Children Become Who They Play With

When I was a kid I had a friend who lived nearby. He wasn’t Jewish. He was, by any measure, a good kid — kind, easy to be around, more mature than his age. We played the way children have always played. Hide-and-seek with walkie-talkies and some elaborate set of rules I no longer remember. Pretend games. Pulling bamboo out of the strip behind his backyard, which for some reason felt like the best thing in the world to do on a given afternoon.

I never felt strange around him. The Jewish thing never really came up — not because it was hidden, but because it had no occasion to. We were two boys with bamboo and walkie-talkies. That was the whole of it.

The friendship ended for the most ordinary reason there is. Someone moved. There was no falling out, no drift, no moment where I felt something pulling at me that shouldn’t have been. It just ended, the way a thousand childhood things end. I think about him sometimes. Maybe we’ll cross paths again.

That friendship cost me nothing Jewish. So when frum parents — and especially Chassidic parents — worry about their children having non-Jewish friends, is the worry overblown? Is it good, bad, or something else entirely?

It is none of those. “Good or bad” is the wrong frame. It depends on what you’re optimizing for and at what stage — and maybe it’s not about the children at all.

The concern, without theology

Strip out every religious word and the concern still stands. The principle comes from secular developmental psychology, not from us:

Children become who they play with.

Judith Rich Harris made this case forcefully in The Nurture Assumption. Her argument — strong enough that developmental psychologists had to take it seriously whether or not they fully accepted it — is that the peer group, not the parents, is the primary vector by which children actually internalize values. If she is even partially right, then a child’s peer environment is not one factor among many. It may be the factor, operating largely beneath whatever parents teach at the dinner table.

That is not xenophobia. It is not “goyim are bad.” It is a claim about how identity forms in young, plastic human beings: they absorb the world they are placed in. The question of who your child plays with stops being a question of social manners and becomes a question of identity formation.

Japan, sealed and opened

The cleanest laboratory we have is Tokugawa Japan. For roughly two and a half centuries — the sakoku period, from 1603 to 1868 — Japan maintained a deliberate cultural quarantine. This was not backwardness or fear of the unfamiliar. It was policy, chosen by a sophisticated society that had watched what European contact was doing to its neighbors and decided it wanted no part of the transformation.

The quarantine held for two hundred and fifty years. Then in 1853 Commodore Perry sailed into the harbor and forced the ports open, and the speed of what followed should frighten anyone who thinks culture is sturdy. Within a single generation, dress, diet, family structure, social hierarchy, religious practice — all of it destabilized. The Meiji elites actually chose the disruption and drove it as state policy; the common people simply absorbed it.

The dramatic shifts people like to cite — collapsing birth rates, youth delinquency — are most cleanly documented in the post-war American occupation wave, not in Meiji. Two distinct waves, one underlying phenomenon. Sustained contact does not merely add to a culture. It reorganizes it. The receiving society does not end up as itself-plus-some-new-things. It becomes something structurally different.

The immigrant pattern

If Japan shows the phenomenon under political pressure on adults, the immigrant story shows it happening quietly, voluntarily, through children — and it is one of the best-documented patterns in all of sociology.

Milton Gordon’s framework breaks assimilation into stages: cultural first (language, dress, custom), then structural (you enter the institutions and social networks of the majority), then marital (intermarriage), then identificational (your very self-concept changes). His central finding was that the stages cascade. Once the next generation enters majority-culture peer networks, structural assimilation follows almost inevitably — and once that happens, intermarriage tends to follow within another generation.

The arc is almost monotonous in its consistency:

• The first generation holds — native language, old customs, a tight internal world.
• The second is bilingual and bicultural and quietly torn.
• The third speaks only the majority’s language and is left with nostalgia that has no content.

It is so reliable it has a name. Hansen’s Law: what the son wishes to forget, the grandson wishes to remember. And there is a real cruelty buried in it. The grandson wants the heritage back — but the living chain that could have transmitted it is gone. He has the longing and not the thing.

Italian-Americans: by the third generation, the Catholicism often nominal, the language nearly extinct, the endogamy collapsed. German-Americans: before the First World War the largest non-English ethnic bloc in the country, with their own papers and churches and schools — essentially erased within two generations. Non-Orthodox Eastern European Jews: the Reform and Conservative movements can be read, in part, as a real-time record of this very process, each institutional accommodation tracking how far peer-network integration had already gone.

What actually drives it

The mechanism underneath all of it is not geography. It is not even persecution. It is who the children’s functional peers are.. Once the density of outside-group peers crosses some threshold, the downstream cultural outcomes become predictable and, within a generation or two, largely irreversible. Japan shows it can happen even to adults under pressure. The immigrant story shows it happening on its own, gently, through the natural social gravity of children toward the majority world around them.

The Jewish anomaly

Every model above predicts that we should be gone.

By Gordon’s stages and Hansen’s Law, a landless people scattered across foreign majorities for millennia should have dissolved many times over. Instead: Babylon, and no dissolution into Babylonian life. The Persian empire. The Hellenistic world — where the Maccabean story is, read plainly, a story about resisting cultural assimilation at the very peak of Greek cultural prestige. Rome. The Christian medieval world and the Islamic one. Emancipation-era Europe, where every prediction said total assimilation was coming — and where it did come for the Reform-trending, but the traditional communities held.

No other diaspora people maintained coherent identity across that time span without a territorial base. Historians do not treat this as a triumphant footnote. They treat it as a genuine problem to be explained.

The explanations they reach for are not one silver bullet but several reinforcing structures. Halacha functions as a behavioral firewall — kashrus, Shabbos, family purity laws make deep social integration structurally difficult without anyone needing to stay ideologically vigilant; you simply cannot fully share the table and the calendar of a society whose food you can’t eat and whose central day you keep differently. Jacob Katz documented this carefully. Then there is text — identity anchored to a portable corpus rather than to soil or to a narrow ethnicity. The Torah goes everywhere; ask what the portable Italian anchor is once you’ve left Italy and the honest answer is cuisine and some customs, nothing with the structural weight of a learning tradition that re-inhabits its own past. There is endogamy, which in the sociological literature is simply the single most powerful long-term preservation variable there is; once intermarriage climbs past roughly a third, communities tend to dissolve within two generations, and traditional Jewish communities held near zero for centuries. And there is persecution — real, but, as Katz is careful to note, insufficient on its own. Persecution can just as easily produce apostasy or despair. What it mostly did was close the exit on the host side, reinforcing a separation halacha was already producing. The proof is in what happened when emancipation finally opened the exit: the non-Orthodox left quickly. The persecution had been doing some of the work all along.

The enclave, and the Chabad twist

The critical variable was never geographic proximity. Jews lived in Rome. In Baghdad. In Amsterdam. Physical closeness to the majority was the normal condition, not the exception. What varied — what determined whether a community held or dissolved — was institutional density and peer-network control. Schools, courts, marriage networks, batei midrash. When those were strong, the functional peer world stayed internal even when the streets outside were not. The walls of the Frankfurt Judengasse were as much social and institutional as physical, and what destabilized things at emancipation was not that the geography changed but that the institutional barriers fell.

Now look at Chabad. We live integrated. Most of us have non-Jewish neighbors on every side; we do not live in sealed-off Jewish-only zones. And yet our school, our shul, our children’s playdates run something like ninety-five percent internal. We are, in effect, an enclave inside an integrated geography. The institutional world is dense even though the physical world is mixed.

We are not unique in having found this structure. Overseas Chinese communities across Southeast Asia held cohesion for generations the same way — live among the majority, but school and marry and worship and do business inside the network — and where it broke down, it broke down when the institutions weakened, not when the geography mixed. The Armenian diaspora rebuilt identity after catastrophe across wildly different host countries on church, school, and endogamy. The Parsis of India held a distinct identity for over a thousand years on strict endogamy and community institutions — and are now collapsing demographically, because endogamy plus low birth rates plus some loosening of marriage norms turned out not to be sustainable. And the Old Order Amish may be the closest structural parallel of all: geographically integrated (they farm among non-Amish), institutionally sealed (no outside schooling past the eighth grade), with rumspringa as an explicit controlled-exposure mechanism — after which something like eighty-five to ninety percent return. Childhood identity formation can be robust enough to survive even a deliberate adolescent experiment in exposure.

Where this leaves us

A Chabad kid with a non-Jewish friend is not a Jew dissolving into a majority culture. The weight of his social world still sits, overwhelmingly, inside the community — the school, the shul, the Shabbos table, the friends who are also at all of those. That weight is exactly the institutional density that held Jews together in Baghdad and Amsterdam. It genuinely mitigates the risk.

Which means the question was never really binary. It is not “non-Jewish friends: yes or no.” It is quantitative and structural: how much weight sits where? And once you ask it that way, a second thing becomes obvious — “a non-Jewish friend” is not one situation at all. It is several very different situations that we lazily call by one name.