This is well said:

It’s possible these respondents are “overthinking” their response to what, on the surface, is a very straightforward question: Am I attracted to this person or not? Indeed, some would argue that there’s no reason to even explain: You can’t argue with your genitals.

But maybe what we call the argument of one’s genitals is, in truth, incredibly — and both consciously and subconsciously — influenced by the cultures in which we grow up as well as our distinct (and equally culturally influenced) ideas of what a “good couple” or “good relationship” would look like. Put differently, we swipe because someone’s “hot,” but we find someone “hot” based on unconscious codes of class, race, education level, religion, and corresponding interests embedded within the photos of their profile.

Essentially, we’re constantly inventing narratives about the people who surround us — where he works, what he loves, whether our family would like him. And more than other dating services, which offer up comprehensive match dossiers, Tinder appears to encourage these narratives and crystallize the extrapolation process and package it into a five-second, low-stakes decision. We swipe, in other words, because of semiotics.

“Semiotics” is, quite simply, the study of signs. The field of semiotics tries to figure out how we come up with symbols — even as simple as the word in front of you — that stand in for a larger concept. Why does the word “lake” mean that massive blue watery thing? Or how does the stop sign, even without the word “stop,” make everyone understand not to go forward?

But signs aren’t always static in their meaning — it’s all about context. Wearing a camouflage jacket can mean that you’re in the military, a hunter, a punk, a redneck, a misogynist; having a shaved head, as a girl, can connote that you’re a radical, a cancer survivor, or a lesbian.

The article has lots of fascinating examples like this one:

The most swipeable woman — no matter if the user identified as straight, gay, queer, or bi — was Yasmin, with an 89% swipe-yes rate, a full 10% higher than her closest “competitor.”

But why? She signified as middle-class (85% believed so); she seemed as if she had finished a four-year college degree or higher (83%). She looks Christian (42%), spiritual (20%), or agnostic/atheist (17%), and reads as either “mixed race” (48%) or black (40%).

Look closer at this image: Yasmin’s teeth are white and straight and her skin is clear. Her shirt is nondescript, but doesn’t read, at least from what we can see of it, as “cheap.” The contrast between the shirt color and house in the background makes her look crisp and clean. Her overarching look is bourgeois, like a model in an issue of Real Simple.

Her eyes are “smizing,” which makes it seem like she’s actually happy, not just posing for the camera, all of which combines to create a feeling of “genuineness.” Her hair seems only the slightest bit unruly — hey, she’s not uptight! — but is also well-conditioned and cared for. She probably has means; she is content; she is educated; you will have something to talk to her about, and she will be pleasant.

But perhaps the most attractive thing about Yasmin, at least according to the simulation, is that her race is ambiguous. In his new book Dataclysm: Who We Are (When We Think No One’s Looking), OkCupid co-founder and data scientist Christian Rudder asserts that “when you’re looking at how two American strangers behave in a romantic context, race is the ultimate confounding factor.” Working with star ratings and messaging data, Rudder found “two essential patterns” of male to female attraction: First, men tend to like women of the same race; second, men “don’t like” black women.

The conclusion: Tinder rejection in this simulation appears to be more about class than race or religion.

If a user self-identified as upper-middle-class and identified the male profile before him or her as “working-class,” that user swiped “yes” only 13% of the time; if they identified themselves as lower-middle-class, the swipe rate rose only slightly to 17%.

If those same users identified the profile before them as middle-class, that number rose to 36% and 39%, respectively. The same trend held true when judging female profiles: If the user identified as upper-middle-class and identified a profile as working-class, the yes rate was 26% — compared with 52% if they identified a profile as middle-class.

Whatever the signs that made someone think that a profile was working-class — McKenzie’s fishing pole, Renee’s dye job and pool pose, Ricky’s tattoos and piercings, John’s tank top, Toby’s camo, Jimmy’s truck — the swipe rates plummeted.

Which isn’t to suggest that poor people are ugly. The vast majority of explanations for the no swipes on all of the above profiles pointed to a perceived lack of common interests: “we’d have nothing to talk about,” “I don’t think our politics would mix,” “nothing in common.” Sometimes those assumptions stem from depicted activities — fishing, body modifications — but some are just the way the mind runs wild with class, weaving the narrative that a working-class person probably doesn’t read books for pleasure, or enjoy art cinema, or seek out microbrews, or go on hikes the way a bourgeois, middle-class person does.

Now, the results of a small sample-size Tinder simulation doesn’t mean that we’re all destined to marry within only our own classes. Data on the tendency to marry within one’s class is difficult to come by, but if relying on education level as an (imperfect) proxy for class, then the rate has decreased dramatically over the 50 years. Even as more and more people marry “across” lines of race and religion, fewer and fewer are willing to cross the education/class line.

Tinder is by no means the cause of this decline. It simply encourages and quietly normalizes the assumptions that undergird it. The Tinderspeak of “we’d have nothing in common,” taken to its natural extension, bolsters and reifies the idea of “two Americas” with distinct values and worldviews, two discrete factions with little impetus to support that which doesn’t necessarily personally affect us or our class.

It’s not as if race and religion aren’t still mitigating factors in our decisions about whom we find attractive, with whom we emphasize, or for whom we feel compassion. Race and religion do matter (and might always), but almost only when they intersect with a class identity that isn’t our own.

Ultimately, this admittedly un-randomized sample seems to suggest that the raw idea of attraction — that knee-jerk “thinking from the genitals” decision — has less to do with our unmentionable parts and much more to do with a combination of our deepest subconscious biases and with our most overt anduncharitable personal politics. And if that’s the case, it’s no doubt the reason why Tinder is so popular, addictive, and ultimately insidious.