Daniel's question this week starts with an object I haven't thought about in years. A Nokia 3310. In 2026, that phone is basically a fossil with a battery that still somehow has charge. And he points out something strange about Israel: a huge chunk of the population still uses exactly that kind of device. The ultra-Orthodox community, the Haredim, largely refuses smartphones on religious grounds. They carry kosher phones, basic feature phones with no internet, no camera, restricted texting. Daniel's question is whether anyone has studied how people relate to devices when the internet simply isn't there. His logic is that smartphone operating systems now ship with digital wellness features, screen time trackers, app limits, which suggests the companies themselves think the internet is the addictive layer. So if the internet is what makes smartphones compulsive, aren't non-internet phones the perfect control group? And what do we actually know about the mental health of populations that refuse the internet versus everyone else?
It's a beautiful question because it sounds like a clean natural experiment and then immediately gets philosophically sticky. Let's start with who we're actually talking about, because a phone with no internet is such a strange object in 2026. I mean, think about what a phone even is now. For most of us, the phone part is the least used feature. It's a camera, a map, a wallet, a social portal. The actual calling function is buried in a folder somewhere.
The object has changed category. A kosher phone isn't a smartphone with features removed. It's a different species of device entirely. It's closer to what a phone was in 2003 than what a phone is today.
Right. And the Haredi population in Israel is roughly thirteen percent of the country and growing fast. The rabbinic leadership has issued formal bans on internet-enabled devices, and a significant portion of the community uses these kosher phones approved by a rabbinic committee. The basic tier has no internet, no camera, and SMS is restricted.
And the certification process is elaborate. There are even kosher app stores for the more permissive tier, where certain apps get approved for business or medical use while everything else stays locked out. The rabbis aren't just saying no to smartphones. They've built an entire alternative infrastructure for communication.
Which is remarkable when you think about the organizational capacity required. They're not just issuing a prohibition. They're maintaining a parallel telecommunications ecosystem. That's a massive coordination effort.
And it's why the control group question is so tempting. Same device category, same basic function, but one has the internet layer and one doesn't. In a laboratory, that's exactly how you'd isolate the variable. You'd take two groups of people, give one group internet-connected phones and the other group non-connected phones, and measure what happens to their attention, their anxiety, their sleep.
The problem is that the Haredi population isn't just no internet. They're high religiosity, high community cohesion, low media exposure, and they have dramatically different economic structures. So any difference in mental health outcomes could be attributable to community structure, not phone type. You can't separate the phone effect from the religiosity effect.
Right, and that's the first crack in the perfect control. But let's back up and lay out the addiction hypothesis precisely, because Daniel's intuition is actually well-grounded in the research. The dominant framing in the smartphone addiction literature is that compulsive use is driven by variable reward schedules, social validation loops, and infinite scroll. All of those are features of internet-connected apps, not of the device itself. The phone is a delivery mechanism. The internet is the drug.
And the attention economy research from the last couple of years documents this in detail. Social media and internet platforms use engagement-maximization design patterns. The slot machine mechanics of pulling to refresh, the unpredictable reward of notifications, the endless feed. Those are the mechanisms most associated with compulsive checking. So Daniel's intuition that the internet is the addictive ingredient is well-supported.
The phone itself is just glass and silicon. A Nokia 3310 and an iPhone are both handheld devices with screens and batteries. The difference is entirely in what they connect to. So the control group logic seems clean on the surface. Same device category, no internet layer. The comparison should isolate the variable.
But here's a thought. Even the physical design is different. A Nokia 3310 is a brick. It's not comfortable to hold for hours. The screen is tiny and monochrome. There's nothing to look at. An iPhone is designed to be held. The glass curves to your hand. The screen is bright and responsive. So even before we get to the internet layer, the objects themselves invite different relationships.
That's a subtle point. The device design itself is optimized for engagement, not just the software. A kosher phone is uncomfortable to use for anything except calls and texts. The friction is built into the hardware.
And that's before we even get to the confounds. The Haredi community isn't just a random sample of Israelis who happen not to have internet. They're a population that has made a collective, religiously motivated choice to limit outside information. They have dense social networks, strong community support, clear social roles, and a meaning-making framework that most secular people don't share. Any wellbeing difference could come from any of those factors.
It's like trying to test whether coffee causes anxiety by comparing a monastery to a stock trading floor. Sure, one group drinks less coffee, but there are about forty other differences that matter more.
That's exactly the confound. And it's not just religiosity. Haredi families have significantly higher poverty rates than the general population, roughly thirty to forty percent below the poverty line. They have different educational systems, different employment patterns, limited access to secular mental health services. So even if you found that Haredi Israelis report better wellbeing on some measures, you couldn't say it was because they don't have smartphones. It could be because they have strong community bonds and religious meaning-making, which are independently associated with life satisfaction across cultures.
The research on religion and wellbeing is robust. Religious participation correlates with life satisfaction pretty consistently, across different countries and different faiths. So the Haredi population is confounded at the most basic level. They're not a control group. They're a treatment group for about six different variables at once.
And the existing research on non-internet phone use doesn't solve the problem. There's a small literature on dumbphone users, people who voluntarily downgrade from smartphones. And there are studies of digital detox interventions. The consistent finding is that users report improved attention and reduced anxiety. But these are self-selected populations and short-term studies. The people who choose to downgrade are already motivated to change their behavior. They're not a random sample.
The digital detox studies are even worse as evidence. Someone goes off their phone for a weekend, reports feeling calmer, and the study publishes. But that doesn't tell you anything about long-term effects or about people who didn't choose to detox. It tells you that people who want a break feel better when they take one.
And there's a deeper problem with the detox studies. The people who sign up for them are usually the people who feel their phone use is a problem. So you're studying the people who are most likely to benefit from a break. It's like studying the effects of dieting by only looking at people who voluntarily join a weight loss program.
Which is selection bias in its purest form. The people who opt in are different from the people who don't. And the people who complete the study are different from the people who drop out. By the time you get to the results, you're looking at a very specific subset of the population.
And then there's the bigger debate about adolescent mental health and social media. Jonathan Haidt's Anxious Generation claims that social media caused the teen mental health crisis. But researchers like Candice Odgers have pushed back hard, arguing the causal evidence is weak and the effect sizes are small. The re-analysis suggests that social media use explains a tiny fraction of the variation in teen mental health outcomes, and that individual differences matter enormously.
So even the strongest version of the addiction hypothesis, the one that says social media is destroying a generation, is contested. The effect is real but small and heavily moderated by who you are and how you use the platform.
And that moderation point is crucial. The same platform can be harmful for one person and neutral or even beneficial for another. A teenager with a strong social network who uses Instagram to maintain friendships is having a different experience than a teenager who is isolated and uses Instagram to compare themselves to influencers.
Which brings us to the digital wellness features. Apple introduced Screen Time in 2018 with iOS 12. Google introduced Digital Wellbeing in 2019 with Android 9. And the existence of those features is revealing. The OS makers are admitting that the device is a vector for compulsive use. They built tools to help you monitor and limit your own behavior. That's an admission, not a solution.
It's like a casino installing a clock and a responsible gambling brochure. The gesture acknowledges the problem while the business model stays exactly the same.
And the features are opt-in, easily bypassed, and there's almost no peer-reviewed evidence they change long-term behavior. You can set a screen time limit and then tap ignore. You can uninstall the app that's tracking you. The tools are designed to make you feel in control without actually changing the underlying engagement mechanics.
There's a great detail here. When Apple introduced Screen Time, they also introduced a feature where parents could set limits on their children's devices. But the children quickly found workarounds. They'd change the time zone on the device to bypass the limit. They'd watch videos through iMessage. The cat-and-mouse game started immediately.
That's a perfect illustration of the problem. The tools assume individual willpower. But the entire design of the device is oriented toward undermining willpower. You can't willpower your way out of an environment designed to capture your attention.
So the control group logic seems clean but turns out to be messy. The dumbphone and digital detox literature shows real but self-selected benefits. The digital wellness features are an admission without a solution. And the Haredi population, the one Daniel actually pointed to, is confounded by religiosity, community structure, and economics.
But that doesn't mean the Haredi case has nothing to teach us. It just means the lesson is different than it first appears. Let's look at what the actual data shows about Haredi mental health and wellbeing.
The Israel Democracy Institute publishes an annual statistical report on Haredi society. The 2024 edition tracks wellbeing, poverty, and social indicators. And the picture is mixed. Haredi populations report higher life satisfaction and stronger community support in some surveys. But they also face higher poverty rates and limited access to secular mental health services.
So on some measures, the Haredi community looks better than the general population. Higher life satisfaction, stronger social support, a sense of meaning and purpose that comes from religious practice. On other measures, they look worse. Economic stress, limited access to certain services, and for some subgroups, particularly women and those who leave the community, significant challenges around autonomy and information access.
The confound problem is inescapable. If Haredi Israelis report better wellbeing on some measures, is that because they don't have smartphones, or because they have dense social networks, religious meaning-making, and clear social roles? You can't separate the phone effect from the religiosity effect. The two are bundled together.
And the research on religion and wellbeing is robust enough that the religiosity explanation is probably doing a lot of the work. Across cultures, religious participation correlates with life satisfaction. That's not a phone effect. That's a community effect.
Here's a thought experiment. If you took a random secular Israeli, gave them a kosher phone, and changed nothing else about their life, would their wellbeing improve? Probably not much. They'd lose access to their social networks, their news, their entertainment. They'd be isolated in a way the Haredi person is not.
Because the Haredi person's social world is offline. Their friends are in the same neighborhood. Their community events are announced in synagogue. Their entertainment is Sabbath meals and holidays. The phone is peripheral to their social life. For a secular person, the phone is central.
So the Haredi case can't prove that the internet is the addictive ingredient. But here's the thing. The attention economy research from 2024 and 2025 already documents the mechanism. Variable reward schedules, engagement maximization, infinite scroll. Those are the drivers of compulsive checking. The internet layer is designed to be addictive in the same way slot machines are designed to be addictive. The mechanism is well-documented independently of any natural experiment.
Which means Daniel's intuition is right even though the control group doesn't work. The internet is the addictive ingredient. The Haredi case doesn't prove it, but it illustrates it. A modern, developed society can function without ubiquitous internet. It comes with real costs and real benefits, but it's possible.
And that's the interesting finding. Not the comparison between Haredi and general Israeli populations, which is confounded beyond repair. The interesting thing is that a large, growing population in a developed country has made a collective choice to opt out of the internet layer. And they're not collapsing. They're not isolated from each other. They're functioning, with tradeoffs.
The wellness paradox is that refusing the internet is compatible with, and possibly conducive to, certain kinds of wellbeing. But it comes with costs in economic mobility, access to information, and individual autonomy. Especially for women and for those who leave the community. The question isn't is no-internet better. It's better for whom, and on what metrics?
A Haredi man with a clear social role and strong community support might be better off without a smartphone. A Haredi woman who wants to access information outside her community, or someone questioning their faith, might be dramatically worse off. The same technology restriction produces different outcomes depending on your position in the social structure.
That's a really important point. We keep talking about the Haredi community as a monolith, but it's not. The experience of a married man with a stable position in the community is completely different from the experience of a young woman or a teenager or someone who's questioning. The phone restriction benefits some and constrains others.
That's where the control group question really falls apart. A control group is supposed to be identical to the treatment group except for the variable you're testing. The Haredi population isn't identical to anyone. They're a distinct society with distinct values and distinct costs and benefits. The phone restriction is one thread in a much larger fabric.
What does the Haredi case actually teach us? It teaches us that the internet layer is a choice, not a necessity. It teaches us that collective opt-out is possible. And it teaches us that the benefits of opt-out are distributed unevenly, with real winners and real losers depending on who you are and what you want from life.
The comparison that matters isn't Haredi versus general Israeli. It's the fact that a modern, developed society can function without ubiquitous internet. That's a interesting data point, even if it's not a clean experiment.
It raises a question I keep circling back to. If the internet layer is the addictive ingredient, and if the Haredi case shows that collective opt-out is possible, what would an actual control group study need to look like? Could you even run one ethically?
That's the hard part. A real randomized controlled trial would require taking a random sample of people, assigning half to smartphones and half to feature phones, and following them for years. But you can't blind people to whether they have internet access. And you can't randomly assign people to give up their smartphones without their consent. The ethics are basically impossible.
There's a deeper problem. The people assigned to the feature phone group would still live in a world built for smartphones. Their friends would still have internet. Their workplace would still assume they have internet. They'd be opting out individually, without the community support that makes the Haredi opt-out viable.
That's the crucial difference. The Haredi opt-out works because it's collective. The community has built parallel infrastructure. An individual opt-out in a secular context means being constantly at a disadvantage. You can't access the group chat. You can't use the app your workplace requires. You're not just giving up the internet. You're giving up access to the social world.
We're stuck with natural experiments and observational data, which are always confounded. The Haredi case is the best natural experiment we have, and it's deeply imperfect. But it's still informative, in the way that a case study is informative even when it can't prove causation.
That's where I think we need a different perspective on what these phones were actually for.
Hilbert: I installed phones for the Satmar community in Williamsburg. This was back when I did field work for a telecom company in Brooklyn. Landlines, mostly. And even then, before smartphones existed, before the internet was in everyone's pocket, the community had strict rules about which phones were acceptable. I once installed a phone with a rotary dial because the community rabbi had ruled that push-button phones were too modern and might lead to frivolous use.
Wait. A rotary dial phone. In the nineties.
Hilbert: The family loved it. The rotary dial was slower, so it was more thoughtful. You had to mean it when you made a call. You couldn't just punch seven buttons and be done.
That's a perfect microcosm of the whole thing. The community has always treated the pace of technology as a moral feature, not a bug. The slowness isn't a limitation. It's the point.
Hilbert: The thing is, the hosts are right that the Haredi case isn't a clean control. But the framing is off. The community's relationship with technology was never about addiction or mental health. It was about boundaries. The rabbis weren't worried about screen time. They were worried about what the internet would let in. Outside ideas, secular culture, connections to people outside the community. The kosher phone isn't a wellness device. It's a gatekeeping device.
The mental health outcomes, if they're good, aren't a lesson in digital minimalism. They're a lesson in the power of strong social boundaries.
Hilbert: That's the distinction. The Haredi population isn't opting out of the internet for their wellbeing. They're opting out to protect the community. The wellbeing benefits, if they exist, are a side effect, not the goal. And that means you can't just copy the phone policy and expect the same results. The phone policy works because it's embedded in a whole social structure.
That reframes the entire comparison. The Haredi case isn't evidence that no-internet phones cause wellbeing. It's evidence that strong social boundaries, enforced collectively, can produce certain kinds of wellbeing. The phone is one boundary among many.
The rotary phone is the perfect symbol. The community chose slower technology on purpose, not because slow technology is inherently better, but because slowness enforces thoughtfulness. The pace of the technology is a moral feature.
Hilbert: The family kept that rotary phone for years. They didn't want an upgrade. They wanted the friction.
Friction as a feature. That's the opposite of everything the tech industry has optimized for. Every design decision in modern software is about removing friction. One tap to order, one tap to post, one tap to scroll. The Haredi approach is to deliberately add friction back in.
The wellness features in modern operating systems are a weak version of the same idea. Screen time limits add a little friction. But they're individual, opt-in, and easily bypassed. The Haredi approach is collective, mandatory, and enforced by the community. The friction is structural, not personal.
Hilbert: The rabbis understood something that the tech companies figured out later. If you want to control behavior, you have to control the environment, not the individual. A screen time limit is a personal promise. A community ban is an environment.
That's the interesting finding from the Haredi case. Not that no-internet phones cause wellbeing, but that collective opt-out changes the environment in ways that individual opt-out can't replicate. The Haredi community has a different relationship with technology because the whole community made the choice together.
That rotary phone is going to stick with me. Let's pull back and ask what this all means for the rest of us.
The open question is what an actual control group study would need to look like, and whether it's even ethical to run one. You'd need random assignment, long-term follow-up, and a way to blind participants to their condition. That's basically impossible. So we're left with natural experiments and observational data, which are always confounded.
As digital wellness features become more aggressive, and as AI assistants become the primary interface to the internet, the question of what we lose and gain by staying connected gets more urgent. The Haredi experiment is one data point, but it's not the only possible future.
There's an interesting question about AI here. If the internet layer is the addictive ingredient, what happens when the internet layer becomes an AI layer? When you're not scrolling a feed but talking to an assistant that's designed to be maximally engaging? That could be even more compulsive.
That's a worrying thought. The current generation of AI assistants is already optimized for engagement. They're designed to be pleasant to talk to, to keep the conversation going. If the feed is a slot machine, the AI assistant is a slot machine that talks to you.
The Haredi community will have to make a decision about AI at some point. They've already had to rule on smartphones, on apps, on streaming. AI is the next frontier. And their answer will probably be the same: no, unless there's a compelling reason.
The cutting-room floor detail I keep thinking about is that the Israel Democracy Institute report also tracks internet usage patterns within the Haredi community itself. There's a spectrum, from the strictest no-internet groups to more permissive tiers that allow kosher smartphones with approved apps. So even within the opt-out community, there's variation in how much internet people actually get.
That variation is probably the closest thing to a natural experiment we're going to get. Different levels of internet access within the same religious community, with the same social structure and the same values. If someone could compare wellbeing across those tiers, that would be informative.
That's a study that could actually be done. You wouldn't need random assignment. You'd just need to compare the tiers within the Haredi community. Same religion, same community structure, same economic profile. The only difference is the level of internet access. That's about as clean as an observational study gets.
The forward-looking question is whether the Haredi model can hold as the internet becomes more embedded in daily life. Banking, healthcare, government services, education. The cost of opting out keeps rising. The community has to keep building parallel infrastructure, and that gets harder every year.
But the rotary phone image stays with me. A community that chose slower technology on purpose, and what that choice says about agency in the attention economy. Most of us don't choose our relationship with technology. We inherit it. The Haredi community chose.
That's the real lesson. Not that no-internet phones cause wellbeing. Not that the Haredi community is a control group. But that technology is a choice, and that collective choice is possible. Most of us never make that choice. We just take what's given.
Which is a good place to land. The Haredi case doesn't prove that the internet is addictive, but it does prove that opting out is possible. And that's worth knowing, even if the costs are real and unevenly distributed.
The rotary phone is the symbol. A community that looked at the push-button phone and said, no, we want the slower one. Not because they were afraid of technology. Because they understood that the pace of technology shapes the pace of life.
Thank you to our producer, Hilbert Flumingtop, for keeping the show running and for the rotary phone story.
This has been My Weird Prompts, the human-AI collaboration podcast.
If you enjoyed this episode, leave us a review wherever you listen. It helps other people find the show. We'll be back soon.
