplified story and, at worst, misled. Objective measures can have lots of measurement errors and be biased (they often fail to capture relevant aspects of democracy). In my tweet thread, I mentioned the fictive measure of “democratic is in the country name”: Completely objective measure, and definitely biased (as autocratic states are the ones typically including this in the name, examples being North Korea or East Germany).

When I asked Lindberg for his response to Little and Meng’s paper he struck some of the same notes, couched in polite, diplomatic language:

I’d like to say first that we’ve always welcomed critique and I’ve always said that in due time something better than V-Dem will come along, in terms of measuring democracy. We think we’ve improved on the previous measures, and added sort of pluralist and other measures of democracy and others should come along. It’s good we hear critique. I think that some of the language used by Meng and Little is misleading. Take the terms “objective” and “subjective” measures. What they call “objective,” I will call “factual” or “observable” — things you can see in the constitution or you can see in the electoral system, you can see turnout and so on. 

That doesn’t mean they are objective in the sense of unbiased or truthful, in two ways. Take something like voter turnout or whether you have term limits. Well, the Chinese Communist Party until recently had a term-limit policy: You couldn’t be in power for more than 10 years. Did that make it democratic? I don’t think so. 

Turnout figures between countries vary by electoral system, China has the highest turnout in the world, and we know that things like whether elections are held on a holiday or not affect turnout. So factual measures have lots of biases in them, it just comes from other sources. So portraying them as objective and therefore unbiased, and telling the truth is not correct, or is misleading. 

Lindberg further suggested that using these “observable measures” to determine whether or not a country has democracy had the effect of “restricting what democracy is to those things that can be seen or measured.” That’s also misleading, he contends: “So many things that have to do with whether you have democracy or not, or more democracy or less democracy, are not seen in those measures.”

He cited the example of the written constitutions of Belarus and the United States, which both state that the legislative and judicial branches have the power to constrain the executive branch and hold it accountable. But in only one of those countries is that power still even remotely meaningful, a distinction that would be lost in Little and Meng’s “objective measures.”

Another example would be the crucial importance of media freedom, meaning that “journalists can write critically about the government without being harassed or thrown into jail.” Authoritarian governments will lie about this, first of all, as Lindberg notes, but there’s also a more fundamental problem: 

The number of journalists being harassed or thrown into jail is zero in a country like mine, in Sweden, most of the time. But it’s also zero today in Russia or North Korea, at least by any official measures, because nobody dares to stick out their necks anymore. So that factual observable measure would be misleading in terms of how much democracy you have.

What you really want to know, which is what we measure with experts, then, is: If a journalist criticizes the government, what’s the probability that that journalist would be harassed or thrown into jail? Because that probability is close to 100% in North Korea and it’s close to zero in Sweden. It used to be close to zero in the United States, I think it is today, again. But you see my point — the pretense that we can make sure how democratic a country is by purely observable facts, I think it’s misleading and false. 

Lindberg added that quasi-dictatorial leaders like Viktor Orbán in Hungary and Recep Tayyip Erdoğan in Turkey have made an effort to “mimic democracy as much as they can by doing the dirty work behind the scenes,” such as by repeatedly changing the constitution or holding elections in which the opposition may win a few token victories but has no chance of winning power.

I suggested to him that one way of understanding Little and Meng’s report is that it reinforces V-Dem’s point about the nature of “third-wave” autocratization, which works by maintaining the forms of democracy as much as possible, so that a country may appear democratic on paper but in practice absolutely is not. 

“Yeah, you said it better than I did,” Lindberg responded, suggesting that this dispute speaks to the reason why he and others launched V-Dem nearly a decade ago. Everyone studying democratization in the academic world, he said, was “frustrated with all these measures of ‘observables,’ because they were not enough. They didn’t show enough of what democracy is, and were misleading in different ways. So we started V-Dem to measure democracy more fully, if you like, and with the unobservable areas that are really critical. So this paper sort of shows why V-Dem is needed.”