Daniel sent us this one — he's noticing something that I think a lot of people feel but never quite articulate. We all know what a parliament does in theory: representatives argue about laws and pass them. But then there's this second chamber — the Senate, the House of Lords, whatever it's called — and suddenly everyone's a bit fuzzy. Why does it exist? What's it actually doing up there? And Daniel's asking two things: how many countries actually run bicameral legislatures, and what commonalities and differences can we see in how these second houses operate.
The timing on this is genuinely interesting, because right now — I mean literally in the past year or so — you've got at least four countries with active reform or abolition proposals on the table for their second chambers. The Czech Republic, Ireland, Uzbekistan, Kenya. This isn't a dusty civics lesson. These are live constitutional fights happening as we speak.
Which is the thing, right? Most people could name the US Senate and the House of Lords, but if you asked them to explain why a second chamber exists beyond "checks and balances," you'd get a lot of hand-waving. And that answer turns out to be surprisingly incomplete.
Surprisingly incomplete is generous. It's wrong for most countries. "Checks and balances" is the American story. Most second chambers exist for completely different reasons — federal representation, territorial representation, ethnic power-sharing. If you're applying the US rationale to, say, the German Bundesrat or the Irish Seanad, you're misunderstanding what you're looking at.
We're going to do what Daniel's asking: map out the global landscape, figure out what these second chambers actually do, and then get into why so many of them are under pressure right now. Because the abolition trend is real — roughly twenty countries have scrapped their second chambers since nineteen fifty — and the ones facing reform debates in twenty twenty-five and twenty twenty-six tell us a lot about what's working and what isn't.
Let's put some numbers on this. As of twenty twenty-four, the Inter-Parliamentary Union tracks roughly eighty bicameral parliaments out of about two hundred countries worldwide. That's around forty percent. But the distribution isn't random at all — it's almost perfectly split along one fault line.
Federal versus unitary.
Nearly every federal state is bicameral — the US, Germany, India, Brazil, Australia, Canada, Ethiopia, Nigeria. The second chamber represents the subnational units. That's the whole point. Meanwhile, most unitary states are unicameral — one chamber, one national legislature. The exceptions are places like the UK, France, Japan, Ireland — unitary countries that keep a second chamber for other reasons.
Which is where it gets messy. Because if you're a unitary state running a second chamber, you're spending money and legislative time on an institution that doesn't have the clean federal rationale. And that's the core puzzle Daniel's pointing at — why do roughly eighty countries keep these things around when they're almost never as powerful as the first chamber, they slow down legislation, and they cost real money to run?
The answer isn't one answer. It's a bundle of distinct rationales that vary by country, and they don't always overlap neatly. What we're really doing here — and this is what makes it interesting — is comparative institutional analysis. We're asking: given that these chambers exist, what job are they actually performing? Not what the civics textbook says, but what function they serve in practice.
Because sometimes the textbook answer and the real answer have almost nothing to do with each other. The House of Lords is officially a revising chamber. In practice, it's also a retirement home for political loyalists and a weird repository of hereditary titles that somehow survived into the twenty-first century. The two descriptions coexist uncomfortably.
That's the kind of thing we want to pull apart. Not "here's how a bill becomes a law" — that's the civics lecture version. We want to look at the institutional design choices: who sits in these chambers, how they get there, what powers they actually have, and why some countries are quietly abandoning the whole idea while others couldn't function without it.
Once you see the global pattern, the obvious next question is: what are these second chambers actually doing? And the answer turns out to be four very different things. They're not all playing the same sport.
Which is the first thing most people get wrong. We assume the Senate is what a second chamber looks like, and everything else is just a weaker version of that. But the US Senate exists for federal representation — Wyoming and California each get two seats, end of story. That's rationale number one: giving subnational units a voice at the national level. Germany's Bundesrat, India's Rajya Sabha, same logic.
The Bundesrat is the one that really breaks the mental model most people have. Its members aren't individual politicians — they're delegates of state governments. Literally cabinet ministers from the Länder. They vote as a bloc per state, and if a state election changes the government in, say, Bavaria, the Bavarian bloc in the Bundesrat flips overnight. It's not a legislature in the normal sense. It's an intergovernmental body that happens to sit in a parliament building.
You're a state premier, you send your ministers to Berlin, they vote as a unit based on what your state government decided. That's not "sober second thought." That's federal power projection.
And that brings us to rationale number two — the one people do reach for when they're guessing: legislative review. The sober second thought. The House of Lords, the Canadian Senate — these chambers exist to catch errors, slow down hasty legislation, and bring expertise that the elected lower house might lack. The Lords has former judges, scientists, business leaders. They can't block legislation permanently, but they can make the Commons think twice.
That's the key distinction. The Lords can delay bills by one year under the Parliament Acts of nineteen eleven and nineteen forty-nine, and they can't touch money bills at all. The Canadian Senate is similar — appointed, not elected, and its job is revision, not confrontation. These chambers are designed to be deferential.
Which brings us to rationale three: territorial or special interest representation. France's Sénat massively over-represents rural communes — tiny villages get a disproportionate voice. Ireland's Seanad has these vocational panels for universities, agriculture, labour, culture. The idea is that certain interests — geographic or professional — need representation that the population-based lower house won't deliver.
The Irish university seats are the ones that always crack me up. Trinity College Dublin and the National University of Ireland each elect senators. It's like if Harvard and Yale got to send people to the US Senate.
Then there's rationale four, which is the most serious and the least discussed: ethnic or regional power-sharing. Bosnia's House of Peoples is explicitly designed so that Bosniaks, Croats, and Serbs each have guaranteed representation. Ethiopia's House of Federation represents the country's ethnic nations and nationalities, and it has the power to interpret the constitution. These chambers exist to prevent majority domination in deeply divided societies.
You've got four completely different jobs — federal representation, legislative review, territorial voice, ethnic power-sharing — and they produce completely different institutions. Which is where the selection mechanism becomes the thing you really want to watch.
This is the most revealing variable. How do you get into the second chamber? If you're directly elected — US, Australia, Brazil, Japan — you've got your own democratic mandate, and you're going to use it. Directly elected second chambers are the ones that pick fights with the lower house. Australia's Senate has triggered double dissolutions. Japan has these "twisted Diets" where the two chambers are controlled by different parties and nothing moves.
Whereas if you're appointed, like Canada or the pre-nineteen ninety-nine House of Lords, you know you're there on someone else's authority. You're not going to block the government's agenda. You'll suggest, revise, delay maybe — but you won't declare war.
Then there's the middle ground — indirect election by subnational legislatures. Germany's Bundesrat delegates, India's Rajya Sabha members elected by state assemblies, South Africa's National Council of Provinces. These people have a mandate, but it's a mediated one. They're accountable to state governments, not directly to voters.
France does this hybrid thing where the Sénat is elected by an electoral college of about a hundred and fifty thousand local officials — mayors, municipal councillors, regional councillors. So it's indirectly democratic, but the electorate is tiny and heavily tilted toward rural areas. Which is exactly why the Sénat over-represents the countryside. The selection method is the policy outcome.
Once you see that, the power spectrum makes sense. The US Senate can block any legislation, confirm appointments, ratify treaties — it's nearly co-equal with the House. Because it's directly elected and represents states. It has the democratic legitimacy to say no. The House of Lords can only delay. Because it's appointed and everyone knows it. The Chinese People's Political Consultative Conference isn't even formally a legislative body — it's purely advisory. The Irish Seanad can delay but not block. The power tracks the democratic legitimacy, and the democratic legitimacy tracks the selection method.
Daniel's question — why do we need this exactly — the answer depends entirely on which of the four rationales a given country is working with. And the selection method tells you, faster than anything else, what the chamber thinks it's supposed to be doing. If it's directly elected, it thinks it's a co-legislator. If it's appointed, it knows it's a review body. The trouble starts when the rationale and the selection method don't match.
That mismatch is where the abolition debate lives. You look at a chamber and ask — does its selection method actually serve its stated rationale? If the answer's no, the reformers start circling.
They've been circling for decades. Since nineteen fifty, roughly twenty countries have just pulled the plug entirely. New Zealand in nineteen fifty, Denmark in nineteen fifty-three, Sweden in nineteen seventy, Peru in nineteen ninety-three, Kyrgyzstan in two thousand five, Senegal in two thousand twelve. The pattern is so clean it's almost a law of political science: small unitary states with appointed or weak second chambers abolish them. No federal state has ever done it.
Because in a federal state, the second chamber is the deal. It's why the federation exists. You can't abolish the Bundesrat without dissolving Germany. The states would walk.
But if you're New Zealand — small, unitary, homogeneous, strong rule-of-law tradition — your upper house was a legislative speed bump nobody needed. They just voted it out. And the sky didn't fall. New Zealand's been unicameral for over seventy years and nobody's clamoring to bring back the Legislative Council.
Which is the counter-model Daniel was hinting at. What do unicameral countries do instead? Sweden has this formal referral process to the Council on Legislation — a panel of judges that reviews bills for constitutionality before they pass. Denmark and New Zealand have unusually strong committee systems where the real scrutiny happens. You don't need a whole second chamber of politicians if you've got robust pre-legislative review and a committee structure that actually works.
That's the thing — the unicameral path works. But it works best in places that are fairly homogeneous, unitary, and have deep rule-of-law traditions. It's not a universal prescription.
Let's talk about who's in the hot seat right now. Because the twenty twenty-five to twenty twenty-six reform wave is interesting. The Czech Senate was only created in nineteen ninety-six, and it's never been popular. Turnout in Senate elections is abysmal — sometimes below twenty percent. Public trust is low. So you've got proposals to either strip its powers or merge it with the Chamber of Deputies entirely.
Then there's Ireland. The Seanad survived an abolition referendum in twenty thirteen — fifty-one point seven percent voted to keep it, which is about as narrow as constitutional survival gets. Now in twenty twenty-five, reform pressure is back. The proposals are actually creative: expand the electorate so every citizen can vote for Seanad members, not just the current tiny franchise of university graduates and politicians. Add diaspora representation. Make it a genuine voice for the Irish abroad instead of what it is now — which is, let's be honest, a patronage mill with a few useful vocational voices mixed in.
The university seats are the weirdest bit. A tiny fraction of the population gets to elect senators because they went to Trinity or the National University. It's like a constitutional fossil that somehow didn't go extinct.
Uzbekistan's an interesting case in the other direction. Their Senate was created in two thousand five, and it's being discussed as a potential model for other Central Asian countries — a second chamber that's indirectly elected by regional councils, designed to represent territorial units in a highly centralized state. Whether that's genuine federalism or window dressing is a separate question, but the fact it's being exported is notable.
Kenya's Senate — created by the twenty ten constitution — has been in a rolling turf war with the National Assembly ever since. The fight is over county revenue allocation. The Senate is supposed to protect county interests, but the National Assembly controls the budget. So you get these moments where the Senate says "the constitution gives us this power" and the National Assembly says "we have the money, good luck." It's a live stress test of whether a federal-style second chamber can work in a country that's still figuring out what federalism means in practice.
That brings us to the counterintuitive finding that I think is the most important thing in this whole discussion. Second chambers that are directly elected tend to be more controversial, more likely to face abolition debates, and more likely to cause constitutional crises than appointed ones. That's exactly backwards from what you'd expect. You'd think democratic legitimacy would make a chamber more secure.
Democratic legitimacy means democratic mandate. And if you've got a mandate, you use it. Australia's Senate is directly elected, and in nineteen seventy-five it blocked supply — refused to pass the budget bills — which triggered the dismissal of Prime Minister Gough Whitlam. That's the most extreme example of a second chamber exercising power to the breaking point. The Governor-General actually removed a sitting prime minister because the Senate wouldn't fund the government.
Australia's had seven double dissolutions — where both chambers are dissolved and a full election is held — specifically because the Senate and the House were deadlocked. Japan has these recurring "twisted Diets" where the upper house is controlled by the opposition and nothing moves. Direct election creates partisan conflict between chambers. Appointed chambers don't do that. They can't — they know they'd lose the legitimacy fight instantly.
The chambers that get called undemocratic — Canada's appointed Senate, the House of Lords — are actually the stable ones. Nobody's storming Parliament Hill to abolish the Canadian Senate because it's causing constitutional crises. It's not causing anything. It's just... there, revising comma splices, occasionally sending a bill back with a polite note.
Canada's Senate reform saga is the perfect case study in institutional paralysis. Forty-plus years of reform proposals — elected Senate, term limits, outright abolition — and nothing happens. Because constitutional amendment requires unanimous provincial consent. Every single province has to agree. One holdout kills it. So instead you get this twenty sixteen "merit-based" appointment process where an advisory board recommends senators and the prime minister appoints them. It changed almost nothing in practice, but it gave everyone something to point at.
Given all of that — the four rationales, the selection mechanisms, the reform pressures — how should a thoughtful person actually evaluate a second chamber when it pops up in the news? Because they do pop up. The Czech Senate makes a move, the Seanad reform debate flares, and most coverage just says "upper house" and moves on.
And that's where a framework actually helps. First: what's the primary rationale? Is this chamber doing federalism, legislative review, territorial representation, or ethnic power-sharing? If you can't answer that cleanly, the chamber's probably in trouble.
Second question: how are members selected, and does that method align with the rationale? If you've got a chamber that's supposed to do sober legislative review but its members are directly elected and campaigning on party platforms — that's a mismatch. They're not going to behave like a revising body. They're going to behave like politicians.
Third: does it have real power or just suspensive veto? That tells you whether the mismatch is an annoyance or a constitutional crisis waiting to happen. Australia's Senate has real power and direct election. That combination produced nineteen seventy-five.
The three answers together predict stability. If the rationale is clear, the selection method fits, and the power level matches the democratic legitimacy, the chamber is probably secure. If any one of those is off, you'll find abolition debates. If two are off, you'll find crisis.
The federalism test is the strongest single predictor. Bicameral, almost certainly. Unitary and small? Unicameral is the norm. The exceptions — France, the UK, Japan, unitary but bicameral — are exactly the ones facing the most reform pressure right now.
For listeners who want to track this as it unfolds, the Inter-Parliamentary Union publishes annual data on parliamentary structures worldwide. And the two to watch in the next year or so are the Czech Senate and the Irish Seanad — they're the test cases for whether the abolition trend continues or stalls.
If Ireland finally expands the Seanad franchise beyond the university clubs and the politicians, that's a real reform. If the Czechs just strip powers and leave the shell, that's managed decline. Either way, they're the canaries.
We've got a framework. But here's the question I can't shake. Look at the global data — almost every low-income country is unicameral. Is that because they can't afford two chambers, or because unicameralism is actually more efficient for development? If you're trying to build a state from scratch, does a second chamber just slow you down?
I think it's both, and they reinforce each other. A second chamber costs real money — salaries, staff, buildings, sessions. If you're a developing country with scarce resources, that's hard to justify. But there's also a governance argument. When you're trying to pass land reform or build infrastructure fast, a revising chamber that can delay your legislation by a year is a luxury you might not want.
Bicameralism as a luxury good. Something wealthy democracies can afford because they've already built the basics and now they want error-correction and deliberation.
That framing gets really interesting when you think about what's coming. Because the "sober second thought" rationale — the whole argument that you need a second chamber to catch drafting errors and unintended consequences — that's exactly the kind of thing AI tools are getting good at. We're already seeing AI-assisted legislative drafting and scrutiny tools emerge. If a machine can flag inconsistencies, conflicts with existing law, and unintended loopholes in seconds, what's left for the second chamber to do?
The political judgment, presumably. The machine can tell you the bill contradicts section four of the existing statute. It can't tell you whether overriding section four is actually the right call. That's still a human question.
But the mechanical review function — the "someone should read this carefully before it becomes law" part — that's increasingly automatable. And if that was your chamber's main job, you've got an existential problem. The House of Lords spends a lot of time on exactly that kind of technical revision.
Which means the second chambers that survive the next twenty years might be the ones with a political rationale that AI can't replicate. Federal representation isn't going anywhere — you can't automate Wyoming's voice in the Senate. Ethnic power-sharing in Bosnia or Ethiopia isn't something a machine can replace. But the revising chamber model? The "we exist to catch errors" model? That one's looking more fragile by the year.
That's where I'll leave it, because I think Daniel's question — why do we need this exactly — is going to get harder to answer for a lot of countries over the next decade. Not because second chambers are suddenly useless, but because the bundle of things they do is being unbundled. Some functions will persist. Some will be automated. And some will just quietly stop mattering.
Thanks as always to our producer Hilbert Flumingtop. This has been My Weird Prompts. Find every episode at my weird prompts dot com, and if you've got a question like Daniel's — something you've noticed but never quite articulated — email the show at show at my weird prompts dot com. We'll be back next time.
Now: Hilbert's daily fun fact.
Hilbert: In the late sixteen hundreds, a now-abandoned theory held that the Aleutian Islands were the visible peaks of a sunken continent connecting Asia to North America, and several European cartographers proposed using a system of floating postal relay stations — anchored barrels with waterproof seals — to establish a mail route across the chain before the landmass fully submerged. The scheme was debated seriously in at least two royal academies and never attempted.
...anchored barrels.
I have so many questions and I'm choosing to ask none of them.
