The literal body politic? Appearance and political affiliation
March 15, 2021 2:16 AM   Subscribe

Democrats and Republicans Can Be Differentiated from Their Faces because Political Attitudes Vary with Physiological Traits. That may sound far-fetched, but associating physiology with ideology is consistent with findings from behavior genetics that political orientation exhibits genetic heritability. See Genetic Variation in Political Participation and Are Political Orientations Genetically Transmitted?
posted by BadgerDoctor (45 comments total) 13 users marked this as a favorite
 
Neophrenology is a thing now? Huh.
posted by flabdablet at 2:51 AM on March 15, 2021 [68 favorites]


The headline for the first study is that Dems and Repubs can be "differentiated from their faces" but that is not what the study looks at. The study is of pictures of people that the people chose to represent themselves with. How much of the differentiation is down to choices of pose, expression, quality etc. rather than physical differences? (This is not me calling for the return of callipers by the way, just wariness that we might once again be tempted to assign to nature that which is culture.)
posted by Gratishades at 3:30 AM on March 15, 2021 [65 favorites]


I think the linked articles are fairly interesting, but I think the framing of this post is problematic, because the first thing idea this evokes is indeed something like neophrenology.

The first study is really not about the physiology of the faces, it's about what I would call countenance. From the abstract:

Republicans were perceived as more powerful than Democrats. Moreover, as individual targets were perceived to be more powerful, they were more likely to be perceived as Republicans by others. Similarly, as individual targets were perceived to be warmer, they were more likely to be perceived as Democrats.

This is really about how politicians and college students express themselves in professional photos, which I think is actually rather interesting.

Edit: What Gratishades said.
posted by Alex404 at 3:32 AM on March 15, 2021 [30 favorites]


Previously.
posted by TheophileEscargot at 3:37 AM on March 15, 2021 [2 favorites]


I think these tiny studies (29 undergraduates and 46 adults among the first and second link) would have made their attempted points a lot more convincingly if they'd stepped out of American politicians for the study and used, say, pictures of Canadian politicians or another country with similar "liberal"/"conservative" parties as ours -- or used topics about more unfamiliar subjects.

Call me "triggered", but yes, after going through the 2000s, I will start sweating more and blinking heavily when I see the terms "Patriot Act", "Iraq War", and "Biblical Truth". And I might not be surprised that many self-identified Republicans do the same when the very concept of "gay marriage" gets mentioned. I doubt we're born that way.
posted by Theiform at 3:45 AM on March 15, 2021 [2 favorites]


Yeah, I'm guessing this is less about "we can tell this thing about people from their faces!" and more about how AI/ML learns to draw conclusions based on the training data.
posted by rmd1023 at 3:52 AM on March 15, 2021 [1 favorite]


"Undergraduates at Tufts can identify politicians from the most recent US senate election from CNN photos at levels slightly above chance."
posted by you at 3:53 AM on March 15, 2021 [40 favorites]


This does feel like it's subject to the WEIRD effect - that whatever interesting effects you find amongst American undergraduates disappear when you move further away from that very unusual and specific group.
posted by Merus at 3:55 AM on March 15, 2021 [14 favorites]


In America, “Liberal” and “Conservative” are tribal identities first, and philosophical positions only after that (if at all; in reality, groups can rationalise positions diametrically opposed from their ostensible philosophical values, such as conservatives demanding extensive state control of social media to fight “cancel culture”). These tend to get conflated with all sorts of other signifiers of membership, from make-up to iconography. While some of this is explicit (such as slogan T-shirts or hair dye), it would make sense that more subtle aspects would be passed on by cultural transmission.

I'm wondering whether one can differentiate between first- and second-generation members of an ideological tribe; i.e., a liberal child of Californian hippies vs. a liberal whose family are red-state evangelical Republicans.
posted by acb at 4:24 AM on March 15, 2021 [8 favorites]


Each photo was cropped to the extremes of the targets' heads (top of head, bottom of chin, sides of hair or ears), converted to grayscale, and standardized for size. To avoid race-based stereotypes, racial minority candidates were excluded from the study. In total, there were 118 candidates: 59 Democrats (n = 15 women) and 59 Republicans (n = 5 women).

Drat. I was going to use this same method in my upcoming book, "A student's field guide to making political generalizations based on white, mostly male faces".
posted by Avelwood at 4:36 AM on March 15, 2021 [31 favorites]


To avoid race-based stereotypes, racial minority candidates were excluded from the study.

In order to save the village it was necessary to destroy it.
posted by flabdablet at 5:26 AM on March 15, 2021 [42 favorites]


The article won't load here (work PC) but I wonder if they account for people like me whose political leanings have changed dramatically over the years. Thirty years ago when I was young I was a hardcore conservative but today Democrats would likely call me a socialist (although my views are a little more complicated than that, but you get the idea.) Has my face changed in 30 years? Apart from the obvious aging, of course.
posted by drstrangelove at 5:27 AM on March 15, 2021 [5 favorites]


Is it the number of guns on the bookshelf behind the face?
posted by pompomtom at 5:29 AM on March 15, 2021 [6 favorites]


Is it the number of guns on the bookshelf behind the face?

Or simply the number of books?
posted by drstrangelove at 5:35 AM on March 15, 2021 [6 favorites]


They might all be bibles (and maybe the odd copy of the Protocols of the Elders of Zion...)?
posted by pompomtom at 5:51 AM on March 15, 2021 [2 favorites]


> Or simply the number of books?

Right wingers have learned the value of a background of books on their zoom interviews. Of course when you take a look the shelves are stocked with a combination of leather-bound books with unreadable bindings and a hefty helping of those books ghostwritten on behalf of Fox News celebrities.
posted by at by at 6:57 AM on March 15, 2021 [5 favorites]


I'm about to dig in, but wanted to stop to say after spending a hellish few months with family, wouldn't an ultra-conservative explain a lot of this?

I can ID politics based on hair alone in the areas where my family lives.

Also I ran across some study a while back talking about how income level is fairly visible in the face, and that too can be a determining factor. The more money someone has, the higher the odds that they're conservative.
posted by liminal_shadows at 8:32 AM on March 15, 2021 [5 favorites]


That second link is just blowing my mind.

First they measure people's physiological response (increased skin conductance and harder blinking) to threatening images ("a very large spider on the face of a frightened person, a dazed individual with a bloody face, and an open wound with maggots in it"). So far, so good.

Then they try to establish that people with more pronounced physiological responses to fear stimuli are also more likely to hold political opinions that "protect the interests of their own social group." This is where it goes completely off the rails. Compare and contrast their definition of the participants' social group (underlined and bolded in the quote below) with that list of political positions:
We identified particular positions on 18 of these policy issues as those most likely to be held by individuals particularly concerned with protecting the interests of the participants’ group, defined as the United States in mid-2007, from threats. These positions are support for

- military spending
-warrantless searches
-the death penalty
-the Patriot Act
-obedience,
-patriotism
-the Iraq War
-school prayer,
- Biblical truth;

and opposition to
-pacifism,
-immigration,
-gun control,
-foreign aid,
-compromise,
-premarital sex,
-gay marriage,
-abortion rights,
-pornography.
Support for these policies absolutely does not indicate a desire to protect the social group defined as "United States in mid-2007". It indicates a desire to protect the social group of white nationalist, christian fundamentalist, male supremacist, warmongering, authoritarian United States people in mid-2007. I... I can't even deal with how stupid this list is, but here's the real punchline that comes immediately after it:
We do not label these collections of policy positions as either “liberal” or “conservative” because we measure only one aspect of ideologies and exclude other aspects such as positions on economic issues. We ... only assert that those most concerned about social protection will tend to be attracted to the particular policy positions listed.
What stunning verbal & logical contortions to avoid saying "we found racists".

If you stick around for the mid-credit funny bit, you may also catch this gem:
race and ethnicity were not controlled because all but one participant was self-identified as white and non-Hispanic
posted by MiraK at 8:42 AM on March 15, 2021 [19 favorites]


It's time for some game theory psuedo-genetic bullshit!
posted by medusa at 8:54 AM on March 15, 2021 [3 favorites]


Neophrenology is a thing now? Huh.

See this small ridge about an inch above the right ear just below where the temporal joins the parietal bone? That's a classic telltale sign of resistance to market reform.
posted by star gentle uterus at 9:01 AM on March 15, 2021 [8 favorites]


The more money someone has, the higher the odds that they're conservative.

Sure, but that's a prime example of correlation does not equal causation. Maybe money makes people more conservative. Maybe being conservative means they are more likely to enter lucrative careers. Maybe having more money and being conservative are both related to a third factor, like white privilege.

That's not even touching the highly biased recruitment and methods. As my mentor used to say, "garbage in, garbage out."
posted by basalganglia at 9:05 AM on March 15, 2021 [3 favorites]


Only racists hate spiders?
posted by Nanukthedog at 9:06 AM on March 15, 2021 [3 favorites]


"Genetic heritability" or my politics are likely to be influenced by the politics of my parents?
posted by 3.2.3 at 9:09 AM on March 15, 2021 [2 favorites]


If this was true and genetics or family history influenced one's political views I'd probably be a not so vaguely racist and anti-LGBTQ+ Mormon that voted conservative and had a career as an officer in the Air Force.

It is physically impossible for me to be any farther from any of those things without being from an entirely different planet or species.
posted by loquacious at 9:13 AM on March 15, 2021 [3 favorites]


Only racists hate spiders?

APPARENTLY

(srsly how was this not the headline racism-denial ruins everything)
posted by MiraK at 9:19 AM on March 15, 2021 [1 favorite]


If you want to find genetic heritability then you'd damn well better be testing genes not phenes. And definitely not people's feelings about phenes in a picture out of a yearbook.
posted by Horkus at 9:37 AM on March 15, 2021


The last link is to a study comparing monozygotic and dizygotic twins, Horkus. It isn’t looking at faces though, plus also it finds stronger genetic association with ?temperament? than with party alignment.
posted by clew at 9:50 AM on March 15, 2021


Agreed, but it's an unrelated study from 2005. I guess I took it as a sort of "previously" to the thesis of the FPP.
posted by Horkus at 10:05 AM on March 15, 2021


"Do Liberals And Conservatives Really Have Different Moral Foundations? Differences May Be Less Clear Cut Than Often Claimed"

"The idea that political conservatives and liberals differ in fundamental ways — in their biology and neurology, personality and moral foundations — has received a good deal of attention. However, cracks have begun to appear in this idea. In 2019, we covered new work finding that conservatives are not in fact more readily disgusted than liberals (disgust has a moral dimension, of course). And the year before, Jesse Singal, a regular Digest guest blogger, covered evidence suggesting that claims about liberal-conservative personality differences have been overblown.

Now a major new review and meta-analysis of research into political orientation and moral foundations — essentially, how people view morality — calls into question some influential earlier conclusions. Writing in the Psychological Bulletin, J. Matias Kivikangas at Aalto University and colleagues report finding support for the idea of some basic moral differences between conservatives and liberals. However, they also conclude that the differences are less clear cut than had been thought, and the results are also less generalisable across regions, countries and political cultures than has been claimed."
posted by Mr.Know-it-some at 10:08 AM on March 15, 2021 [3 favorites]


other signifiers of membership, from make-up to iconography. While some of this is explicit (such as slogan T-shirts or hair dye)

I can ID politics based on hair alone in the areas where my family lives.


This hair business is intriguing; I'd love some more details. Men and women? I mean, distinguishing the freaks from the hard-hats was easy in the late 60s and 1970s, but times change.
posted by Rash at 10:11 AM on March 15, 2021 [1 favorite]


AFAICT from a quick read the first paper is not making claims that the the facial differences are physiological, let alone genetic.

Gratishades, in the second comment here, was pretty quick off the mark in pointing to the fact that the photographs are self-selected (in one study by candidates for political office) and it's quite plausible that the cues are "just" in expression and other changeable traits.
posted by mark k at 10:12 AM on March 15, 2021


Christ, the skull fondlers are back at it again.
posted by MartinWisse at 10:13 AM on March 15, 2021 [8 favorites]


Cameras patrol the streets taking snapshots of passersby, sort sort sort. There. Fixed it for you. No more expensive campaigns and voting booths to decide what's already decided. Power to the People! We have your best interests at heart. You can trust us. We are Them.

(Yeah, well, I don't know what you call them, but I know what they look like.) What could go wrong?
posted by mule98J at 10:36 AM on March 15, 2021


The study is of pictures of people that the people chose to represent themselves with.

I bet the venn diagram of Republicans and "When you're about to get a hot take from the local news comment section" is a circle.
posted by Your Childhood Pet Rock at 10:53 AM on March 15, 2021 [2 favorites]


I didn't see any talk about controlling for age, is this just "looks old = probably republican"?
posted by ymgve at 11:41 AM on March 15, 2021 [2 favorites]


Christ, the skull fondlers are back at it again.

I know, right? This is seriously right up there with phrenology.
posted by Nanukthedog at 12:15 PM on March 15, 2021 [2 favorites]


"Undergraduates at Tufts can identify politicians from the most recent US senate election from CNN photos at levels slightly above chance."


Just turn the photo 30 degrees.
posted by Nanukthedog at 12:21 PM on March 15, 2021


I don't know if y'all have been following the wrong people on Twitter but the Andrew Sullivans and Steven Pinkers of the world have been jumping the I-don't-want-to-hide-my-racism-anymore ship for the just-asking-the-hard-questions pseudoscience liferaft for some time now.

They've christened it "Quillette." I hope it sinks.

I'm certain some of this trash will show up there in the form of gross mischaracterizations, but I'll never find out one way or t'other, I guess.
posted by klanawa at 12:53 PM on March 15, 2021 [1 favorite]


Huh, that's not even the same "let's tell people's politics from their face" paper I was expecting. There was a more recent one, from the same person who did the "gay / straight face" paper before that.

https://pluralistic.net/2021/01/15/hoover-calling/#phrenology

Similar to Gratishades, I'm immediately skeptical of the claim that it's wholly about your face, not hundreds of tiny things about how you present yourself in a photo.

Like, for an extreme example, imagine Butch Cassidy taking a duck-lips selfie. There's a lot of cultural expectations around how people pose for photographs.


Is maybe the idea of political inclinations having a biological source a tempting explanation if you feel people are completely entrenched? Is it one of those situations where it's nice to have a simple explanation even (or especially) if it's one you can't do anything about?
posted by RobotHero at 5:32 PM on March 15, 2021


And the year before, Jesse Singal, a regular Digest guest blogger, covered evidence suggesting that claims about liberal-conservative personality differences have been overblown.

Jesse Singal is one of a handful of writers whose claims I always treat with skepticism.
posted by Merus at 7:30 PM on March 15, 2021 [2 favorites]


Would it work better on feces?
posted by RobotVoodooPower at 9:23 PM on March 15, 2021


Having just now checked whether ratemypoo.com is still a thing, it would appear that the answer is Yes.
posted by flabdablet at 9:31 PM on March 15, 2021


Political scientist here. I hate this vein of research, it speaks of faddism, and I bet if you poke and prod it the results don't hold. Thanks.
posted by MisantropicPainforest at 10:14 AM on March 16, 2021 [1 favorite]


I would rather test the grandmotherly hypothesis "If you keep that expression your face will stick that way", or the more literary version "By forty every man has the face he deserves".
posted by clew at 10:18 AM on March 17, 2021 [2 favorites]


maybe after 40 generations of assortative voting/mating we might see some speciation :P
A Close-Up Picture of Partisan Segregation, Among 180 Million Voters - "Democrats and Republicans increasingly live apart, all the way down to the neighborhood level."[1,2]
posted by kliuless at 12:03 PM on March 22, 2021


« Older healing trauma with art   |   Only Some People Get Full Dimensionality and... Newer »


This thread has been archived and is closed to new comments