Ninety-two days out from an election that half the country has already stopped arguing about and started resigning itself to, Daniel sends us a prompt that is less a question than an indictment with a bibliography attached.
He's been building this one for years. You can hear it in how tight the argument is.
He says he's spent episode after episode sharpening his complaint about the Israeli political system, and he wants us to take it the last mile. The complaint is this. Too many parties. All of them campaigning on the big emotive national things, security, identity, who belongs here and who doesn't. Almost none of them with anything resembling a plan for rent and groceries and the price of a flat.
And his diagnosis of the left is the part he's honoring us for, which is a little nerve-wracking.
He puts it in two halves. The collapse of the left had a very specific moment in history, and it can also be largely explained by simple demographics. Tel Aviv is the biggest metro area in the country. It's also a sliver of it. And then he goes further. He says the real situation isn't just division, it's that we've ended up with several discrete groups, each with a fundamentally different picture of what the social contract should even be. The ultra-Orthodox world, large parts of which reject the legitimacy of the Zionist state and won't serve in its army while drawing its benefits. The Tel Aviv world, outward-looking, invested in Israel's reputation abroad, deeply Zionist and committed to serving the state. The National Religious, straddling the gap. And Arab Israelis, a portion of whom dispute the country's fundamental right to exist as a Jewish state.
And that's where he brings in the word everybody uses.
Tribes. He says it's the description that gets worn out from overuse, and then he points out it echoes into Gaza and the West Bank too, wherever we talk about who governs and how, without an indefinite occupation.
Then he asks the actual question.
He asks whether the Western democratic model, the thing the state was built on, might be fundamentally ill-fitting to this part of the world. And he wants us to go find examples. Countries, current or historical, that found some alternative arrangement for holding several groups with incompatible visions of the social contract. He'll take anything. States within states. Secession, even. He says the exact parallel almost certainly doesn't exist. He wants to know if a wide enough lens turns up anything instructive anyway. So let's start with why the system produces this many parties in the first place.
Here's the immediate scene, and it's worth taking seriously because the mechanics of it are the whole first half of this. The Twenty-Fifth Knesset dissolved on the seventeenth of July. That's the first Israeli legislature since nineteen eighty-eight to actually finish a full four-year term without collapsing early. Everything else in the last three decades fell before its time. This one served its sentence.
And the reward for good behavior is thirty-eight lists.
Thirty-eight lists filed by the eighth of September deadline. About twelve are expected to cross the threshold and take seats. Seven million three hundred thirty-seven thousand eligible voters, which is up roughly five hundred forty-eight thousand from 2022. Sixty-one seats out of a hundred twenty is a majority, and nobody is going to get there alone.
And the campaign itself, what's it actually about?
Security. Judicial reform. The October seventh inquiry. Haredi conscription. That's the spine of it. Cost of living does surface, and it's real, an April report put Israeli costs about twenty percent above the wealthy European countries, and Bennett's been running an all-out war on monopolies line, and there's a New Economic Party running with the reservists on fighting tycoons and the cost of living. But it's at the margins. It's the thing parties mention in the last thirty seconds of a debate before they go back to the thing they actually want to talk about.
So the portrait Daniel's painting is accurate. That's worth saying once out loud. It's not a feeling he has. It's the observable shape of the thing.
It's the observable shape. The mechanics explain the fragmentation, though. That part's not mysterious at all, and it's where I'd want to start.
The mechanics explain the fragmentation. They don't explain why the fragmentation hardened into four groups that don't agree on what the country is for. Those are two different questions, and I think we keep collapsing them into one.
Agreed. Let's do the mechanics first, because it's the cleanest part of the whole story. Israel runs closed-list proportional representation, single nationwide constituency, three point two five percent threshold. And when political scientists look at that, they classify it as an extreme case. In the Laakso-Taagepera literature, which is the standard framework for this, the effective number of parties scales with district magnitude and inverse to threshold. Israel maxes the first and minimizes the second.
Translate.
District magnitude is a hundred twenty. There's no district-level gatekeeping, no first-past-the-post anywhere, no local races. It's one country, one ballot, one list, and anything clearing three and a quarter percent gets a proportional slice of a hundred twenty seats. Three point two five is low. In most parliamentary systems, you'd see four, five, some higher. Israel's floor is just high enough to punish a party that spreads itself thin, and just low enough to reward a party that speaks hard to a narrow slice.
Which is why you get thirty-eight.
Which is why you get thirty-eight, and which is why the strategic advice to any new party is always the same. Find the passionate niche, own it entirely, do not try to be everybody's second choice.
And the second consequence of the low threshold is that no party campaigns to actually win. They campaign to be the kingmaker.
That's the part I keep coming back to. With a hundred twenty seats and sixty-one needed for a majority, the real audience for any party's campaign is not the voter in the middle. It's the future coalition partner. And that changes the rational message completely. You're not persuading a median voter. You're signaling to Ben-Gvir or to Lapid or to Abbas that you're the ally they can't govern without, which means you lead with the thing that's unmistakable and emotive.
So cost of living policy is a public good and security and religion are club goods.
Exactly that, and it explains the whole surface of the campaign. If I'm a small party with four seats and I want invitations to coalition talks, I don't publish a housing paper. I shout the thing that makes me legible. And once I've built my brand on the shout, my voters don't want me to trade it away. And that's what locks this in across cycles.
The left collapsed. Where do we actually date it?
If you want a single data point, Labor in 1992 under Rabin, forty-four seats. Labor in 2022 under Merav Michaeli, four. Barely clearing the threshold. That's the clearest single frame you can put on it.
That's a party that governed the country for its first three decades.
And it's not just Labor. This is the observation from earlier in the show that Daniel's referencing. In April 2019, the non-right Zionist parties held forty-nine seats, forty-one percent of the Knesset. By December 2020, they were polling at twenty-six. That's a forty-seven percent fall in twenty months. Not twenty years. Twenty months.
And the response to it was consolidation. Labor and Meretz merged in July 2024 into The Democrats under Yair Golan, explicitly to stop the bleeding.
Explicitly. The name tells you the theory. If you stop being two small parties and become one moderate-sized one, you get above the noise. The problem is that merging parties doesn't manufacture voters. It just re-organizes the ones you already had. Shmuel Rosner put it in a way that's stuck with me ever since I read it. He says the left always plays with a bad hand because its pool of potential votes is small.
Pool of potential votes. That's the demographic half of it, and it's the part I find most interesting because it's the part nobody inside the system wants to talk about.
Rosner says roughly forty percent of Israelis are center and left, and the Tel Aviv tribe specifically, the left-leaning one, is shrinking and growing older. Both adjectives matter.
Shrinking. And growing older.
The right-wing Knesset quip is that an ambulance siren is either a left-wing voter dying or a right-wing voter being born. It's an ugly line, and it's also the arithmetic. The most electorally dense left constituency is also the one with the lowest birth rate and the highest outward migration among the young. Demography is not a policy. You can't run a campaign that fixes it.
I think that's the part that gets missed. When people talk about the collapse of the Labor party they usually reach for ideology. There was a moment. There was Oslo, there was the Second Intifada, there was a credibility collapse on security. All of that's true. But you can have the correct ideology and still lose if the pool is draining underneath you. Tel Aviv is the largest metro area in the country and it's still a slice of it. Not the majority.
Rosner's phrase about a bad hand. It's not that the left plays badly. It's that the deck is stacked.
And I'd add one layer to that. The demographic answer is the more interesting because it's the one that isn't reversible by any policy in any reasonable timeframe. A party that fixes its message can grow back. A party whose base isn't being replaced can't. That's the asymmetry between what the left is doing and what the Haredi world is doing.
Which leads us to the four groups, and I think this is where the episode should sit for a while, because Daniel's framing of it is more precise than the popular version. He's not saying Israeli society is divided. He's saying the groups don't share the same social contract. Four different contracts, four different ideas of what the state is for.
Walk through them.
Haredim first. By the numbers, projected at about twenty-four point four percent of the population by 2050. Growing at four point two percent a year, which is the highest of any population in the developed world. Fifty-seven percent of the community is under age twenty, against thirty-one percent nationally. Haredi male employment at fifty-three point nine percent in 2024. Haredi matriculation eligibility at sixteen percent.
That last number, sixteen, and fifty-three point nine, those two together tell you the whole story of the coalition fight that keeps happening.
They do. And it's worth saying what the fight is really about, because it's not about conscription as a bureaucratic category. It's about obligations versus benefits. The state provides. The community draws. And in significant parts of the community, there's a rejection of the state's own legitimacy as a Zionist entity, which is the deepest version of the disagreement. And the polling is moving fast. Support for Haredi conscription went from sixty-seven percent in January 2024 to eighty-four point five percent in January 2025. The Supreme Court declared continued exemptions unlawful in June 2024. And UTJ and Shas repeatedly threatened to bring down the government over it, which they didn't do, but they got very close multiple times.
The one thing I'd add on those numbers, because it's the piece that turns it from a political fight into a structural problem. Haredim are twenty-three point five percent of those designated for military service today. By 2050, that's forty percent.
It's fifty percent of the pool that's the crunch point.
And if the picture doesn't change, non-integration costs on the order of ten percent of GDP, roughly a hundred sixty billion shekels. That's not a coalition bargaining chip. That's the arithmetic of a system that either changes or doesn't work.
The Tel Aviv tribe next. Daniel's formulation here is good. Not a monolith, but by and large outward-looking, concerned with Israel's international reputation, more secular, still deeply Zionist, and committed to the concept of serving the state. That's the world where military service is the norm, not the exception, where the kids do the gap year, where the university degree gets pulled from the American model, where the international coverage gets read.
And the size problem. Tel Aviv is the largest metro area in the country. It's also, I keep wanting to say this, a sliver.
Sliver. None of the four groups is a majority. That's the structural fact underneath the whole episode. If any one group was a majority, this would be a different country with a different politics.
The National Religious cohort.
Straddling. And this is where the coin flips hard, because the National Religious world is not the same as the National Religious leadership. Smotrich's Religious Zionist Party has merged with Feiglin's Zehut. Four far-right parties are running on three lists, and all of them are calling for expulsion of Palestinians from Gaza and the West Bank. That's the leadership position. The cohort itself is bigger than the leadership and doesn't all go there. But the leadership is loud and it's running in a system that rewards emotionally maximal signals.
So the group that by some accounts is the bridge between the Tel Aviv secular world and the ultra-Orthodox world is currently being represented by people who campaign on the least bridge-like position available.
That's the story of 2026 politics in one sentence. Everyone is optimizing for the primary audience, and the primary audience is the hardest part of your own base.
And Arab Israelis.
Joint List and Ra'am running separately this time. Ra'am's Mansour Abbas has made the most interesting move in Israeli politics over this cycle. He's calling it a pivot, from Islamism toward an independent civic platform, and he's running with Yoav Segalovitz, who's a former police brigadier general, which is not a natural pairing on paper.
Segalovitz. That's the one experiment in the whole country that's actually running a cross-tribal test.
Abbas is trying to make the case that an Arab party can be part of a governing coalition, not as a vote for the opposition to lean on when it can't form its own government, but as a legitimate coalition partner. And a former police brigadier general running with him is a very deliberate message.
And the anti-Netanyahu camp's response has been. Well. The Doha Institute's read is that they insist on relying on a Jewish majority in the Knesset and, except for The Democrats, refuse to include or rely on Arab parties, which makes a sixty-one-seat government difficult if not impossible without changing that stance.
Which is where the tribes framing gets interesting, because it cuts both ways. Likud's 2026 slogan is "There is no middle ground. It's right or Palestine." That's the same logic from the other side. Either side can only imagine a coalition that excludes the other's essential partners.
And the same tribal arithmetic echoes into every conversation about Gaza and the West Bank. Who governs, who administers, how do you do it without an indefinite occupation, and every one of those answers turns out to require working with a group your own base treats as illegitimate.
So that's the diagnosis. The mechanics explain the fragmentation. The demographics explain the shape. And the four groups each have a vision that can't be reconciled by any single policy shift.
Daniel's real question is whether history offers anything better. And I think it starts with a model Israel is already running.
Consociationalism. Arend Lijphart's framework, codified in the late sixties and developed since. Four features. Grand coalition, mutual veto, proportionality, and segmental autonomy. And the whole point of those four together is that groups who can't agree on the basics still get to govern together by not requiring agreement. You concede a stable share and a veto on the things that matter most to you, in exchange for not trying to win everything.
And the examples.
Belgium. The Netherlands under pillarisation from 1917 to 1967, where the country was structured around Catholic, Protestant, and secular pillars that ran their own schools and radio stations and unions, and nobody lived inside anyone else's. Switzerland's the more durable case. Cyprus from 1960 to 1963, which failed. Lebanon, which I want to come back to. Northern Ireland post-1998. Bosnia post-Dayton. North Macedonia after the Ohrid Agreement in 2001. Burundi after Arusha in 2000. Interwar Czechoslovakia, which had a multi-ethnic democracy before it got taken apart from outside.
And the paradox.
Wikipedia's consociationalism article lists Israel among consociational systems. It is not an aspirational entry. It's a description.
Israel already has proportionality. Already has segmental autonomy in education and personal status, the four streams running their own schools, the religious courts running marriage and divorce for their own communities. Already has a de facto grand coalition norm, because with eleven or twelve parties, every government is a grand coalition of some kind or it doesn't exist.
Which is the finding I want to stop on. Israel is already consociational. It is already running the model that the consociationalist literature recommends for exactly this kind of country. And it is experiencing maximal internal division anyway.
So the model doesn't solve it.
The model doesn't solve it. It might make it worse. That's the honest reading of the evidence and it's the one I'd want to sit with before we go anywhere near recommending anything.
Say why.
Horowitz's centripetalism is the alternative school, and his core objection is that consociation reifies ethnic divisions. It freezes them into institutions. If the state says the Maronite Christian community runs the presidency, the community now has a permanent bureaucratic stake in staying Maronite and staying separate. You've taken a cleavage that might have been fluid and you've made it the load-bearing wall of the country. Yascha Mounk made the argument two ways in his 2022 writing. He says power-sharing buys temporary peace and can worsen tensions over the long run, and the standard citation is Lebanon, where the confessional system became a machine for elites to capture state institutions and turn them into clientelist operations.
So the pitch is. Take a divided country. Freeze the divisions into the constitutional architecture. And the architecture then perpetuates the divisions that required it in the first place.
That's the sharpest version of the critique, and it lands.
And confessionalism is closer to what Daniel's describing anyway.
It's the sharper cousin of consociationalism. Lebanon's the textbook. Maronite Christian president, Sunni prime minister, Shia speaker. Fifty-four Christian seats and fifty-four Muslim seats in the parliament, in practice sixty-four each. Eighteen recognized communities, each with autonomy in education. The point is that the slots are reserved, not elected.
And Iraq post-2003.
The informal version. Kurdish president, Sunni speaker, Shia prime minister. Nobody wrote it into the constitution as cleanly as Lebanon did, but that's how governments have been assembled since the invasion.
And Daniel asked about states within states. Secession.
This is where I have to be honest. I looked for the modern functioning example of a state-within-a-state arrangement and I couldn't find one that maps onto Israel's situation. Not because it doesn't exist in abstract but because the examples that do come up, the Vatican, the Sovereign Military Order of Malta, are micro-territories with very specific historic origins, not multi-group countries that decided to split into nested sovereigns. On secession, the examples tend to be either peaceful and specific, Norway and Sweden in 1905, or violent and unresolved. And the honest answer is that the formal menu of alternatives for a country with four discrete groups inside one territory is consociation and confessionalism, not secession. That's a real gap in what Daniel asked for and we should flag it rather than fudge it.
So the closest formal analogue to what he's describing is the one that isn't dividing the country in the first place. It's reserving the slots.
Reserving the slots. And it's the one that gives you the best cautionary tale.
Then what's actually instructive here, not as a model but as an observation about where the actual constraint lives in Israel? Because if Israel is already consociational and still fracturing, then the model isn't the binding problem.
The binding problem is that the anti-Netanyahu camp won't rely on Arab parties. Except The Democrats. That's it. That's the one constraint that would have to move before any cross-tribal coalition becomes arithmetically possible, and it's not moving, and it's not moving for reasons that are deeper than the electoral system.
And Ra'am's pivot is the one live experiment running against it. Abbas is testing whether an Arab party can sit in a coalition, not as a prop but as a member. If it works, it changes the arithmetic. If it fails, it confirms the arithmetic.
If it works, it changes the arithmetic. And the two demographic clocks run on opposite schedules. Haredi growth is a slow, accelerating curve. The left's decline is a sliver getting thinner. Neither is reversible by policy in any short timeframe.
Which means the renegotiation of the social contract is coming whether anyone plans it or not.
And there's a striking signal about what kind of renegotiation people actually want. Seventy-one percent of Israelis, in a December 2025 poll, said they'd support a constitution instead of the Basic Laws.
Seventy-one percent.
On a question that in any other country would be the boring procedural one, and in Israel it's the whole point. Because Basic Laws are a legislated document that any Knesset can amend by majority, and a constitution is a two-thirds document that requires the groups to actually negotiate a shared floor. The appetite is there. The mechanism to get from here to there is not.
Let me be careful about how we land this, because Daniel explicitly said he doesn't want a fix. He wants to know what history offers.
The most useful thing history offers is not a model to copy. It's a warning about each model's cost. Consociation costs you the fluidity of the divisions. Confessionalism costs you competitive politics within groups. Centripetalism costs you the guarantee that any group's essential interests are protected. Secession, where it works, costs you the shared state entirely. Every one of those is a real cost. None of them is free.
So exact parallels to Israel are almost certainly non-existent. He said that. And the pleasant surprise is that he was right. But the menu of alternatives is real, and the incentive to think about them is not going away, because the demographic clocks don't care whether anyone plans.
The twenty-seventh of October is when we find out whether the anti-Netanyahu camp can bring itself to rely on Arab parties. That's the first real test.
I've been waiting all episode to hear how the tribes framing sounds from a man who's been in one.
It's a fine word. I don't dispute it. I'm just saying it doesn't go deep enough. Because I was in a splinter group once, and I want to tell you what it actually takes to run one.
Sixteen of us. That was the number. We broke off from a bigger group in March, and the whole thing came down to a motion about seating arrangements. Which side of the room we sat on during the general meeting, and whether the admittance table sat inside the doorway or just outside it. Motion three, second paragraph, and the wording was "the table shall be considered internal to the proceedings." Internal. That was the word. It passed by nine votes to seven, and the seven of us who thought the table sat outside the room doorkept the minutes.
The minutes.
We took the minutes with us. That's how you know a split is real. Who gets the book.
The group was sixteen, this splinter group.
Sixteen at the moment of the split. Then my cousin Yossi joined us. And he resigned four days later. So we were down to sixteen again, and so were they. Sixteen and sixteen. Two full sets of officers. Two treasurers, two secretaries, two chairs, and a joint dispute about who kept the kettle. I was elected treasurer of the breakaway group in April. I was elected treasurer of the original group the following week, because they'd forgotten to hold their election and I was the only one who showed up.
You were treasurer of both.
Eleven months. Nobody noticed. I signed off on both sets of books. I had a stamp for each. I kept the stamps in different drawers.
Nobody ever asked why the breakaway group's finances were reconciled by the same person.
It came up once. At the annual meeting. Someone raised a point of order about whether the treasurer was eligible to serve in both groups, and the chair ruled it out of order on the grounds that the question presupposed a relationship between the two groups, which we had never formally conceded. So we carried on.
And the founding document.
Written on the back of a takeaway menu. I still have it. There's a picture of a chicken on the front and Article Four on the back, which says the group recognizes the authority of the founding meeting and reserves the right to sit on whichever side of the room it damn well pleases.
That's the whole constitutional settlement of a new polity.
It ran for two years and four months. It dissolved over a dispute about whether the treasurer, me, had properly minuted the decision to buy a second kettle. Turns out we hadn't.
The trivia question is what happens when a group defines itself entirely by not being the other group.
You get two groups that are identical to each other and go to great lengths to prove they aren't. Sixteen and sixteen. Two kettles, two stamps, two sets of minutes. And both of them gradually stopped talking about the table.
Did they eventually merge?
No. They stopped meeting, mostly. The last time I went was in 2019, and there were four of us. All four had been officers of both groups at some point. So in the end the tribes got their wish. They were separate and they were identical.
That's it. You can have the split and the split doesn't produce two different things. It just produces two of the same thing.
Two sets of books. Same page count.
Let's pull this back to the twenty-seventh.
The thing I'm left holding is this. If Israel is already consociational, and it's still fracturing, the problem isn't the model. It's that the model has nothing to sit on. Consociation works when everyone beneath the architecture agrees on a minimum floor. And the whole point of the four contracts Daniel described is that they don't. So the interesting question for the twenty-seventh is whether any of the four groups is willing to negotiate that floor, or whether the entire campaign turns again on the belief that the other side can be permanently excluded.
Seventy-one percent want a constitution. Which is the closest thing the country has to a demand for a floor. But a constitution only works if the groups that write it can walk away with something. Which means somebody has to lose something.
The two clocks keep running. The Haredi share keeps growing. The left keeps thinning. Which means the negotiating position of each group is going to be very different in ten years than it is now.
Which is a reason to have the conversation sooner rather than later. Not that anyone asked us for a recommendation. Daniel didn't.
He didn't.
It's not just politics. Politics is what you get when the underlying thing hasn't been settled.
To Daniel's credit, he asked something that's unusually hard to reduce to a slogan without flattening.
Which is what we do every day, so.
Speak for yourself.
Thanks to Hilbert Flumingtop, our producer, who has been at this desk, and who has, at every stage of this conversation, refrained from explaining which committee he was treasurer of in 2017.
Two.
For more along these lines, there's episode one forty-eight, Why Can’t You Fire Your Local Politician; episode fourteen oh eight, Israel Wartime Readiness Field Guide; and episode twenty-one, Is Your AI Secretly American. This has been My Weird Prompts.
If you want to send us your own prompt, you can do it on Telegram at t dot me slash MWP listener bot. And you can find every episode at my weird prompts dot com.
See you tomorrow.
