A quote card floated past me on Instagram. Dostoevsky’s face, that heavy beard, and the line stamped underneath: two intelligent people cannot fall in love; true love needs one idiot. It’s built to make you nod. And the engineer in me wanted to nod, because it flatters exactly the kind of person who scrolls past quote cards feeling faintly superior to them. Then I did the thing the card is daring you not to do, and actually thought about it.

Dostoevsky never wrote it, by the way. It’s a modern aphorism wearing his face, the way half of Instagram’s “Dostoevsky” is. But I don’t really care that it’s fake. Fake quotes are a dime a dozen. What caught me is that this one is making a claim, an actual empirical claim about smart people and love, and the claim is testable. So is it true? If you’re the clever one in the room, or you keep ending up with people who are, are you quietly doomed?

Short version: the quote is wrong about intelligence and weirdly right about judgment. Let me show you the gap, because the gap is the interesting part.

Smart people pair with smart people. That’s the default.

Start with the thing the card gets backwards. If two intelligent people supposedly can’t fall in love, you’d expect intelligence to be scattered randomly across couples, or for smart people to keep pairing off with their opposites. Neither is true. Intelligence and education are among the traits where partners resemble each other most. Spouses correlate somewhere around 0.45 to 0.51 on education,¹ and couples measurably line up on IQ too.² “Likes attract” is just the boring, well-documented default.

And the whole “opposites attract” idea the card leans on? It mostly isn’t real. A 2023 study paired a meta-analysis of 22 traits with a look at 133 traits across tens of thousands of couples, and the picture is lopsided in one direction: almost everything correlates positively between partners. Out of 133 traits, exactly three came out negative. The strongest match of all wasn’t looks or height, it was shared political and religious values, at around r = 0.58. Education sat high too, and IQ landed around 0.44.³ People pair up with people like them, and “like them” very much includes how their mind works.

So as a literal statement, “two intelligent people cannot fall in love” is exactly backwards. Two intelligent people falling in love is the default case, the thing the data expects.

Two caveats I owe you before I lean on this too hard. The data measure who pairs up and marries, which is the closest thing we can actually count to who falls in love. And nearly all of it comes from Western samples (Finnish, Dutch, British, American), so I’m really describing the world those studies can see, not an arranged marriage in a village. Within those limits, though, the pattern is about as solid as social science gets.

But there’s an honest question hiding under the dumb one.

Here’s where I have to be fair to the card, because there’s a real question tangled up in the silly one: what happens when the intelligence is lopsided? When one person is clearly smarter than the other?

The internet’s favorite answer is a stat that gets passed around as fact: something like “for every 16-point rise in IQ, a woman’s odds of marrying drop 40%, while a man’s go up 35%.” I went looking for where that comes from, because it sounds too clean. It traces back to one real study, a Scottish cohort born in 1921 whose childhood IQs were measured at age 11 and then checked against who’d married by mid-life. The actual finding is directional and sex-split: higher childhood IQ went with more marriage for men and less for women.⁴ But the crisp “40% / 35% per 16 points” isn’t the study’s number at all. It appears to trace to 2005 UK press coverage that turned the study’s odds ratios into round percentages, and it’s been quoted as gospel ever since.⁵ So: real signal, but from one dated cohort raised under the marriage norms of the 1950s, and dressed up in a statistic the researchers never actually printed. Handle with tongs.

And the bigger problem with treating that as a law: the thing underneath it is dissolving. Let me be honest about the move I’m making, because it’s the hinge of this section. I’m treating IQ, education, and earning as three stand-ins for one signal, status, and the strongest evidence I have is specifically about education and divorce, not IQ and marrying. With that caveat said out loud: the lived version of “one of us is smarter” has always been mostly about schooling and earning, and that gap has flipped. Marriages where the wife is the more educated one used to be the ones most likely to break up. In recent US data that penalty has straight-up vanished, and those marriages went from about a fifth of the total to over 30%.⁶ What looked like a law of attraction was a social convention with an expiry date, and the date has passed.⁷

The twist: intelligence is the wrong variable.

Here’s the part that actually rearranged how I think about it. We keep arguing about intelligence because it’s the variable the meme handed us. But when you ask what actually predicts whether love lasts, intelligence barely shows up. Temperament does.

Ty Tashiro, who wrote a whole book on this, puts emotional stability at the top of the list, kindness right behind it, and notes that we systematically overrate the flashy stuff, looks and money, while underrating the quiet trait that does the real work: not being a reactive, anxious, moody partner.⁸ (I’ll flag one thing, because I went and checked: the viral “86% of a marriage’s success is who you pick” number that gets pinned to this book is a mangled search-engine summary. The real 86% is how satisfied couples are in year one, before it slides under 50% by year seven.⁹ The “who you pick matters more than how hard you work” idea is genuinely his; that particular number isn’t.)

Now put that next to the word the card actually leans on: overthinking. The unspoken assumption is that smart people overthink, and overthinking corrodes love. But the corrosive kind of overthinking has a name in the research, and it isn’t intelligence. It’s rumination, and rumination is a flavor of neuroticism, not intellect. Psychologists literally split “thinking a lot about yourself” into two different dispositions: rumination, the anxious, threat-driven kind that tracks with neuroticism and distress, and reflection, the curious kind that tracks with openness.¹⁰ Same surface behavior, two different engines. And across the personality research, neuroticism is the single trait most reliably tied to being unhappy in a relationship, at around r = −0.22, well ahead of the rest.¹¹ Nobody has found the equivalent for IQ. Nobody has found a comparable, well-replicated effect for IQ. The meme takes “anxious and can’t stop appraising” and mislabels it “smart.” They are not the same thing, and conflating them is the whole error.

And yet the card fingers something real.

So the meme is empirically busted. Smart people pair with smart people, the asymmetry is a fading social fact, and what wrecks love is temperament, not IQ. I could stop there and feel clever. But that would be dodging the reason the line lands on people, which is that it’s pointing, clumsily, at something true.

Because love does require you to switch something off. And this isn’t only a metaphor, you can watch a version of it on a brain scan. In a small fMRI study, people who were deeply in love showed deactivation in a set of regions the brain normally leans on for negative judgment, for critical social assessment, for “mentalizing,” the machinery for sizing up other people’s intentions and flaws.¹² Love runs on a push-pull: it dials down the circuits that would ordinarily be busy appraising the other person. Semir Zeki, one of the researchers, put it about as bluntly as a scientist can: the passion of love is mirrored by “a suspension of judgment… a relaxation of judgmental criteria by which we assess other people.”¹³

The philosophers got there a couple of centuries early, without a scanner. Stendhal called it crystallization, and his image for it is perfect. Throw a bare winter branch into an abandoned salt mine, leave it a couple of months, and pull it back out “covered with brilliant crystals,” every twig glittering, “impossible to recognise the original branch.” That, he said, is the mind in love: it keeps drawing “new perfections in the object of its love” out of thin air.¹⁴ Schopenhauer, grumpier about it, said the whole thing is the species running a con on the individual, “implanting a certain illusion” so that the lover serves the species while thinking he’s serving himself. Then the spell lifts, the lover “finds himself cheated,” and sees that “the individual was here the dupe of the species.”¹⁵ Different mood, same mechanic: love requires the appraising intellect to be, briefly, fooled. Even Pascal, doing the older reason-versus-heart version: “the heart has its reasons, of which reason knows nothing.”¹⁶

That’s the kernel the card is groping at. The “idiot” it’s praising was never about a low IQ. It’s the capacity to stop appraising. To idealize on purpose. To commit before the proof is complete, which is the only way anyone ever commits, because the complete proof never arrives.

So what’s the actual failure mode?

If love needs a suspension of judgment, you can see how someone gets into trouble: by never suspending it. By staying in appraisal mode past the point where appraisal helps, running the cost-benefit one more time, waiting for a certainty that never quite arrives.

And notice this has nothing to do with IQ. Appraisal is a capacity anyone can overuse. What turns “I’m still weighing it” into a permanent condition is temperament, the same neurotic rumination from before, not raw horsepower. A secure, brilliant person leaps. An anxious one keeps grading, at any IQ. The meme watches someone grade and grade and assumes the grading is the intelligence, so it prescribes a dumber partner. Wrong diagnosis.

Which is exactly where Alain de Botton lands the correction, and it’s the sentence that made all of this click for me. The problem was never thinking too much, he says. You can’t think too much about love, you can only “overthink badly.”¹⁷ The fix for the over-appraiser isn’t to get dumber. It’s to think well: to accept that “compatibility is an achievement of love; it must not be its precondition,”¹⁸ and that the person worth choosing isn’t the one who ticks every box but the one who’s “good at disagreement,” good at negotiating difference. Maturity here is knowing when the analysis has done its job and it’s time to leap.

So the cure for the smart person in love isn’t the one idiot the meme prescribes. You don’t need to import a dumb person to balance you out. You need to be able to run your own intelligence and then, at the right moment, set it down.

There’s a nice irony to end on. Dostoevsky, whose face got borrowed for a quote he’d have hated, actually wrote the great novel about the holy idiot in love: Prince Myshkin, in The Idiot, a man so guileless the world reads his goodness as stupidity. And Dostoevsky’s real line on love, the one he actually wrote, is the opposite of the card’s cheap little surrender. “Love in action is a harsh and dreadful thing compared with love in dreams.”¹⁹ Not a suspension of effort. Labor. The card wanted an idiot who feels less and thinks less. Dostoevsky’s idiot feels more, and does the work anyway.

Maybe that’s the version worth keeping. Love doesn’t need you to be stupid. It needs you to be brave enough to stop grading, and then stubborn enough to stay.

References

Footnotes

1. Gonggrijp et al. (2023). The mechanism of assortative mating for educational attainment, Frontiers in Genetics. Spousal correlations for education were 0.51 in Finland and 0.45 in the Netherlands, with education and intelligence among the traits showing the strongest partner resemblance. ↩

2. Escorial & Martín-Buro (2012). The role of personality and intelligence in assortative mating, The Spanish Journal of Psychology 15(2). Across 244 couples, “strong assortative mating for intelligence” (spouse correlations around .41 to .45), much higher than for personality. ↩

3. Horwitz, Balbona, Paulich & Keller (2023). Evidence of correlations between human partners based on systematic reviews and meta-analyses of 22 traits and UK Biobank analysis of 133 traits, Nature Human Behaviour 7(9). Partner correlations were overwhelmingly positive; of 133 UK Biobank traits only three were negative. The strongest was political values (r ≈ .58); educational attainment r ≈ .55 and IQ r ≈ .44. ↩

4. Taylor et al. (2005). Childhood IQ and marriage by mid-life: the Scottish Mental Survey 1932 and the Midspan studies, Personality and Individual Differences 38(7). In an early-20th-century cohort, per 1 SD of childhood IQ the odds of ever marrying were 1.35 for men and 0.42 for women: higher IQ meant more marriage for men, less for women. ↩

5. Women, IQ and Marriage (2005), Marginal Revolution, echoing 2005 UK press coverage of Taylor et al. The widely-quoted “40% drop for women / 35% rise for men per 16-point IQ increase” is a newspaper translation of the study’s per-standard-deviation odds ratios, not a figure the paper itself reports. ↩

6. Schwartz & Han (2014). The reversal of the gender gap in education and trends in marital dissolution, American Sociological Review 79(4). US marriages where the wife was more educated were once more divorce-prone, but that association “has disappeared in more recent marriage cohorts”; such marriages rose from about 20% (early 1970s) to over 30%. ↩

7. Van Bavel, Schwartz & Esteve (2018). The reversal of the gender gap in education and its consequences for family life, Annual Review of Sociology 44. Review synthesis: marriages in which the wife has more education “were more unstable in the past… this is no longer true.” ↩

8. Ty Tashiro, The Science of Happily Ever After (2014). Tashiro ranks emotional stability (low neuroticism) as the top partner trait for lasting satisfaction, with agreeableness/kindness next, and argues people overvalue physical attractiveness and money while undervaluing these traits; intelligence is not among his recommended priorities. ↩

9. Eric Barker, “The Science Of Happily Ever After: 3 Things That Keep Love Alive” (2014), TIME, summarizing Tashiro. The genuine “86%” is first-year marriage satisfaction, which falls under 50% by year seven; the “86% of success = partner selection” phrasing that circulates online is a search-engine artifact, not a claim in the book. ↩

10. Trapnell & Campbell (1999). Private self-consciousness and the five-factor model: distinguishing rumination from reflection, Journal of Personality and Social Psychology 76(2). Self-focused thought splits into rumination (a neuroticism-linked, distress-associated disposition) and reflection (an openness-linked, curiosity-driven one); the corrosive kind is the neurotic one, not the intellectual one. ↩

11. Malouff et al. (2010). The Five-Factor Model of personality and relationship satisfaction: a meta-analysis, Journal of Research in Personality 44(1). Across 19 samples, neuroticism was the strongest Big-Five correlate of (dis)satisfaction at r = −.22, ahead of agreeableness (.15), conscientiousness (.12) and extraversion (.06). ↩

12. Bartels & Zeki (2004). The neural correlates of maternal and romantic love, NeuroImage 21(3), and the UCL press release. Both forms of love deactivated regions tied to negative emotion, social judgment and “mentalizing,” via a push-pull that “deactivat[es] networks used for critical social assessment.” ↩

13. Zeki (2007). The neurobiology of love, FEBS Letters 581(14). Zeki describes the passion of romantic love as “mirrored by a suspension of judgment or a relaxation of judgmental criteria by which we assess other people, a function of the frontal cortex.” (Quote confirmed via reproductions; the primary PDF is paywalled.) ↩

14. Stendhal, On Love (1822), trans. Woolf & Woolf, Ch. II. A bare bough left in the Salzburg salt mines emerges “covered with brilliant crystals… impossible to recognise the original branch”; crystallization is “the operation of the mind which, from everything which is presented to it, draws the conclusion that there are new perfections in the object of its love.” ↩

15. Schopenhauer, The Metaphysics of the Love of the Sexes, trans. Haldane & Kemp, Ch. XLIV of The World as Will and Idea. Love works by “implanting a certain illusion in the individual” so that serving the species feels like serving oneself; when it lifts, “the individual… was here the dupe of the species.” ↩

16. Pascal, Pensées, Lafuma 423 / Brunschvicg 277: “The heart has its reasons, of which reason knows nothing.” (In context Pascal is writing about faith, not romance, but the line has long outgrown its original subject.) ↩

17. Alain de Botton, “Why You Will Marry the Wrong Person” (talk transcript). “You can’t think too much. You can only overthink badly. But there is no such thing as thinking too much about emotions.” ↩

18. Alain de Botton, “Why You Will Marry the Wrong Person” (2016), The New York Times. “Compatibility is an achievement of love; it must not be its precondition,” and the person best suited to us is “the person who can negotiate differences in taste intelligently, the person who is good at disagreement.” ↩

19. Dostoevsky, The Brothers Karamazov, Book II ch. 4, trans. Garnett. Father Zosima: “Love in action is a harsh and dreadful thing compared with love in dreams.” ↩