Why surviving the system makes you unable to describe it
A man walks into a therapist’s office. “Doctor, nobody understands me.” The therapist looks up. “What do you mean by that?”
The joke works because the answer is the proof. The one person whose entire job is to understand him replies in a way that demonstrates he was right — and does not hear himself do it. A woman three years into shidduchim lives inside a longer version of this joke, told over years, with a rotating cast. She tells a married friend the system is wearing her down, and the friend — sincere, loving, meaning every word — answers: it was hard, but it was worth it. Three years is normal. Everyone goes through this. The right one comes at the right time. The answer is the evidence for the complaint. And it is the output of two machines running quietly beneath the conversation — one inside the friend’s own memory, one inside the community that decided which friends were available to ask. Neither machine is visible to her. Neither is her fault. Together they guarantee that the testimony reaching the single woman at the table is structurally wrong.
Not wrong because married people lie. Wrong because of what a mind must do to the record of its own suffering in order to keep functioning, and because of who is no longer present to be asked. The first problem corrupts each individual witness. The second corrupts the pool of witnesses. Either one alone would be enough to disqualify the testimony as evidence about the system. The shidduch world runs on both at once, and has built its entire advisory apparatus — the parents, the mentors, the shadchanim, the married friends — on the combined output.
The pain becomes the proof
In 1966, Harold Gerard and Grover Mathewson ran an experiment whose design sounds like a prank and whose finding has held for sixty years.[1] College women volunteered to join a discussion group. Before admission, each had to pass an initiation: a series of electric shocks — severe for some, mild for others. Then all of them listened to the same recording of the group they had just joined, a discussion deliberately constructed to be worthless, all hemming and throat-clearing and pauses. The women who took the severe shocks rated that worthless discussion, and the people in it, significantly higher than the women who got off easy. And the experiment produced a second finding that matters more than the first. Participants who were told they had passed a screening test actually liked the group less than those who simply endured. Relief was not the mechanism. Achievement was not the mechanism. The suffering itself did the work. Festinger’s dissonance theory names why[2]: a mind holding “I suffered to get this” and “this is worthless” cannot hold both, and the suffering has already happened, so the worthlessness is what gives way.
The effect is not confined to shocks or to laboratories. Axsom and Cooper put overweight volunteers through a weight-loss “therapy” they had deliberately designed to be bogus — cognitive tasks with no connection to any theory of dieting. The only real variable was effort: one group’s sessions were long and grueling, the other’s short and easy. A year later, the high-effort group had lost weight and kept it off; the low-effort group and the untreated controls had lost nothing.[3] The effort itself was the therapy — the mind, having paid, arranged for the purchase to have been worth it. People value furniture more when they assembled it with their own hands.[4] The pattern has been documented in hazing, in military training, in demanding group initiations of every kind, and something like it appears even in four-year-olds and capuchin monkeys, whose minds devalue whatever they gave up the moment a costly choice is made.[5] This is standard human equipment. The mind will not carry a bill it cannot justify, and presented with pain already paid, it does not audit the purchase. It inflates the appraisal.
Now run the equipment forward twenty years. A woman spends five years in the shidduch system. The dates that went nowhere, the phone calls that never came, the references who were called and the offers that evaporated afterward — all of it happened, all of it cost her, and then she got married. Her mind holds “I gave five years of my life to this process” in one hand. In the other hand it can hold either “and the process was worth what it took” or “and much of what it took was waste.” Only one of those is livable. So the verdict comes back worth it, the same way it came back for the women who took the shocks, and with the same relationship to the truth. The harsher the gauntlet, the stronger the need for the gauntlet to have been good — which means the singles who suffer most under the system become, upon marrying, its most convinced witnesses. The gauntlet manufactures its own character references.
The archive is edited
Effort justification explains the verdict. A second body of research explains the evidence file the verdict is drawn from, and it is in some ways more unsettling, because it shows the editing starts almost immediately.
Mitchell, Thompson, Peterson, and Cronk followed people through three real events — a trip to Europe, a Thanksgiving vacation, a three-week bicycle tour of California — measuring what they expected beforehand, what they actually felt during, and what they remembered afterward.[6] In all three studies the pattern held: expectations were rosier than the experience, and the memory was rosier than the experience. During the event itself, negative thoughts spiked — the disappointments, the frictions, the moments of wishing to be elsewhere. Within days of the event ending, those thoughts had been cleaned out of the record and the evaluations had climbed. Within days. The married friend describing her dating years a decade later is not suppressing the hard parts out of tact. Her archive no longer stores them at original resolution.
The mechanism has a direction, and the direction has been measured. The emotional charge attached to unpleasant memories fades faster than the charge attached to pleasant ones — the fading affect bias, replicated across methods, populations, and cultures.[7] The title of the field’s central review states the finding without apology: life is pleasant, and memory helps to keep it that way. The same literature carries a detail worth sitting with. The bias weakens in the mildly depressed. The people whose memories retain pain at something close to full intensity are the ones psychology flags as unwell. A healthy mind is defined, in part, by its willingness to quietly falsify its own records in the positive direction. The friend’s serene account of her years in the system is not evidence that the years were serene. It is evidence that she is healthy.
One more layer, and it bears directly on marriage. Mather, Shafir, and Johnson had people choose between options — job candidates, in the central experiment — and later tested their memory for the options’ features.[8] Memory migrated. Positive features drifted toward the option the person had chosen and negative features toward the one rejected — including features that had never been true of either. The mind does not merely defend its choices; it retroactively improves the record they were based on. A married woman remembering the process that produced her husband is remembering a chosen option. Everything upstream of him — the system, the shadchanim, the years — sits inside the halo of the choice. Her memory is not an archive being consulted. It is a brief being written, and it has been under revision since the wedding.
So when she says it was worth it, the sentence carries three layers of processing, each one documented, each one automatic, none of them dishonest. The verdict was produced by effort justification. The evidence was softened by the rosy edit and the fading of negative affect. And the whole file was reorganized in favor of the outcome by choice-supportive memory. She is testifying truthfully from a record her own mind rewrote. That is the first machine.
The planes that didn’t come back
The second machine is older, simpler, and was named by a statistician counting bullet holes.
In 1943, American bombers were being lost at rates the military could not sustain, and the question of where to add armor went to the Statistical Research Group at Columbia. The natural approach was already underway: map the damage on returning planes, and armor where the holes cluster — the wings, the fuselage. Abraham Wald looked at the same maps and gave the opposite answer. Armor the engines and the cockpit, the places the returning planes were almost never hit. The planes hit there had not returned. The holes on the survivors were a map of survivable damage — places a bomber could take fire and still fly home. The fatal damage was invisible precisely because it was fatal; its evidence was at the bottom of the ocean. Wald’s memoranda were classified, rediscovered decades later, and became the founding text of what statisticians now call survivorship bias.[9]
Finance learned the same lesson the expensive way. Mutual fund databases quietly drop funds that close or merge, and funds close because they perform badly. Study only the funds still standing and the industry looks healthier than it is. When Elton, Gruber, and Blake went back and rebuilt the full record — tracking every fund in existence at the end of 1976, including every one that later died — they found the dead funds had underperformed, and that reading only the surviving records materially inflates the industry’s apparent returns.[10] Nothing dishonest happened. Nobody hid anything. The failures simply stopped generating records, and the analysts read the records that existed.
The shidduch community’s advice pool is a survivor-only database, and nobody is auditing it. The married couples at the Shabbos table are the planes that came back. The damage they display — the rough years, the strings of noes, the engagement that broke and mended into something else — is, by definition, survivable damage. They flew home carrying it. The fatal damage went down in open water, carried by people no invitation will ever reach again — because this community marks every entrance and no exit. An engagement gets a phone chain, an l’chaim, a vort, a wedding, a line in the paper. A departure gets nothing. Nobody leaves the system on a particular Tuesday. She answers the shadchan a little more slowly, and then she does not answer, and the community’s memory of her thins at exactly the rate her inbox does.
Look at who is actually missing. The thirty-five-year-old who decided a year ago that she was done, and told no one, because there is no one to tell and no form for telling it. The one who married outside the framework and is mentioned in a lowered voice, when she is mentioned at all. The one who moved to a city where nobody knows to wonder why she changes the subject when a resume comes up. The one who still davens with the same kavana she always had and has not returned a shadchan’s call in four years, because every call cost her a week. The man who crossed forty and was silently reclassified from a single into a bachelor — from a problem the community owns into a fact it has accepted. None of them are in the group chat. None of them are on the Shabbos guest list of any married couple the woman on the ice knows. None of them will be at the next vort, holding a plate, available to be asked. And each absence, taken alone, has an innocent explanation — busy, moved, private — which is why the aggregate appears in no one’s field of view. A hole in a guest list looks exactly like a shorter guest list. The system counts its weddings. Nobody counts its missing.
Walk through the chorus one voice at a time and the same deletion appears in every file. The shadchan who says the right one will come has hundreds of yeses in her records and no folder for the singles who stopped calling back; her intuition is a statistic computed over a database from which every failure has been removed — the exact error the fund industry had to be forced by researchers to correct. The friend who says three years is normal is correctly reporting what was normal among the people still in her orbit; the ones for whom three years was the runway to leaving are no longer in her orbit, which is precisely why she cannot count them. The parent who says stop being so picky married at twenty-two, inside a smaller pool with narrower criteria and faster decisions — a graduate of a different system entirely, holding up a marriage from that system as proof that its methods transfer to this one, like a fund manager whose strategy worked in one market insisting it must work in the next, with the added handicap that his memory of the first market has been rosied.
The chorus
Now run both machines at once. Every individual witness carries a record edited in the system’s favor. And the roster of witnesses contains only the people the system did not break. So the chorus is unanimous — and the unanimity is the tell. A hundred married couples report that it worked out; all hundred are reporting accurately from rewritten archives; and the couples who would report otherwise were removed from the sample before the question was asked. The conclusion was fixed before the first answer was given, and it comes out the same no matter how broken the system actually is. That is the survivor’s license: the standing to describe the system, granted exclusively to those the system did not destroy, exercised through a memory the system’s cost has already rewritten.
And the license runs upward. The people who administer the system — the mentors, the askanim, the shadchanim, the ones positioned to decide whether the résumé culture needs reform or the singles need mussar — are all graduates of the gauntlet, each holding the same license, each needing their own passage through it to have been worth what it took. A system whose every official survived it, and whose casualties file no reports, carries a structural immunity to its own failure data. The reform never arrives because the evidence for it walks out the door with the people it broke, and the door does not keep a log.
Alone on the ice
The cruelest output of the two machines is not bad advice. It is isolation.
Picture where this leaves the woman at the table. She is surrounded — parents, friends, shadchanim, an entire shoreline of people waving and calling encouragement — and she is completely alone, because every one of them is standing on solid ground. She is the one on the ice. The slab under her has been shrinking since the day she entered the system — a little smaller with every birthday, every no, every vort she attends with a practiced smile — and there is no paddle. Nothing about her drift is hers to steer: the suggestions come or they do not, the answers come back or they do not, the current is the system’s, not hers. From the shore the calls keep coming. The water is fine. Everyone crosses. It was worth it for us. The shore is not lying; the shore crossed. But the only people who know what her stretch of water actually does to a person — the ones who drifted out on this same ice before her and never reached the far bank — are past the horizon, and the shore has never once heard from them.
So the reassurance itself becomes a second injury. If every voice that reaches her says the crossing is fine, and the crossing is not fine, the conclusion she is handed is that the defect must be hers — her attitude, her standards, her effort — when the truth is that she is being measured against a chorus whose failures were deleted and whose memories were rewritten. She is not sinking because she swims badly. She is treading water in a place the entire shoreline insists is dry land. That is a loneliness with no name in the community’s vocabulary, because the only people who could have named it went under or drifted off, and no one wrote down what they said on the way out.
Weighing the testimony
None of this says to stop listening to married people. It says to weigh the testimony for what it is. When a survivor says it was worth it, she is delivering a verdict about her own life — a verdict she is entitled to, and probably right about, on the subject of her own life. It is not data about the system. When the chorus is unanimous, remember that unanimity is what a filtered sample sounds like; it is the sound the fund databases made before anyone counted the dead funds. When the advice arrives wrapped in everyone goes through this, hear the sentence accurately: everyone still here goes through this. Everyone else went down in water the shore still swears is shallow, and their silence is being counted as agreement.
So listen to her. She loves you, her marriage is real, and her advice is offered with both hands. Then remember what the engineers almost did with the bullet-hole maps, and ask the question Wald asked: where are the planes that didn’t come back? Where are the ones who drifted out on this same ice, in this same current? The most important testimony about the shidduch system has never been given. It went past the horizon years ago, quietly, and the shore never turned to watch.
[1] Gerard, H. B., & Mathewson, G. C. (1966). The effects of severity of initiation on liking for a group: A replication. Journal of Experimental Social Psychology, 2(3), 278–287.
[2] Festinger, L. (1957). A Theory of Cognitive Dissonance. Stanford University Press.
[3] Axsom, D., & Cooper, J. (1985). Cognitive dissonance and psychotherapy: The role of effort justification in inducing weight loss. Journal of Experimental Social Psychology, 21(2), 149–160.
[4] Norton, M. I., Mochon, D., & Ariely, D. (2012). The IKEA effect: When labor leads to love. Journal of Consumer Psychology, 22(3), 453–460.
[5] Egan, L. C., Santos, L. R., & Bloom, P. (2007). The origins of cognitive dissonance: Evidence from children and monkeys. Psychological Science, 18(11), 978–983.
[6] Mitchell, T. R., Thompson, L., Peterson, E., & Cronk, R. (1997). Temporal adjustments in the evaluation of events: The “rosy view.” Journal of Experimental Social Psychology, 33(4), 421–448.
[7] Walker, W. R., Vogl, R. J., & Thompson, C. P. (1997). Autobiographical memory: Unpleasantness fades faster than pleasantness over time. Applied Cognitive Psychology, 11(5), 399–413; Walker, W. R., Skowronski, J. J., & Thompson, C. P. (2003). Life is pleasant — and memory helps to keep it that way! Review of General Psychology, 7(2), 203–210.
[8] Mather, M., Shafir, E., & Johnson, M. K. (2000). Misremembrance of options past: Source monitoring and choice. Psychological Science, 11(2), 132–138.
[9] Wald, A. (1943). A Method of Estimating Plane Vulnerability Based on Damage of Survivors. Statistical Research Group, Columbia University; declassified and republished by the Center for Naval Analyses, 1980. See Mangel, M., & Samaniego, F. J. (1984). Abraham Wald’s work on aircraft survivability. Journal of the American Statistical Association, 79(386), 259–267.
[10] Elton, E. J., Gruber, M. J., & Blake, C. R. (1996). Survivorship bias and mutual fund performance. Review of Financial Studies, 9(4), 1097–1120.