Physiognometry


The idea that you can tell things about a person from their appearance underlies the entire point of dressing up, wearing make-up, and being concerned with our appearance. There are probably some things you might be able to assume from a person’s looks, but it’s still going to be pretty unreliable: you can’t tell someone’s a terrorist because they “look muslim” because “looking muslim” is a vague concept to begin with.

What about people who look like criminals?

She has an honest face, doesn’t she? Young and pretty, maybe she’s a prostitute? Nope:

Crime: malicious injury to property and wounding with intent to do grievous bodily harm. When a police officer arrived to arrest Esther Eggers for malicious damage she attacked him, causing serious injury. Eggers was sentenced to 12 months prison. Aged 22.

What can we tell about her other than that her clothes are more expensive?

Convicted of conspiracy to procure an abortion. Lillian Boland worked as a secretary for an illegal abortionist who operated out of a dentist’s surgery on Oxford Street, Paddington. Boland protested her innocence and ignorance of the ‘doctor’s’ work; however, the court decided she must have had detailed knowledge of the business and handed her a suspended sentence of 12 months hard labour.

It’s funny to me because I see these clothes as “vintage” and know that “vintage” is expensive. I am sure that Ms Boland, were she alive today, would be shocked at what that outfit would fetch on Ebay.

Dalton has an impressive scar on his face. Must be a violent man. Or maybe someone who worked near dangerous machinery – which is to say pretty much anyone who wasn’t an office-worker in 1910.

Special Photograph no. 129. A cropped print of this photograph appears in a police photo book from the 1920s, annotated in pencil “magsmen”, with no further information offered.

A “magsman” is a con artist. Perhaps this pair were politicians.

Ellis looks like a bad hombre.

Special Photograph no. 86. The precise circumstances surrounding this picture are unknown, but Ellis is found in numerous police records of the 1910s, 20s and 30s. He is variously listed as a housebreaker, a shop breaker, a safe breaker, a receiver and a suspected person. A considerably less self-assured Ellis appears in the NSW Criminal Register of 29 August 1934 (no. 206). His convictions by then include ‘goods in custody, indecent langauge, stealing, eceiving and throwing a missile.’ His MO includes the entry ‘seldom engages in crime in company, but possessing a most villainous character, he influences associates to commit robberies, and he arranges for the disposal of the proceeds.’ It adds that he has the nicknames ‘Curley’ – his hair is thinning – and ‘Deafy’, as he is by then quite deaf.

Indecent language! Fucking hell, you can get convicted for that!?!

Valerie Lowe, 15

Valerie Lowe and Joseph Messenger were arrested in 1921 for breaking into an army warehouse and stealing boots and overcoats to the value of 29 pounds 3 shillings. The following year, when this photograph were taken, they were charged with breaking and entering a dwelling. Those charges were eventually dropped but they were arrested again later that year for stealing a saddle and bridle from Rosebery Racecourse. In 1923 Lowe was convicted of breaking into a house at Enfield and stealing money and jewellery to the value of 40 pounds

Special Photograph no. 1313. Munro is listed in the NSW Police Gazette, 1924 as charged, along with Harris Hunter, with receiving stolen goods to the value of 536 pounds 4 shillings and 1 penny, the property of Snow’s department store.

I’ll stop here – there are about a hundred of them at the Syndey Police Museum [sydney museum]

Though they were taken in a situation of dubious consent, in somewhat difficult conditions, they amount to a brilliant series of portraits.

One thing we learn from these portraits is that it’s really hard to tell anything about a person from the clothes and face. Galton used mug shots like these to layer into composite images, looking for common features – and robo-profilers are still pursuing Galton’s mistake [stderr]

This fellow may look like a creepy pedophile but he’s Cesare Lombroso, a criminologist and physiognomist who tried to demonstrate that criminality was an essential part of some people’s nature [wikipedia]  He was a professor of psychology and eventually ran an asylum: just the kind of person who you’d want for the job – someone willing to convict you based on your appearance. Apparently they didn’t have mirrors or irony meters back in Lombroso’s time.

There were earlier physiognomists, including Kaspar Lavater [wikipedia] who was working with sketches, before the invention of photography. There was also a Paris prefect of police, whose name I forget (and I am not going to call dad about it!) who built a card catalog that prefigured J. Edgar Hoover’s FBI catalog of culprits – the catalog was indexed with multiple indices: if you were looking for a footpad with a scar on their face you would cross-index the two and it would produce a list of suspects.

One thing that’s unavoidable, from the collection of mug shots: there is a distinct class bias. There are no rich people at all and the criminals are mostly guilty of subsistence crimes – trying to survive in the world – a world where the apparatus of policing exists to track you and make sure you stay in your place. The punishments seem to be as arbitrary then as they do now – you can steal millions in a stock scam and be in a low security prison for 6 months, or you can get a year’s labor for being an abortionist’s secretary.

It’s all wasted effort, but some of the artifacts that remain are quite beautiful.

------ divider ------

I’ve been a fan of photography my whole life, and I spent about a decade doing my own wet plate ambrotypes: mixing the silver nitrate, nitric acid, and water – coating plates with nitrocellulose dissolved in ether and spiked with cadmium bromide, etc. The entire process fascinates me, and – at this point – I can look at most photos and tell you a lot about the processes that were used. These mugshots look like early celluloid film (gelatine chlorobromide emulsion) shot with a 5×7″ press camera using natural light, probably developed in an old developer like Agfa Rodinal (which is an aminophenol that partially stains the halides as it reduces them)   This is just a delicious process and looking at these pictures makes my hands twitch to get them wet in the darkroom again.

Comments

  1. says

    The thing is, it’s not a completely crazy idea. There are disorders that manifest both as a set of visible features and mental ones.

    It turns out to be wrong, as far as anyone can tell, but it’s not crazy.

    That big plastic brain of ours apparently makes a surprising amount of stuff arise out of nurture.

  2. says

    Alphonse Bertillon? Cesare Lombroso? Probably Bertillon, with his system of biometrics. Have you spent time at Crimino Corpus?

    The book The Killer of Little Shepherds, by Douglas Starr, features Bertillon quite heavily, about the early serial killer, Joseph Vacher.

  3. says

    Andrew Molitor:

    The thing is, it’s not a completely crazy idea.

    Yes, it is. A very bad idea, based on biases. It’s nothing more than placing current bias and bigotry on a pedestal, and proclaiming “science!” The early efforts in this regard can be forgiven, there was a lot of ignorance in play. There’s no excuse anymore.

  4. starskeptic says

    Seems we could have saved Mr. Lombroso a great deal of work; lock up all the poor people…or some sort of fence or something…

  5. chigau (違う) says

    Andrew Molitor #1
    The are disorders that manifest both as a set of visible features and mental ones.
    For example?

  6. says

    Andrew Molitor:

    @caine, you’re not understanding my comment,

    I understood you just fine. Wrong, but not crazy, yes, yes. You are highly prone to saying stupid shit on the ‘net. Perhaps you should think about what that says about you.

  7. says

    Amolitor@#1:
    There are disorders that manifest both as a set of visible features and mental ones.

    Since those tend to be uncertain (to be accurate one would have to ask) it seems to me to be easier to assume that it’s an unknown, rather than trying to correct for it.

  8. chigau (違う) says

    Andrew Molitor
    Yes, it is sometimes possible to determine that a person has Down Syndrome just by looking at them.
    How about criminality?

  9. says

    Well yeah, the idea is wrong. As far as anyone can tell the isn’t anything that is visible which correlates with criminality.

    If downs correlated with homicidal rages rather than a remarkably pleasant disposition, they’d be on to something, albeit a small something.

  10. jrkrideau says

    Sounds like Bertillon but was not his system an identification system not one that inferred behaviour? Enough physical measurements would give you a “unique” identification? Basically the same prinicple as in fingerprinting?

    Either or both “may” work. At the moment I don’t know of any convincing studies of either.

    Forensic “sciences” have a long tradition of having dubious or no validity.

    The polygraph, for example, seems closer to some kinds of applied shamanism or witchcraft than science.

  11. Brian English says

    I saw the pics, and thought you were showing some generic USian photos, then my underpowered processsor of a brain must have an index somewhere (I have a good memory and most processing tasks are just a form of caching ;) and it reminded me of the NSW photos of criminals, and so it was those photos….

    There are disorders that manifest both as a set of visible features and mental ones.

    Fetal alcohol syndrome…

    Galton used mug shots like these to layer into composite images

    I knew of Galton because of regression, and have heard that he was a bit of Victorian adventurer and wrote that periods Lonely Planet(s). But I didn’t know that he was into psysiognometry. No surprise I guess, I thought he was a pretty cool dude, but it turns out that most cool dudes are into racism and sexual harrasment in my experience….

  12. jrkrideau says

    @ 11 chigau
    How about criminality?
    No, because of the question of what is “criminal”. Shooting and killing anther person is homicide. Depending on the circumstances you may get a medal or a life sentence.

    You may be able, as John M suggests, to predict a tendency to emit a certain behaviour but whether or not it is criminal depends on the existing criminal code, societal norms and probably other things that I cannot think of at the moment.

  13. Pierce R. Butler says

    Suppose we could make useful conclusions from some facial features – say, a prominent jawbone often correlates to a dominant/independent personality.

    We would always find exceptions and come up with false inferences (so, a visible overbite -> subordinate personality – but then how do you explain Mitch McConnell? And why doesn’t Jay Leno rule the world?). Also, any such study, once published, would get warped by media and public prejudices, with various individuals catching hell from casual passersby, HR departments, and of course cops (“study of mug shots shows people with eyes closer together than average arrested more often”).

    Even if researchers could find something there, ethics panels would have abundant reason to abort such projects at a very early stage.

  14. says

    I can tell at a glance from 100′ the gender of many cats, and thereby make some good guesses at personality traits. I can identify the breed of many dogs and some horses, and thereby make some similarly good guesses at personality. None of this is 100% accurate, of course, but it’s going to be better than random.

    When phrenology was happening, it was not at all obvious that we couldn’t do some of the same things for people. Yes, it’s true, the research was lousy, and often turned into simply coding prejudices. So it is with humans when they’re doing research. It’s an ugly business, and this was uglier than some. But still, the hypothesis was not idiotic. It was in fact quite reasonable — humans, it turns out, are kind of an outlier species. The research was done, falsified, mis-represented, re-studied, used for evil, and finally discredited. As it so often is with science.

    Related, the idea that personality and behavior traits are gender linked. Trotting this idea out for people will get you savaged in many many places, I assume this being one of them. But for virtually every animal other than people, it’s not merely reasonable, it’s also obviously true. Humans again seem to be some weird outlier species here in the degree to which chromosomal makeup does not correlate with personality/behavior (and, of course, the very topic is unstudiable because reasons, so it will be 100 years before anyone actually finds out what links there are, if any).

    It was *wrong*, but not *crazy*.

    Doing it again, now, is arguably crazy. Doing the same thing over and over expecting, etc.

  15. says

    The photos, by the way, are indeed lovely. The out of focus backgrounds especially appeal to me, for some reason, and the ones where the subject is mysteriously crammed into the corner.

  16. Brian English says

    I can tell at a glance from 100′ the gender of many cats, and thereby make some good guesses at personality traits.

    You may be able to tell the sex of cats from 30m, and make some good guesses that they’re cats. Not sure about personality traits. I happen to be slave to 8 cats, and know their sexes, but only 1 is a conservative male, as in, he’s a male (neutered) cat and acts like a male and I can’t guess that from the way he ponces about. The rest are just sick killers who are quite happy to be ‘pussies’ and be quite asexual.

    When phrenology was happening, it was not at all obvious that we couldn’t do some of the same things for people.

    sort of like arguing about unicorns. There wasn’t an evidential basis to phrenology, there was a social, descriminatory basis, it’s not like Europeans seem to be pretty smart, other not so much.

  17. says

    Andrew Molitor@#18:
    The out of focus backgrounds especially appeal to me, for some reason,

    Probably a 3rd generation lens design, wide open. Nowadays we’d say something ecstatic about the “bokeh” or something. I agree!

    and the ones where the subject is mysteriously crammed into the corner.

    The gallery’s description says something about that the photographer allowed the subjects a bit of leeway in how they posed. So some of the poses are genuine self-expression while others maybe represent the criminals’ view of how they think society wants them to represent themselves to it.

  18. says

    @Brian English on the one hand, sure, it’s like arguing about unicorns.

    Still, there are many topics from history where the standard response is, basically, some hysterical shrieking about racism and privilege. This is ahistorical, it is the opposite of critical analysis. I oppose this, vigorously, and whenever I get a chance. I happen to think that a critical understanding of history can co-exist with an acknowledgement of the flaws of the people who shaped that history. I happen to think that a critical understanding of history is an inherently good thing to have, which is basically why I oppose the hysterical shrieking.

    Note that the Standard Response basically boils down to:

    EVERYTHING WAS SHIT AND THEY WERE ALL ASSHOLES BUT I AM RIGHT AND YOU SHOULD AGREE WITH ME AND NOT THINK TOO MUCH ABOUT ANYTHING JUST AGREE AND IF YOU AGREE HARD ENOUGH EVERYTHING WILL BE ROSES BUT IF YOU DON’T AGREE HARD ENOUGH YOU WILL DIE IN A FIRE

    which is also the Standard Song of tyrants. It’s just a coincidence, but there you have it.

  19. says

    Brian English@#19:
    There wasn’t an evidential basis to phrenology, there was a social, descriminatory basis, it’s not like Europeans seem to be pretty smart, other not so much.

    The tough thing about these pseudosciences is that they do have some kind of underlying mechanism – like the bits I posted from Poppenoe’s book on Applied Eugenics there was an underlying truth – it was just the wrong truth.

    There was an evidentiary basis, it was just wrong. Makes me think about Evolutionary Psychology – this is the core issue of philosophy and science, really: epistemology.

  20. says

    jrkrideau@#15:
    You may be able, as John M suggests, to predict a tendency to emit a certain behaviour but whether or not it is criminal depends on the existing criminal code, societal norms and probably other things that I cannot think of at the moment.

    I’d say it has more to do with how good a lawyer you can afford. Ask folks like Jeffrey Epstein and Martha Stewart.

    Of course, if criminality is dependent on your war-chest for legal defense, then there’s really no such thing as “criminality” at all. As I mentioned in the OP, there are no wealthy people in the pictures. One can argue that the wealthy are going to be disproportionately underrepresented because of demographics, but – it’s complicated.

  21. Siobhan says

    Why would the police need an “excuse” for profiling? They already do it with no consequences. Stop-and-frisk? Walking while trans? Police departments keep getting sued over this shit and yet they still do it. Hell, every time I’ve gone flying I’ve been “randomly” selected for additional screening thanks to those absurd body scanners that are quite convinced my cock is an explosive device in waiting.

    @Pierce:

    Even if researchers could find something there, ethics panels would have abundant reason to abort such projects at a very early stage.

    And rightly so.

  22. says

    I have not been asked as such, but it has been gently suggested that the world would be improved in ways that matter to me if I were to take my opinions elsewhere, and so I shall.

    Best to all.

  23. says

    Andrew Molitor@#17:
    I can tell at a glance from 100′ the gender of many cats

    That’s cheating, since cat fur genetics are sex-linked!

    I get your point, that there may be some outward signs of something but you’re walking down the rathole (apologies to Caine’s rat team!) that eugenicists did: there are a great deal of things that are environmental/social that affect outcomes far far more than the tiny number of genetic factors. And, not to attempt Argumentum Ad Lauper “Money changes everything.”

  24. says

    But – not to distract everyone with the flashy thing….

    I kind of expected that we’d all observe the relationship between clothing and criminality/social expectations. Clothing as a signal for wealth (and all that goes with it, signalling trustworthiness, etc) is deeply embedded in all the mug shots. If you look at Ellis – who is manspreading conveniently – he’s trying to look snappy and dapper – what is he signalling? I’d love to know, for example, if the mug shots are of the suspects as they looked when they were picked up, or if they got a chance to neaten up a bit first?

    Clothing as a signal for wealth, and wealth as a proxy for trustworthiness/stability, is particularly relevant to the current political environment in the US, where a demoagogic conman is now president of the US based on his apparent wealth (which may be a false signal)

  25. says

    Marcus:

    I kind of expected that we’d all observe the relationship between clothing and criminality/social expectations.

    How we judge on appearance is such a mix of so many micros – cosmetics, hair, jewelry, scent, clothing, stance are all processed so rapidly, we think they really don’t matter to us, but of course, they do. Then again, we often use such things to deny we could possibly be racist: It’s nothing to do with skin colour, it’s all that face jewelry, or that hoodie, or whatever.

  26. says

    Anyway, we all tend to be judgmental asses, and are always coming up with new justifications for very old behaviour.

  27. says

    One thing we learn from these portraits is that it’s really hard to tell anything about a person from the clothes and face

    “Faces of meth” excepted.

    Lombroso’s claptrap about appearances and crime is akin to the rightwingnut attitude towards obscenity: “I know it when I see it.” They can’t define it, so they slap the label on anyone they don’t like – most often, that’s the poor, non-whites, and the powerless in society. Why target those who could fight back, even if they are the ones more dangerous to society (e.g. catholic priests)?

  28. says

    Shiv@#25:
    Why would the police need an “excuse” for profiling? They already do it with no consequences. Stop-and-frisk? Walking while trans? Police departments keep getting sued over this shit and yet they still do it. Hell, every time I’ve gone flying I’ve been “randomly” selected for additional screening thanks to those absurd body scanners that are quite convinced my cock is an explosive device in waiting.

    I agree. In the US it’s blatantly unconstitutional, but they get away with it by targeting only the less wealthy – the wealthy have private planes, frequent flier clubs, TSA precheck, etc. Fuck everyone else.

  29. says

    Intransitive@#31:
    They can’t define it, so they slap the label on anyone they don’t like – most often, that’s the poor, non-whites, and the powerless in society.

    I think (per my reply to Shiv@#25) that’s just a fig-leaf. It’s power that’s the issue; the label goes on anyone who’s not obviously rich and privileged.

  30. jrkrideau says

    @ Marcus Ranum

    I’d say it has more to do with how good a lawyer you can afford.

    No, the enforcement of sanctions for a “criminal act” is a different matter from my intended meaning. Then is when you need a good lawyer or the equivalent in the local social and legal system. The ability to bribe local police and/or judiciary may help in some countries?

    Even if you manage to avoid legal sanctions because of a good lawyer, people still will classify you as a crook or moral degenerate …. Just one right or well-connected enough to game the system.

    To go back to my original homicide idea. In the recent shootings in Paris, the person who shot and killed the police officer committed a criminal act. The police officer who killed the attacker was performing their duty.

    Slavery was generally legal and socially acceptable in the Roman Empire; it was illegal and socially reprehensible in 20th Century Canada. And so on.