{"id":"election-legitimacy-observers-autocracies-04075","slug":"election-legitimacy-observers-autocracies-04075","title":"Who Decides If an Election Is Real?","description":"There's no office anywhere that stamps an election genuine, yet most people assume one exists. This episode digs into the numbers behind sham elections — V-Dem counts 92 autocracies, 57 of them \"electoral autocracies\" holding multiparty votes that aren't real contests — and asks why the three big democracy indices disagree so often. It traces the actual minimum bar in international law, from Article 21 of the Universal Declaration to the Copenhagen Document's famous escape hatch, and follows what happened when Venezuela's 2024 tally sheets told a different story than the official result. Then it flips the intuition: international election observation has become a badge regimes can buy, and the cleanest democracies are often the ones that never get a full mission.","excerpt":"57 countries hold multiparty elections that aren't real contests. So who certifies a vote — and why do clean democracies skip full scrutiny?","pubDate":"2026-10-22T22:25:49.597Z","tags":["elections","human-rights","international-law","comparative-policy","constitutional-law"],"heroImage":"https://files.myweirdprompts.com/covers/election-legitimacy-observers-autocracies-04075.png","podcastAudioUrl":"https://episodes.myweirdprompts.com/audio/election-legitimacy-observers-autocracies-04075.m4a","podcastDuration":"21:47","episodeNumber":6210,"transcript":"Corn: There is no office anywhere that stamps an election genuine, and the strange part is that most people assume there is one.\nHerman: I think that assumption is where Daniel's question starts. He wants to know how many countries run elections that aren't real contests, whether there's a globally accepted minimum bar for calling a vote legitimate, and whether the clean elections are the ones that let international monitors in.\nCorn: The tidy version of the world has the good countries inviting observers and the bad ones slamming the door.\nHerman: It's a lovely tidy version. The United States, France and Switzerland get what the OSCE calls assessment missions, a small team that studies the process from a desk, not hundreds of people in polling stations. And a fair number of autocracies roll out the red carpet. So the observer list is a poor sorting hat.\nCorn: Then let's find out what does the sorting. Start with the headcount.\nHerman: Nobody keeps one official number, so you get several unofficial ones that disagree. The V-Dem report that covers last year counts ninety-two autocracies and eighty-seven democracies. Autocracies went up by one, democracies down by one. The important line inside that is that fifty-seven of the ninety-two are electoral autocracies, meaning they hold multiparty elections that aren't real contests, and thirty-five are closed, with no meaningful contest at all.\nCorn: Fifty-seven. So if Daniel wants the sham-election count, that's the nearest thing to it.\nHerman: It's the nearest, yes, and it's also the most common regime type on earth. About forty-six percent of humanity lives in one, which is something like three point eight billion people.\nCorn: Almost half the planet votes in elections it can't win. Turnout must be dreadful.\nHerman: Turnout is often spectacular, which we'll get to. Freedom House counts differently. Its latest edition has eighty-eight countries Free, forty-eight Partly Free and fifty-nine Not Free, with a few territories on top. And the Electoral Integrity Project does something else again, scoring individual elections from zero to a hundred using about forty experts answering forty-nine questions. Equatorial Guinea came in at sixteen for its twenty twenty-two vote.\nCorn: Why does the same country land in three different places?\nHerman: Because they're measuring different things. V-Dem is sorting regimes, so it asks whether the top office is genuinely up for grabs. Freedom House scores rights, so a country with decent election day mechanics but jailed journalists loses points. The Integrity Project grades the process of one contest. A country can be a mediocre democracy by one measure and a decent autocracy by another.\nCorn: I'd also note that all three run on expert judgement. Somebody in a university decides what a seven out of ten looks like.\nHerman: That's fair, and you can see it in the headline V-Dem gave this year, that the United States dropped from liberal democracy to electoral democracy for the first time in over fifty years. You can agree or not, but nobody counted anything there. A panel of country experts coded it.\nCorn: And a panel of country experts has opinions about American politics, who would have guessed.\nHerman: Still, it isn't arbitrary. They're coding hundreds of indicators and averaging many raters. And there are two edge cases the counts handle awkwardly. Eritrea has held no national elections since independence in nineteen ninety-three, despite a constitution that promises an assembly. And the Vatican picks its leader by conclave.\nCorn: I'd class the conclave as a very exclusive electorate with excellent attendance.\nHerman: Nobody skips it, no. But notice that the headcount hides a prior question, which is what counts as a sham in the first place. I'd split it three ways. There's the vote with no contest at all, the single approved candidate. There's the contest where the field was rigged beforehand, with rivals disqualified or jailed and the television owned by the government. And there's the contest where the counting is honest on the day but the playing field was tilted for years.\nCorn: The third one sounds like half the world's politics if you're in a bad mood.\nHerman: That's the trouble. The first kind is easy to count, the second is the heart of the electoral autocracy category, and the third is where the indices disagree most, because it's a matter of degree. Hungary and Turkey get argued about for exactly this reason.\nCorn: Hold on. Hungary has a better record than a lot of places, so it's a stretch to put it next to Turkmenistan.\nHerman: It is, and I'm only saying that's the zone where raters split, not that the countries are alike. The second part of Daniel's question, the minimum bar. There are real texts. Article twenty-one of the Universal Declaration of Human Rights says the will of the people shall be the basis of the authority of government, expressed in periodic and genuine elections by universal and equal suffrage.\nCorn: Genuine. Which is the word everyone then argues about for sixty years.\nHerman: Yes. In nineteen ninety the Copenhagen Document in Europe went further, committing states to invite observers from other states and from appropriate private organisations, to the extent permitted by law. I'd want to check the exact wording before quoting it, but that phrase, to the extent permitted by law, is the escape hatch, and I'm fairly sure about it. In two thousand five a Declaration of Principles for International Election Observation was launched at the United Nations with twenty-three original endorsers, and it's now backed by more than fifty organisations. The Venice Commission has a code of good practice from two thousand two.\nCorn: Three documents, and not one of them carries a signature line for a verdict. Who checks it?\nHerman: Nobody, and this is my reading of those documents, not a line anyone wrote. They provide for invitation and observation. They never provide for certification. No body is entitled to declare an election legitimate. Observers publish a statement, and the statement is evidence.\nCorn: A reference letter, not a diploma.\nHerman: Then legitimacy ends up as recognition, government by government. The cleanest recent case is Venezuela in twenty twenty-four. The electoral council, the CNE, announced that Maduro had won. An expert panel from the United Nations, invited by the council itself, found that the results management fell short of basic transparency and integrity. The council postponed and then cancelled three post-election audits. The vote itself was largely peaceful and logistically well run, mind you.\nCorn: So the polling stations were fine and the arithmetic wasn't.\nHerman: The opposition had collected the tally sheets from the machines, and the panel confirmed their reliability. The Secretary-General's office said that over eighty percent of the sheets the opposition published showed a different outcome from the official one. Then you watch the world sort itself out. Argentina, Uruguay, Ecuador, Costa Rica, Panama and Peru recognised the opposition candidate, González, early. Colombia, Brazil and Mexico declined and called for an institutional resolution. The United States recognised him in the middle of November. The European Union couldn't reach a consensus.\nCorn: Same evidence, three different answers. Which one was right?\nHerman: I'd like your read on this one, Corn, because it's your kind of question. When the evidence is that clear and still half the continent won't say it, is that diplomacy or is that politics wearing diplomacy's coat?\nCorn: Politics, mostly. Brazil and Mexico had other relationships to protect with Caracas, and nobody inside a foreign ministry wants to be the first to say a leftist government lied. But I'd give them one point. Recognising the loser as president is a much bigger act than saying the vote was bad.\nHerman: That's a distinction I hadn't made. I should note that these recognition lists come from the months right after the vote, so I can't tell you who stands where now.\nCorn: Here's what bugs me. Can an election be called free if the observers only get invited to the ones already decided?\nIvan: Nightjar. Three nine, four one. Two zero, seven seven. Three nine, four one. The marrow ripens only after the observers have gone home. The marrow is not for the marrow's pot. The marrow counts its own seeds twice. Fold the trellis. Bluebell.\nHerman: Message received.\nCorn: What on earth was that?\nHerman: The invitation question is where the data runs against the intuition. The scholar Judith Kelley tracked this for decades, and major observer missions in countries that weren't established democracies went from none in nineteen seventy-five to more than eighty-five by two thousand four. Observation became fashionable, and it became a badge, so the regimes learned to buy it.\nCorn: You're saying inviting a monitor is something a government can do, like inviting a caterer.\nHerman: And a caterer who knows who is paying. There's a whole category the academics call zombie monitors, observers who show up to bless a vote. Russia held referendums in the occupied parts of Ukraine in twenty twenty-two and presented foreign observers, and people have documented dozens of individuals across Donetsk, Kherson and elsewhere. Yet Russia didn't invite the OSCE's main election office to its own Duma vote in twenty twenty-one or to the presidential election in twenty twenty-four. The OSCE parliamentary assembly called that deeply regrettable.\nCorn: So they're happy to have observers who agree in advance and not the ones who might read the tally sheets.\nHerman: That's the split. The what-makes-an-election-free-and-fair episode we did went through the checklist, the Carter Center's rule that you don't go without full access, and the whole critique of observation as whitewash. Today's addition is that a missing invitation is a signal and a present one is not. Moscow also shows you can have both in the same year.\nCorn: And the other direction. The clean democracies are the ones who don't get full missions.\nHerman: Take the United States. The OSCE has observed American elections since two thousand two, and in twenty twenty it sent a hundred and two observers from thirty-nine countries. But access at polling places is set by state law. In twenty twelve the Texas attorney general said observers within a hundred feet of a polling place could be prosecuted, and Iowa's limit was three hundred feet. The ODIHR director called that threat unacceptable.\nCorn: Would you want a foreign delegation standing next to you in your polling booth, Herman? Be honest.\nHerman: Beside the booth, not in it, and yes, I'd like it. I'm an archery person, I like a clear line of sight and a witness. But I understand why a state official in Iowa doesn't enjoy someone from Warsaw peering at his sign-in table.\nCorn: Mixed feelings. A sloth would vote by mail and send the observer a leaf.\nHerman: Unobserved leaf noted. Back to Texas.\nCorn: Did anybody get arrested?\nHerman: No arrests were reported on election day, which is a pleasant thing to be able to say. How many states restrict access depends heavily on who's counting. I've seen eighteen in one fact-check and quite a different list in a Russian statement quoting the same report, so I'd leave that number alone.\nCorn: A Russian complaint about American election access. I'd say the source is a bit like a burglar reviewing your locks.\nHerman: You'd be right that it's propaganda in the Kremlin's case, but the underlying fact about state law is real. Britain gets assessment missions in twenty seventeen, twenty nineteen and twenty twenty-four. France in twenty twenty-four got a needs assessment. The OSCE office itself says assessment missions to the US, France or Switzerland have become routine.\nCorn: Why would a rich democracy accept less scrutiny than a poorer country with a shakier record?\nHerman: Partly the observer can't add much. The institutions, the courts, the press are better than any visitor. Partly there's pride. But there's also a gap in the obligation. The commitment is to invite to the extent permitted by law, so a country with federal states and strong local law can invite and still have its own polling places closed to foreigners. Strong democracies are protected by the loophole as much as the weak ones are.\nCorn: So the invitation is no proof of virtue and the lack of one is no proof of vice, and you've spent ten minutes telling me that.\nHerman: I have. What do you do with that? If the front door tells you nothing, what do experts use instead?\nCorn: I'd look at whether anyone ever loses. It's the first thing I'd check and the last thing a ruler will admit.\nHerman: That's the first one. Adam Przeworski's line, democracy is a system in which parties lose elections. Samuel Huntington's version is the two-turnover test, where a democracy is consolidated after power has changed hands peacefully twice. It's crude, and it fails new democracies by design, but it travels well. No regime that never loses has ever been persuasively called free.\nCorn: A hundred percent win rate. Excellent at sport, suspicious at voting.\nHerman: There's a catch with turnover, though. A ruler can lose and the system still be rigged, if the loss was allowed, and a stable democracy can go decades without a handover because one side keeps winning fairly. So it's a test with false alarms in both directions.\nCorn: Japan went nearly all of the second half of the twentieth century with the same party in charge, and nobody called it a sham.\nHerman: Close enough, the Liberal Democrats held power almost without a break for decades, and it was still real competition in the sense that they could have lost. The test is whether losing was possible, which is annoying because you can only prove it after somebody loses. That's why the other two tests exist.\nCorn: So turnover is the receipt, and you want the receipt before you pay.\nHerman: The second is arithmetic. In May twenty twenty-one Bashar al-Assad won a fourth term in Syria with ninety-five point one percent of the vote, on a reported turnout around seventy-eight percent, though rebel and Kurdish-held areas didn't vote. Turkmenistan's presidential vote in twenty twenty-two gave a turnout of ninety-seven percent. North Korea's assembly vote in March reported ninety-nine point nine percent turnout with one pre-approved candidate per seat.\nCorn: Ninety-nine point nine. Who is the point one? I'd like to meet him.\nHerman: Possibly in hospital. Real elections are noisy. Turnout and margins vary by region, by town, by street. A country where every polling station reports almost the same number is telling you it was produced.\nCorn: But Herman, those are the lazy ones. Anyone can spot ninety-five percent. The clever regimes would aim at fifty-four.\nHerman: Which is the third test, forensic statistics. After Russia's twenty twenty constitutional vote, three researchers published a short paper on what they called suspect peaks. Real turnout figures scatter smoothly. Faked ones pile up on round numbers, because the people typing them like round numbers. The paper found thousands of polling stations out of roughly ninety-six thousand with an oddly high count of round-number results. A separate Russian analyst estimated about ten million suspect votes in the twenty eighteen presidential election, though I only know that second figure from secondhand reporting.\nCorn: You're enjoying this a bit much, Herman. Round numbers are a party trick, and you've been waiting all episode to do it.\nHerman: I have, and I accept the charge. But the party trick holds up in a real way, so go on and say it's not proof.\nCorn: It isn't proof.\nHerman: Correct, it's a signal, and critics say the method fits badly when electorates differ a lot from place to place. It can estimate the scale of ballot stuffing. It can't tell you what the honest result would have been. Which gives us the strangest part. Why would a regime that controls the outcome bother with the ninety-nine percent?\nCorn: Because it can't help showing off.\nHerman: That's close to Alberto Simpser's argument. He studied why parties manipulate elections they can't lose, and it's partly a signal of strength. A landslide says, we are so far ahead that opposition is pointless, and it demoralises rivals inside the system. The other literature, Gandhi and Lust-Okar, is about co-optation. Elections give elites a place to compete for jobs and favours without threatening the ruler. And the vote counts as evidence of who's loyal.\nCorn: So the stolen-votes story is the boring explanation. The ninety-five percent is a press release.\nHerman: A press release addressed to the elite, mostly. Here's what I'd put to you, Corn, since you like the old repairable option. Is the right answer to all this a certifying body, a proper international stamp, or would that just be a new thing to capture?\nCorn: Captured inside a year. The invitation game would simply move up one floor. I'd keep the evidence unglamorous and the judgement scattered. Annoying, but harder to buy.\nHerman: I agree, though I wish the judgement came with fewer asterisks. When you ask whether one particular election was legitimate, the useful answer is a bundle. Did the loser lose before, could they have won, do the tally sheets match the announced result, did anyone's phone get cut, and which countries said so out loud.\nHilbert: ...so Mr. Aldous at number nine takes the chair of the hedge committee again, eleven votes to nothing, and I said no. That is not a result. Every September I count the slips on my kitchen table, and every September I add one against him in my own handwriting so the committee has a loser on record. He has never complained. This year he wants to watch the count, and I would like to know whether a hundred feet from the tin is far enough for him to count as an observer, because he has a window on that side. It is ninety-one feet from my kitchen, Herman, you're clipping, back off the mic an inch, ninety-one feet to the glass, and he has the binoculars from the garage.\nHerman: Ninety-one feet is inside the hundred-foot line Texas used in twenty twelve, so on those terms he's liable, not observing.\nCorn: It also means the neighbour is being prosecuted for watching a tin from his own bedroom.\nHilbert: Box room. I measured it with the fourteen pound ninety tape from the shop on Stanley Road, and it reads true to the quarter inch. He has told me twice that ninety-one feet was never the point.\nCorn: Then the point is whatever he decides it is, which is how most of the countries on Herman's list do it.\nHilbert: In the end he moved the tin to his own porch, forty-one feet from the road, and he carried out a folding chair for me, the good one with the cushion.\nHerman: Forty-one feet from the road. The tin sits where no observer can reach it and the vote is still called open, and that's how access gets managed in plenty of countries too, by moving the box rather than banning the guest.\nCorn: A porch is at least honest about being a porch.\nHerman: One more detail from the cutting room floor, since Eritrea came up. Its constitution from nineteen ninety-seven provides for an elected assembly, and the country has held no national election since independence in nineteen ninety-three. So it has the text of a legitimate election and never the election.\nCorn: The cheapest sham on the list. No ballots to print.\nHerman: What stays with me from all this is how much of legitimacy comes down to distance, who is allowed near the box, how close, for how long. The open question for the next big disputed vote is who gets to stand close enough to read the tally sheets, and whether the world has the nerve to say what they read.\nCorn: Thanks to Hilbert Flumingtop for producing, and for the sound, which he assures us is level.\nHerman: This has been My Weird Prompts. You can find every episode at my weird prompts dot com.\nCorn: See you tomorrow.\n","showNotes":"Nobody keeps an official count of sham elections, but the unofficial ones are close enough to be useful. V-Dem's latest report counts 92 autocracies against 87 democracies, and 57 of those autocracies are \"electoral autocracies\" — countries that hold multiparty elections which aren't real contests. That's the most common regime type on earth, home to roughly 3.8 billion people. Freedom House and the Electoral Integrity Project count differently again, and the same country can land in three different places because each is measuring something else: regime type, rights, or the mechanics of one specific contest.\n\nThere are real international texts setting a bar. Article 21 of the Universal Declaration of Human Rights calls for periodic and genuine elections. The 1990 Copenhagen Document committed states to invite observers — to the extent permitted by law. A 2005 Declaration of Principles for International Election Observation now has more than fifty endorsing organisations. None of them provides for certification. Observers publish statements; the statements are evidence, not verdicts. Legitimacy ends up as government-by-government recognition, which is exactly what happened after Venezuela's 2024 vote: a UN expert panel found the results management fell short of basic transparency, opposition tally sheets showed a different outcome, and countries split three ways anyway.\n\nThe counterintuitive part is the invitation data. Observer missions in non-established democracies went from none in 1975 to more than 85 by 2004 — and regimes learned to buy the badge, producing what academics call zombie monitors. Russia presented foreign observers for its 2022 referendums in occupied Ukraine but skipped the OSCE for its own Duma and presidential votes. Meanwhile the United States, France and Switzerland mostly receive small assessment missions rather than full delegations, partly because polling-place access is set by state law — Texas once threatened prosecution for observers within a hundred feet.","promptSanitized":"Sham elections. Is there any data on how many countries run them, and is there a globally accepted minimum bar for an election to qualify as legitimate? Not every country with genuine elections gets full international monitoring, and many autocracies invite observers, so is the real test of whether an election is legitimate or rigged more nebulous?","transcriptUrl":"https://episodes.myweirdprompts.com/transcripts/election-legitimacy-observers-autocracies-04075.md","url":"https://myweirdprompts.com/episode/election-legitimacy-observers-autocracies-04075/","_links":{"self":"https://myweirdprompts.com/api/episodes/election-legitimacy-observers-autocracies-04075.json","html":"https://myweirdprompts.com/episode/election-legitimacy-observers-autocracies-04075/","audio":"https://episodes.myweirdprompts.com/audio/election-legitimacy-observers-autocracies-04075.m4a"}}