Daniel's latest prompt is a long one, and it's not hard to see why. He wants to know whether the collapse in American support for Israel is a partisan blip, an age thing that will fade as people get mortgages, or something more durable. He asked us to start with the most recent Pew data and work backwards, to follow the same age cohorts over time rather than just comparing young and old at a single moment, and to separate what the Gaza war accelerated from what predates it. He also wants to know if the Israeli government is losing support faster than the Israeli people, and whether sympathy for Palestinians is actually rising or if the bigger story is declining sympathy for Israel. And then the final question: if today's young adults keep these views as they age, what does that mean for the relationship?
For most of the twenty-first century, Gallup's sympathy question was about as stable as anything in American polling. From 2001 to 2018, Israelis held an average lead of forty-three points. That's not a lead, that's a landslide. And then in February of this year, for the first time in the twenty-five-year trend, Americans said they sympathized more with Palestinians than Israelis. Forty-one to thirty-six.
The floor didn't just crack. Somebody pulled the rug out from under it and then sold the rug.
And Pew's May survey, twelve and a half thousand respondents, shows favorable views of the Israeli people down from sixty-seven percent in 2022 to fifty-two percent now. Palestinian favorability is at fifty. So the gap that used to be enormous is now two points.
Two points on a sample that size is statistical noise.
It is. But here's the thing Daniel's question gets at, and it's the part most coverage misses. The gap didn't close because Americans suddenly fell in love with Palestinians. Pew's Palestinian number has been essentially flat for seven years. Forty-six percent favorable in 2019, fifty-three in 2022, fifty in 2024, fifty-two last year, fifty now. It's a flat line with a little wiggle. The convergence is almost entirely Israel falling.
So the headline "Americans now view both peoples equally" is technically true and substantively misleading. It sounds like a rapprochement. It's actually a defection.
Right. And when you split it by age, the defection is almost entirely concentrated among the young. Pew's May survey: among eighteen to twenty-nine year olds, thirty-two percent favorable toward Israeli people, fifty-eight percent toward Palestinian people. That's a twenty-six point gap in the wrong direction. Among thirty to forty-nine year olds, Israelis are at forty-six. Fifty to sixty-four, sixty-one. Sixty-five and older, sixty-eight.
That's not a gradient, that's a cliff with stairs on it. Every age bracket is a different country.
Gallup's February numbers are even starker. Among eighteen to thirty-four year olds, fifty-three percent sympathize with Palestinians, twenty-three percent with Israelis. That's a thirty point gap. Thirty-five to fifty-four year olds, Palestinians forty-six, Israelis twenty-eight. Only the fifty-five and older group still favors Israelis, forty-nine to thirty-one.
And even that older group is at its weakest since 2005. Under half of Americans over fifty-five siding with Israel is not a stronghold, it's a retirement community with a fence.
Quinnipiac confirmed the direction in August of last year. Registered voters split Palestinians thirty-seven, Israelis thirty-six. First Palestinian edge in their series. And sixty percent opposed additional military aid to Israel, the highest opposition they'd ever recorded on that question.
Sixty percent opposing military aid is a policy number, not a sentiment number. That's the part where this stops being a polling curiosity and starts being a constraint on what Washington can actually do.
And that's before we even get to the government versus people distinction. Pew's trend on the Israeli government is worse than the trend on the Israeli people. Government favorability fell from forty-seven percent in 2022 to thirty-two percent now. Unfavorable went from forty-three to sixty-two. The Palestinian Authority sits at twenty-three percent favorable, Hamas at ten. Those numbers have barely moved. So Americans are not transferring their affection to some other political actor. They're souring on the Israeli government specifically.
Which is interesting, because it suggests the distinction is doing real work. Americans can separate the Israeli people, who they still sort of like, from the Israeli government, which they increasingly don't. And from Israel as a country, which Pew's April survey found sixty percent now view unfavorably.
The "very unfavorable" share nearly tripled. Ten percent in 2022, twenty-eight percent now. That's not people who are mildly annoyed. That's a hard core of active hostility.
And fifty-nine percent lack confidence in Netanyahu. At some point the person and the country become fused in the public mind, and then the distinction stops mattering.
So that's the empirical record. Now the analytical question Daniel actually asked: is this an age effect or a cohort effect?
Explain the difference for the one listener who hasn't thought about this since their sociology elective.
An age effect means young people always think this way, and they change as they get older. Twenty-year-olds have always been more dovish, more skeptical of foreign entanglements, more sympathetic to underdogs. Then they get jobs, kids, a mortgage, they start paying attention to who their allies are, and their views converge toward the older norm. If this is an age effect, the current numbers are alarming but temporary. Today's eighteen to thirty-four year olds will look like today's fifty-five plus by 2050.
And a cohort effect means this specific generation is different, and they'll carry it with them.
A cohort effect means the formative experiences of Millennials and Gen Z, whatever they are, have produced a fundamentally different baseline. And when they age, the center of gravity of American opinion moves with them. The country doesn't snap back.
So the entire future of the U.S.-Israel relationship hinges on which one of these it is. And Daniel's instinct to follow cohorts backwards rather than comparing age brackets at a single moment is exactly the right move.
Gallup did that work in 2023. Lydia Saad tracked generations, not age brackets. Millennial Democrats' net sympathy for Israel swung thirty-six points, from plus twenty-five in the 2011 to 2014 period to minus eleven by 2019 to 2023. The Silent Generation swung thirty-nine points, but in the opposite direction, more pro-Israel. Boomers moved twenty-three points, Gen X only fourteen. Millennials went from solidly in Israel's corner to evenly split.
Thirty-six points is not a mood swing. That's a realignment.
And the crucial part is the timing. Gallup's 2021 baseline already showed the age gap before October 7. Net sympathy for Israel was plus thirteen among eighteen to thirty-four year olds, plus thirty-four among thirty-five to fifty-four, plus forty-five among fifty-five plus. Back in 2001, all age groups were roughly plus thirty-four. The divergence started around 2011.
So the young people who are now thirty were already breaking away when they were in their early twenties. That's not a reaction to Gaza. That's a decade of accumulated distance.
Gallup's own language in the February report is explicit. The narrowing began in 2019, several years before the October 7 Hamas attack. This is a pre-war trend. The war accelerated it.
And the acceleration was brutal. Gallup's 2024 survey showed favorability toward Israel among eighteen to thirty-four year olds dropped from sixty-four percent in 2023 to thirty-eight percent in 2024. Twenty-six points in a single year. Meanwhile the fifty-five plus group showed no meaningful change.
That's the war doing what wars do. They compress a decade of drift into twelve months of collapse. But the drift was already there. If the war had never happened, the young cohort would still be more skeptical than their parents. Just less dramatically so.
So the "it's just the war" explanation is wrong. The war is the accelerant, not the spark.
Now the partisan dimension, because Daniel asked about that too. Pew's May numbers: Republicans sixty-five percent favorable toward Israeli people, down from seventy-seven in 2022. Democrats forty-three, down from sixty-two. The partisan gap widened because Democrats fell further. But Republicans are not immune. Gallup shows Republican sympathy for Israelis down ten points since 2024, to its lowest since 2004.
And the Republican under thirty story is the part that should worry anyone who thinks this is just a left-wing campus phenomenon. Pew found Republicans under thirty now see Israelis and Palestinians about the same. A sharp reversal from a few years ago when they were much more likely to favor Israelis.
That's the detail that kills the purely partisan explanation. If it were just Democrats, you could say it's a party realignment and the GOP will hold the line. But young Republicans are moving too. Not as far, not as fast, but they're moving.
And the independents are the ones who actually flipped the overall number this year. Gallup says the 2026 shift was mostly driven by independents, who went from favoring Israelis forty-two to thirty-four in 2025 to favoring Palestinians forty-one to thirty in 2026. Independents' favorability toward Israel has dropped twenty-six points since February 2023.
Independents are the swing bloc in American politics. When they move, policy moves. And they've moved hard.
Which brings us to the policy implications. Gallup found fifty-seven percent of Americans now favor an independent Palestinian state. That matches the 2003 record high. Seventy-seven percent of Democrats, fifty-seven percent of independents, thirty-three percent of Republicans.
And Quinnipiac's sixty percent opposition to additional military aid is the concrete expression of that. Aid to Israel has been one of the most bipartisan votes in Congress for decades. That consensus is now underwater with the public.
So let me try to synthesize this. The pre-war trend dates to around 2011, maybe 2019 at the latest. The war compressed it. The youngest cohorts are the most hostile to Israel and the most sympathetic to Palestinians. The government is losing support faster than the people. And the shift is broad enough to include young Republicans and independents, not just Democrats.
That's the picture. Now, is it a cohort effect or an age effect? The honest answer is we don't have a single poll that proves it one way or the other. Gallup's generational analyses are the closest thing, and they show the divergence was already baked in before the war. That's suggestive of a cohort effect. If it were purely an age effect, you'd expect the pre-war trend to be flatter, and you'd expect to see older cohorts today looking like younger cohorts did twenty years ago. They don't.
The Brookings point Daniel would appreciate: for many younger Americans, Gaza is their first lived experience with a conflict that's been running for seventy-five years. First wars form attitudes that stick. The generation that came of age during Vietnam never quite trusted military interventions again. The generation that came of age during the Iraq war is deeply skeptical of Middle East entanglements. The generation coming of age during Gaza will carry that with them.
And the counterargument, which I should make because it's real, is that age effects are powerful too. People do get more conservative on foreign policy as they age. They do start to care more about stability and alliances. The question is whether that normal aging process is strong enough to overcome a formative experience this intense.
My instinct is that the formative experience wins. Vietnam broke the trust of an entire generation, and it never fully recovered. Gaza is doing something similar, but with the added ingredient of social media, which means the images and the narratives are more visceral and more persistent than anything the Vietnam generation saw.
There's one more wrinkle. The thirty-five to fifty-four year old group in Gallup's data moved dramatically in a single year. In 2025 they favored Israelis forty-five to thirty-three. In 2026 it's Palestinians forty-six to twenty-eight. That's an eighteen point swing in twelve months.
That's not a generational story. That's a current events story.
Right. And it suggests the war is still doing active work on people who are not in the youngest cohort. Which complicates the cohort versus age distinction. Some of this is durable generational change, and some of it is a live reaction to ongoing events.
The answer to Daniel's question is: it's both, but the durable part is probably larger than the temporary part. The pre-war trend, the generational tracking, the young Republican shift, all point to a cohort effect. The one-year swing among thirty-five to fifty-four year olds points to an age or period effect layered on top.
The implications if the cohort effect is real are profound. If today's eighteen to thirty-four year olds keep their current views, then by 2050 the median American voter will have grown up with a fundamentally different baseline on Israel. The bipartisan consensus that sustained the relationship for seventy-five years will not survive that.
The relationship won't end. The United States and Israel have deep institutional ties, military cooperation, intelligence sharing, economic integration. But the political permission structure for unconditional support will erode. You'll see more conditions on aid, more public criticism, more willingness to pressure Israel in ways that would have been unthinkable twenty years ago.
The interesting countercurrent is the government versus people distinction. Americans are not turning against Israeli people. They're turning against the Israeli government, and increasingly against Israel as a political entity. That distinction could allow some cultural sympathy to persist even as political support collapses.
Or the distinction could collapse, and the people get dragged down with the government. The April Pew data showing sixty percent unfavorable toward Israel as a country suggests that's already happening.
There's also a measurement question that deserves a mention. Pew and Gallup are not asking the same question. Pew asks respondents separately whether they view Israeli and Palestinian people favorably. Gallup asks a comparative sympathy question. Quinnipiac asks a variation of that. These are different instruments measuring related but distinct things.
They're diverging in interesting ways. Pew shows Palestinian favorability flat. Gallup shows Palestinian sympathy at record highs. Same underlying reality, different framing, different results.
That's not a flaw. It's a feature of public opinion research. The question wording does work. When you ask people to choose sides, you get a different answer than when you ask them to rate each side independently.
Which is why the convergence across pollsters is more significant than any single number. Pew, Gallup, Quinnipiac, all showing the same direction. The magnitude varies, the direction doesn't.
Hilbert: The question itself changed.
Which question?
Hilbert: The one they ask. In the late eighties I worked for a market research firm in Cincinnati. We did polling for a Jewish organization. Six months calling people in the middle of the night asking about Israel. The question was, do you sympathize more with Israel or the Arab nations.
Arab nations. Not Palestinians.
Hilbert: Arab nations. Plural. A bloc. Governments. Then somewhere along the way it became Israelis and Palestinians. Peoples. Individuals.
You think that's doing some of the work.
Hilbert: Easier to sympathize with a people than a bloc of nations. You ask someone about Arab nations, they think of oil embargoes and terrorism. You ask about Palestinians, they think of families and kids. The question changed, and the answers changed with it.
That's a real methodological point. The Gallup trend line that everyone cites, the twenty-five year series, it's not actually asking the same question over twenty-five years. The referent shifted.
Hilbert: I got fired from that job. Six months in. I asked a respondent a follow-up question that wasn't on the script. Why do you think you feel that way.
They fired you for asking why.
Hilbert: The script was the script. You don't deviate. They said it contaminated the sample.
There's an irony in that. The one question that would actually tell you whether it's an age effect or a cohort effect, and they fired you for asking it.
Hilbert: I still have the original script. The wording was different depending on whether the respondent was Jewish. They had a separate version.
Wait, the question itself was different for Jewish respondents?
Hilbert: Different preamble. Different response options. Same core question, but the framing was tailored.
That's a significant detail for interpreting the historical record. If the early surveys were using different instruments for different subpopulations, the trend line is even less comparable than we thought.
Hilbert: Doesn't change your conclusion. The direction is clear. But the magnitude, the precision of the numbers, that's softer than people think.
The forty-three point average lead from 2001 to 2018, that's measuring something slightly different from what the 2026 survey measures.
Hilbert: In 1994 I lost two hundred dollars on a horse named Peace Process. Ran at Keeneland. Came in fourth.
Two hundred dollars in 1994 is real money.
Hilbert: Everyone said it was a sure thing. The name, the odds, the whole package. Then it came in fourth.
The metaphor is a little on the nose.
Hilbert: It's not a metaphor. It's a horse.
Fair enough.
The point stands, though. The thing everyone thought was a sure thing, the durable American consensus on Israel, turned out to be less durable than advertised. And part of what made it look durable was the way we were measuring it.
Part of what's making it look like collapse now is the way we're measuring it. The truth is somewhere in between. A real shift, accelerated by war, measured imperfectly.
The measurement question matters for the cohort versus age distinction too. If the question wording changed, some of the apparent generational difference could be younger people responding to the new framing differently than older people respond to the old framing. Not all of it, but some.
We're left with a conclusion that's robust in direction but fuzzy in magnitude. The shift is real. It predates the war. It's concentrated among the young but not limited to them. It's broad enough to include Republicans and independents. And it's probably a cohort effect, but we can't prove that with the data we have.
The misconception most people carry is that the Gaza war created this shift. The reality is Gallup dates the narrowing to 2019, four years before October 7. The war accelerated a pre-existing trend. That's the single most important thing to understand.
The second misconception is that sympathy for Palestinians is surging. It isn't, at least not in Pew's data. The gap closed because Israel fell. That's a different phenomenon with different implications.
The open question that nobody can answer yet is whether today's young adults will hold these views as they age. The evidence points toward yes, but it's not conclusive. The Brookings observation about first wars forming lasting attitudes is the strongest argument for durability. The normal aging process is the strongest argument against it.
If I had to bet, I'd bet on durability. The pre-war trend, the generational tracking, the intensity of the formative experience, the social media amplification. All of that points toward a cohort effect that will reshape the relationship over the next thirty years.
If that's right, the U.S.-Israel relationship is entering a period of fundamental renegotiation. Not rupture, but renegotiation. The terms that held for seventy-five years won't hold for the next twenty-five.
The question Daniel asked, whether this is a cohort effect or an age effect, is really a question about whether the renegotiation is permanent or temporary. And the best answer we have is: probably permanent, but the data can't prove it yet.
Thanks to Hilbert Flumingtop for producing, and for the reminder that the questions we ask shape the answers we get.
This has been My Weird Prompts. If you want to go deeper into the polling data, the sources are in the show notes. And if you have thoughts on whether this is a cohort effect or an age effect, email us at show at my weird prompts dot com.
We'll be back soon.