Supercentenarian records show patterns indicative of errors and pension fraud (2020)

https://www.biorxiv.org/content/10.1101/704080v2

by bookofjoe • 3 years ago

456 275 3 years ago

jedberg3• 3 years ago

This is actually really fascinating. If only humans had rings we could measure!

To me the biggest finding is the lack of people 90-99 in the same areas. Where are the supercentarians coming from with a lack of 90+ pipeline?

nearbuy2• 3 years ago

Supercentenarians are people over 110 (centenarians are over 100). Only about 1 in 100,000 people live this long. Around 1 in 5 people live to 90. An order of magnitude more supercentenarians than average (through fraud or error) wouldn't make a noticeable difference in the number of people living 90-99, even assuming all supercentenarians were once counted as being 90-99.

jedberg3• 3 years ago

Yes I understand that, but if you are saying some area is a "blue zone" where people tend to live longer, you'd expect an outsized number of 90-99 (and 100-109) for the same reasons that there are 110+, if those reasons are related to the geography (which is the claim with blue zones).

londons_explore• 3 years ago

After a person has died, I suspect science could measure their age pretty accurately with enough effort.

For example, there are certain cells that don't multiply after birth (eg. some nerve cells). One could presumably date carbon atoms in their DNA...

Or parts of the body that don't regenerate - like tooth enamel.

I suspect with the right type of imaging, you'd probably find 'tree rings' in things like fatty deposits in arteries too.

dontlaugh• 3 years ago

Even cells that don’t multiply still accept nutrients. Pretty much no atoms in our bodies stick around for too long. It’s why carbon dating largely measures the time an organism died.

pcthrowaway• 3 years ago

I agree we could find markers of age, but I think they'd be nothing like tree rings, which are created by the freeze-thaw cycles and the fact that they are fixed in one place and exposed to the elements in ways humans are not.

Although, that begs the question, if a potted tree were to be placed on a cruise ship, which always sails to warm weather, would it fail to develop rings?

btilly4• 3 years ago

People have looked for such things. https://en.wikipedia.org/wiki/Epigenetic_clock describes the best one found.

But, unfortunately, all the ones we know of can be skewed by environmental factors. And it is likely that the exact same environmental factors which make someone live longer will also make them appear younger than they are. Which makes them particularly unreliable exactly for the oldest people on record.

user070223• 3 years ago

apperantly Dodo birds has marking on the bones[0] which researchers interpert it as time they struggled to find resources (simillar to tree where you move between wet and dry season). I guess one could cross reference times of femine to a skeleton with known lifetime to see if it shows up in humans as well.

[0] https://youtu.be/Juci-kAqjes?t=219

ip262• 3 years ago

Maybe supercentarians are more likely when, for whatever reason, there aren’t any cheeky young upstarts breathing down their necks :)

mannyv2• 3 years ago

This article is amusing because some countries don't have a good way of notifying institutions about deaths (like the SS death master file)...and there are lots of issues associated with dying (taxes, inheritance, loss of benefits, etc).

In those countries they just don't report the death, sometimes for decades. I used to joke that the government should have a celebration of centenarians and see how many of them actually show up.

PaulHoule• 3 years ago

I used to kid that I couldn't get a security clearance because I had relatives in Eastern Europe collecting social security for dead people.

civilized• 3 years ago

The SS death master file has been broken for a decade now. Institutions that need death information have to work with private vendors that aggregate data from a variety of sources.

mannyv2• 3 years ago

Yeah, once states were able to opt-out (due to medical/privacy issues) it basically stopped being a canonical reference. But it's still pretty good for the most part.

meetingthrower• 3 years ago

Ah yes, anybody remember the lovely insurance companies who had PERFECT data for stopping annuity payments, but somehow couldn't find the records to pay life insurance claims?

cperciva• 3 years ago

I used to joke that the government should have a celebration of centenarians and see how many of them actually show up.

IIRC a version of this happened a while back in Japan; the mayor of a town decided to visit some of the oldest residents, only to find that none of them were still alive.

Animats• 3 years ago

That turned into a sizable pension fraud scandal in Japan.[1][2]

[1] https://www.nytimes.com/2010/08/15/world/asia/15japan.html

[2] https://www.economist.com/banyan/2010/08/06/the-case-of-the-...

jandrese• 3 years ago

It has long been noted that the oldest people in the world are clustered in countries that didn't keep paper birth records.

cwmma• 3 years ago

or might have had something happen to the records (cough japan cough)

Tade02• 3 years ago

My maternal grandparents are in their mid 90s, my paternal grandfather passed away a week before her 97th birthday.

A lot can happen within a century and there's usually hardly anyone alive to confirm some facts. My paternal grandmother's birth certificate was gone before the end of WW2 and when asked she would give an age six years younger - initially it was to prevent a scandal because not only was my grandfather of lower status (Pah! "Just" a doctor! Scandalous, I daresay!), he was younger.

As for my maternal grandparents their age is confirmed by their marriage certificate(would be harder to obtain one with this date as a younger person), children in their late 60s and living siblings of my grandpa, of whom there are seven.

sseagull• 3 years ago

Family history is funny like that. I have a few (usually female) ancestors who, according to censuses, just don’t age at quite the same rate as other people :)

I find it kinda charming. Gives some insight into who they are, and that fear of aging is a pretty universal feeling.

tasty_freeze• 3 years ago

My grandmother was born in 1899 in Ireland, and moved to the USA in 1920. In the early 1980s, she came clean: "I don't want to spend eternity with a lie on my gravestone. I was born in 1894."

Tade02• 3 years ago

I only learned the truth right after she passed away and my father was free to discuss this - the year was 1911, not 1917 - clouded by two wars' worth of lost records.

Y_Y2• 3 years ago

It's not so bad, a headstone engraving will usually be gone after a couple of hundred years.

hn_throwaway_997• 3 years ago

I know some experts have poo-poo'ed the theory that Jeanne Calment's daughter impersonated her ( https://en.wikipedia.org/wiki/Jeanne_Calment#Scepticism_rega... ), but I don't find their arguments convincing. I'm not saying it's settled, but I think the evidence strongly points to her not being as old as she says, and I think the "experts" downplay the evidence for the switch theory too strongly, even if it is largely circumstantial:

1. The age gap between Calment and the next oldest person is the strongest evidence. It's very, very hard to ignore this.

2. As outlined in this article, there is a strong economic motive for the swap.

3. I think there is other evidence (e.g. some of the photos) that wouldn't be that strong on their own, but just add to the weight of the other, stronger pieces of evidences that there was a swap.

GalenErso• 3 years ago

1. The age gap between Calment and the next oldest person is the strongest evidence. It's very, very hard to ignore this.

That doesn't mean her record isn't legit. Michael Phelps won 23 Olympic gold medals. The most decorated Olympian after him won 9.

There can be a big gap between the top performer and the second best performer. I'm sure a professional statistician would like to chime in.

ianferrel3• 3 years ago

"Number of Olympic medals" is a very poor analogous concept here, because it's not a directly measurable and variable trait, it's a complex tail-end derivative of several factors.

If you looked at "swimming speed", for example, which is the simple directly measurable thing, you would find that while Phelps is faster than other swimmers, , he is only a tiny bit faster than the next fastest swimmer, not 2.5 times faster than the next fastest swimmer.

ihaveajob• 3 years ago

The size of the data sample is massively different. Only a few thousand people have earned any Olympic medals, so variance is expected to be higher than in the "years lived" metric, where we have literally billions of data points.

hn_throwaway_997• 3 years ago

One thing I would highlight, because this is what I've seen the French experts in this situation do, is look at each piece of evidence in isolation and say "Ha, this alone doesn't prove it."

And, to that point, I agree, it doesn't prove it (and, to clarify, I don't think the controversy is proven one way or the other). But I think it's wrong to look at each piece of evidence by itself - it's the totality of all the evidence that makes me extremely skeptical of the overall claim. E.g. if the age gap was the only piece of evidence, it wouldn't alone lead me to believe there was a swap, but all the evidence that Novoselov and Zak present together have convinced me of their theory.

roflyear• 3 years ago

There's lots of people alive not an incredible amount shooting for swimming world records.

anonymouskimmer10• 3 years ago

The current, known, oldest living dog is almost 1.5 years older than the second oldest. The current one is also still living, so could exceed this number. https://en.wikipedia.org/wiki/List_of_longest_living_dogs

Proportionately this compares quite favorably to Calment, being about double her spread from the next oldest human.

Murfalo• 3 years ago

Conspiracy-theorist goggles on... this is also fraud?! (I have no evidence, just pointing out the possibility)

hn_throwaway_997• 3 years ago

If you think the data for humans is marginal, just imagine what it is for dogs. I absolutely don't think some "list of oldest dogs" is in any way exhaustive given that the vast, vast majority of dogs worldwide don't have accurate records.

naniwaduni2• 3 years ago

The gaps in the ages of longest-living dogs are quite large, however: the second-oldest dog is in turn almost two years older than the third, there's four years between #4 and #7 (with the exact gaps between #4-#7 unclear, but there should be at least two gaps of >1 year). This strongly suggests the gap can be attributed to sparsity of data.

The gap between Calment and #2 is over 2.5x the next largest gap in the records (1 year 80 days between #4 and #5).

wjholden• 3 years ago

Maybe this works like a reversed Poisson distribution?

In running races, it's not uncommon for the winner to be minutes ahead of the runner-up, who is only seconds ahead of the 3rd. After they cross the finish line, several more cross soon thereafter, and then the bulk of the runners come streaming in steadily. Eventually, the bulk of the runners pass the line and we start to see stragglers who were far behind the group, trickling in one by one slowly.

Workaccount2• 3 years ago

For me one of the strongest bit against this is that her daughter would have had to pull the wool over everyone in their towns eyes, or they would have had to have been in on it too.

jquery6• 3 years ago

Well, either something like that happened... or she won a lottery with a 10 billion to 1 chance. My money is on the former. I've read enough stories about people assuming false identities to understand there's a number of ways to pull something like that off.

kevinmchugh4• 3 years ago

The time between the first Calment death and Calment becoming notable is long enough - over 50 years - that it's plausible that people who were in on the scam/aware of the assumption of the mother's identity had already died. Especially since ww2 happened in the interim

Raffray is the biggest sticking point in this fwiw - he had time and incentive to debunk Calment's age, and had some preexisting relationship with her before they entered their contract.

mabbo2• 3 years ago

Would she?

After the mother's death, she keeps going about her life as usual, goes by her original name, does all the normal things she did before. On official documents she lists her mother's name but around friends or neighbours goes by whatever they want to call her. Maybe a lot of the neighbours know what's going on, but hey, we all want to avoid taxes, right? So it goes.

And gradually she starts going by the other name instead. People nod and understand if she corrects them. And eventually everyone around her that might prove otherwise dies off, moves away, or loses contact. Decades go by. Soon everyone remembers that older spritely woman who's always been around here.

Any photos or evidence that might prove her story false, well they go missing. The ones that confirm it stick around.

This is how all legends start, really.

naniwaduni2• 3 years ago

How easy is it to win a world particular world record by such a commanding margin past a cluster of runners-up? Easier than pulling the wool over a town's worth of people's eyes?

btilly4• 3 years ago

There is little evidence that she ever spent much time at her official address. Living in remote villas, as she mostly did, and simply going by "Madame Calment", as she apparently also mostly did, few would have paid attention to her given name or any discrepancy between that and official records.

When her extraordinary claims of old age became famous many decades later, well, have you ever seen family members arguing what really happened many decades ago? Doubts would have been hard to sustain.

https://www.liebertpub.com/doi/10.1089/rej.2018.2167 digs into this a bit.

whimsicalism• 3 years ago

I think you all are underestimating how much of a smoking-gun the magnitude of the gap is.

deanCommie2• 3 years ago

All the discussion makes it seem like the difference would be of 20 years or something.

But she died at 122. There are 63 people who lived at least 115. I would say if it's reliable that humans can live up to 115, it's not a huge difference to my mental model on longevity if they CAN'T (yet) live to 120.

hn_throwaway_997• 3 years ago

Look at the list of longest living women:

https://en.wikipedia.org/wiki/List_of_the_verified_oldest_pe...

If you look at, say, the top twenty, the gap between each woman tends to be on the order of weeks to a few months. The gap between Calment and the runner up is over 3 years. Statistically, this is a giant chasm for continuous data like this.

canjobear• 3 years ago

Also she had her personal documents and photos burned, a damning piece of evidence which is mentioned without comment here: https://en.wikipedia.org/wiki/Jeanne_Calment#Age_verificatio...

masklinn5• 3 years ago

There is nothing damning about it? Half of HN would immediately start throwing paper into the BBQ if the city requested access to personal documents.

masklinn5• 3 years ago

2. As outlined in this article, there is a strong economic motive for the swap.

Which one? We're not talking about Sogen Kato here, we're talking about Jeanne Calment.

Calment's daughter is supposed to have died in 1934, did the swap happen then? But there was no economic motive at all, the Calment family was part of the city's wealthy upper class and french social security was only introduced in 1945.

So what else, did the Calment's daughter fake her death somehow (despite her father, mother, husband, and son remaining public figures), then take over after the primary financial coup, the sale of the apartment to the notary in 1965, after what would have been the death of her husband and son both?

Not to mention the notary's strong incentive to uncover such a fraud: the flat was purchased in annuities, he ultimately paid twice the value of the flat (and waited 20 years before being able to use it, when Calment finally moved to a retirement home).

hn_throwaway_997• 3 years ago

No, the economic incentive was not pensions, it was avoidance of death duties. It is precisely because Calment's family was wealthier that this motive is stronger:

https://yurideigin.medium.com/jaccuse-why-122-year-longevity...

tyingq3• 3 years ago

Avoiding inheritance taxes, or distribution of a will that might not have favored the daughter?

jquery6• 3 years ago

Extraordinary claims require extraordinary evidence. The extraordinary evidence simply isn't there, there's too much fog in which some thing could have happened where the daughter assumed the identity of her mother, for whatever reason. It could've been taxes, could've been something temporary that turned into something permanent. Hell, maybe she was a bored housewife and thought that cosplaying as a grandma would be fun. I'm sure she had a grand old time as a "90 year old" that was able to act like a spritely 60 year old.

I simply don't believe she was the oldest, although I do acknowledge there is a very slim possibility of it being true.

This is a scenario where science fails us, I think. Social scientists have "proved" she was the oldest with "99% accuracy". But Bayesian statistics would pose the question, "if the odds of being the world's oldest person are 1 in 5 billion, and you took a test that was 99% accurate that confirmed you were the oldest, what are the odds of you being the oldest if the test is positive?" It turns out the odds of you being the oldest person are still astronomically unlikely... you have better odds of winning the PowerBall than of being the world's oldest person.

neaden12• 3 years ago

The extraordinary claim here is that a 34 year old woman would impersonate her mother for decades, including living with her father, and no one would ever notice or say anything. Your perspective is like arresting anyone who wins the lotto for fraud, because it's more likely that they would cheat then just happen to guess the right numbers.

pcrh2• 3 years ago

Your Bayesian estimations make no sense. When the alternative is that it is the daughter who survived to 98 (which is already old) rather than the mother surviving to 122, the question should be phrased as "if you are older than 98 what is the probability that your age is 122".

mlcrypto• 3 years ago

At this point I'm willing to believe any theory that the "experts" are scrambling to to hide

gifnamething4• 3 years ago

experts have poo-poo'ed the theory

Because the experts are in a joke, pseudo-scientific field in which the most prominent person is the man who verified her age with laughably poor methods and has a vested interest in not being proven wrong.

neaden12• 3 years ago

I was initially prone to believeing this, but the more I looked into it the more I realized how little sense it would make.

For your points: 1. This was the thing that initially made me skeptical, but it's just not enough to say. Sure living 3 years more then anyone else is pretty unlikely, but at the same time it's not proof. 2. The swap would have had to happen so early, when Yvonne was just 34 years old she would have had to impersonate her mother, pretend to be married to her father, and fool everyone in the small town she lived in. That is just not a credible thing. And the alleged payoff to avoid taxes doesn't seem big enough to be worth the immense effort that this swap would have had to go through. 3. Photos are fairly worthless for these sorts of things. Different angles, lighting, camera techniques all can make people look very different.

At the end of the day I find the alleged swap to be less believable then someone just living a long time

BobbyJo• 3 years ago

When you impersonate someone, you don't have to fool everyone, you only have to fool your mark. No one else is likely to know or care what you're doing.

goodcanadian2• 3 years ago

Also, even if you take the position that the age gap is so statistically unlikely as to be impossible[1], it tells you nothing about where the error is. It does not imply a swap. A swap is one possibility, but extraordinary claims require extraordinary evidence, and I don't see any credible evidence of a swap having occurred. Isn't it more likely that her age is simply wrong by a few years?

[1]Really, in statistics, extraordinarily unlikely is not the same thing as impossible.

jobigoud• 3 years ago

Sure living 3 years more then anyone else is pretty unlikely, but at the same time it's not proof.

The previous record at that point was 116 years, that's a 6 year gap. If you look at the distribution now it follows a fairly nice curve, except for that one far outlier.

fool everyone in the small town she lived in

If I recall correctly they were living in a remote location and didn't meet with a lot of people.

neaden12• 3 years ago

I'm replying to myself to lay out the timeline of Yvonne's death, since that is when people claim this switch happened. So first Yvonne's husband Joseph requested leave from the military and was granted it because his wife was sick, we have the record of that from the military. There is a picture of Yvonne at a sanitarium for TB patients around this time period, further establishing she had TB and was ill. Then the priest administered last rights and Yvonne died. There was a funeral mass that many people attended and a viewing of the body at the family home. This is in newspapers from the time. There are no accounts of Jeanne being ill in this time period.

So for this switch to have happened we would have had to have the daughter sick, then the mother secretly gets sick just in time for the daughter to recover. The mother dies and they come up with this plan to dodge taxes, even though presumably it would be risky and Joseph could lose his position as an officer in the army if they are caught. They get the town's doctor, priest, and newspapers to go along with the fraud. All of this to avoid a tax of about 6% of Jeanne's wealth, a considerable sum but do you really think it would be worth all the risk? At the end of the day, because of her status as the world's longest living person Jeeanne's life has been studied mroe then any other supercenturginarian, and no one has ever found proof that anything happened. We have census records from multiple decades, marriage and childbirth records with the government and church, and multiple mentions in newspapers.

hn_throwaway_997• 3 years ago

and fool everyone in the small town she lived in.

As others have pointed out, I think that is absolutely the wrong way to think about this. Yvonne Calment (or Jeanne) died in 1934. It's quite possible/likely that originally they were only trying to deceive the tax authorities. I also wouldn't discount how the chaos, upheaval and destruction of WWII (Arles was bombed heavily) would have made it easier for Yvonne to more "publicly" impersonate Jeanne in later years.

janmo2• 3 years ago

"but the more I looked into it the more I realized how little sense it would make."

True, but what is the probability for someone to become the world oldest person to ever have lived, and then to also outlive the second oldest by 3 years. It is 1 in 100 billions.

You can read this paper that did some math on this https://www.liebertpub.com/doi/10.1089/rej.2019.2227

It concluded that the probability of an identity switch to be 99.99%

littlestymaar2• 3 years ago

1. The age gap between Calment and the next oldest person is the strongest evidence. It's very, very hard to ignore this.

This is in fact a very weak evidence, that only appears to be a strong one because of a common epistemological mistake, let me explain:

The odds of such an age difference are extremely small according to a certain model of how people age, such a model has proven to be pretty accurate for the majority of the population, but that says nothing about the outliers. For biological entities like humans, one only need a single exceptional mutation or pathology to be a super-human of some sort, not fitting into the model at all. In fact, the tallest man in history is also a statistically impossible outlier, yet there's no doubt about his existence and actual height, he just happened to have a rare condition that caused him to grow up to a disproportionate height. (And if you grabbed a Guinness book of records, you'll find these kinds of things in almost every category related to human anatomy or physiology).

Confusing the models with reality is a very common mistake in the history of science, often committed by people having a math background instead of a physics one, but not only. A very famous example is how French explorer Dumont d'Urville was ridiculed when he described the rogue waves he witnessed in the Indian Ocean, because such a wave would be statically impossible. It turned out that the physical model of waves at the time was just too simplistic, as rogue waves do in fact exist (and AFAIK we're still looking for a proper model explaining the phenomenon entirely).

philegibbs4• 3 years ago

Aging does not work like height. There are several biological processes that contribute to aging so one gene mutation would not make someone live longer. We can make people taller by giving them growth hormone. If aging was as simple we would have easy ways to slow aging by now.

ars• 3 years ago

I know that lots of people would discount this, but it's interesting to note that Jewish tradition holds that the maximum lifespan for a person is 120 years.

And Calment is the only person that might have exceeded this. To me that's another bit of evidence toward the impersonation theory.

bentley• 3 years ago

This belief is commonly held by Christians as well. It comes from Genesis 6:3: “The LORD said, ‘My Spirit will not strive with man forever, because he also is flesh; so his days will be one hundred twenty years.’”

At that point, Genesis had described several individuals as having lifespans in the hundreds of years, most famously Methuselah at 969 years. This is sometime before the Flood. Note that interpreting this verse as a lifespan limit is not universal—an alternative reading (possibly a more popular one) is that 120 years in this context refers to how many years were left until the Flood.

After the Flood, Genesis continues to describe individuals with long lifespans, but they decline rapidly. Abraham, for example, died at 175, ten generations after Noah. Later, in the post‐Exodus era, Deuteronomy says that Moses died at age 120.

paulpauper• 3 years ago

1. The age gap between Calment and the next oldest person is the strongest evidence. It's very, very hard to ignore this.

The age gap is is just a few years. not inconceivable at all.

tim3332• 3 years ago

I had a read of Yuri Deigin's blog post on it, "J’Accuse…!", and found it fairly convincing. It has lots of photos of mum and daughter and you can amuse yourself figuring which is which https://yurideigin.medium.com/jaccuse-why-122-year-longevity...

Here's maybe the key photos, showing the daughter, mum and the lady who lived on. Seems to me the daughter had a wider face and she's the one who lived on https://imgur.com/a/DsqLWdR

janmo2• 3 years ago

Probabilities can be tricky.

Many people here in the comments point out how unlikely it is that Jeanne Calment would have been able to pull off such an identity fraud. Let's say you believe there is a 0.00001% chance she could have done it.

Now compare that to, the probability of becoming the world oldest person to have ever lived, it is 1 in billions. Add to this that she outlive the second oldest person by 3 years. The probability is now somewhere in the 1 to 100 billions.

Then there is actually a 99.99999..% chance that Jeanne Calment was indeed a fraud.

The only way to prove her record beyond the reasonable doubt would be to do an extensive DNA testing with her deceased and living relatives as such tests have a one in billions accuracy as well, but this hasn't been done.

Here is an interesting paper on the subject: https://www.liebertpub.com/doi/10.1089/rej.2019.2227

philegibbs4• 3 years ago

This summary misses a lot of recent developments in the case that were published recently in a book "Jeanne Calment, the Secret of Longevity Unravelled"

New evidence includes examples of signed documents showing that her signature changed suddenly in 1933, about a year before Yvonne is supposed to have died. There is also testimony from the family of a doctor in Switzerland that it was Jeanne who was ill with tuberculosis in 1931, not Yvonne who had recovered by then. There is also analysis of her recorded interviews showing that she often spoke more like Yvonne.

troebr• 3 years ago

Jeanne Calment was a famous French supercentenarian (died in 1997 at 122), and there were suspicions around her age: https://en.wikipedia.org/wiki/Jeanne_Calment#Scepticism_rega...

philegibbs4• 3 years ago

New evidence for fraud is detailed in the book "Jeanne Calment, the Secret of Longevity Unravelled"

ilamont2• 3 years ago

The woman widely recognised as the longest-lived human in history may have stolen her identity as part of an elaborate tax evasion scheme, a group of researchers have claimed.

https://www.independent.co.uk/news/world/europe/world-oldest...

ilamont2• 3 years ago

In 1965, Raffray, a lawyer in the southern French city of Arles, thought he had hit on the real-estate version of a sure thing. The 47-year-old had signed a contract to buy an apartment from one of his clients “en viager”: a form of property sale by which the buyer makes a monthly payment until the seller’s death, when the property becomes theirs. His client, Jeanne Calment, was 90 and sprightly for her age; she liked to surprise people by leaping from her chair at the hairdresser. But still, it couldn’t be long: Raffray just had to shell out 2,500 francs a month and wait it out.

https://www.theguardian.com/science/2019/nov/30/oldest-woman...

QuercusMax• 3 years ago

The part about birthdates being on first of the month or divisible by 5 seems pretty weak to me. Records weren't great back then and many very old people may not actually know their true birthdate.

saveferris2• 3 years ago

There is some statistical thing about fraud and the frequency of certain numbers being made up. I don't recall it specifically but made up amounts, dates, number have certain clusters of numbers vs what normally occurs.

edit; didn't get all that was in my head out :-) So, it could be made up or support the fact that actual docs or good record keeping weren't a thing.

nebalee• 3 years ago

https://en.wikipedia.org/wiki/Benford%27s_law

actinium226• 3 years ago

I think you're thinking of Benford's law: https://en.wikipedia.org/wiki/Benford%27s_law

It's pretty strange, basically if you have a document with a bunch of numbers (say, some company's quarterly report), look at the leading digit of all the numbers. For some reason, numbers with a leading digit of 1 show up more often than those with a leading digit of 9 (i.e. 1,234,567 is more likely that 987,123).

You'd think there wouldn't be any particular pattern in the leading digit, I mean why should there be? But observational data seem to suggest a pattern.

So Benford's law can be used as a leading indicator of fraud. If you apply it to a quarterly report and there's an unusually high distribution of 8's, for example, then while you can't be certain that it's fraud, it might be flag to an inspector/regulator to take a closer look.

twblalock• 3 years ago

If you are going to make up a fake birthday and you think you might have trouble remembering it, picking a nice round number might make it easier.

lb1lf• 3 years ago

In Norway, the authorities had a problem with our SSN equivalent a few years ago - for decades, it had been SOP to assign any immigrant with unknown birth date the birth date January 1st.

Eventually, they ran out of valid SSNs for Jan 1st births in some years. (The number is on the form DDMMYY XXXYY, where XXX is assigned sequentially and YY are control digits.)

Hence, for any given date the system can accommodate 1000 people, plenty in a country with some 1000 births a week. Until you start assigning a certain date to people with unknown DOB, that is. They are now assigned a random date.

gifnamething4• 3 years ago

Records weren't great back then and many very old people may not actually know their true birthdate.

Exactly why they can't be trusted!

mfer3• 3 years ago

Taking data for a whole country and drawing conclusions about a specific region may not be so good. For example, in the paper there is a lot of talk about bad data in the US but when they turn to the blue zone in the US there is no issue with the data. It's considered well kept.

For other places, like Sardinia, they don't explain whether the variations (1st of a month, etc) are happening at the local level. Can the data there be trusted? We are left to draw conclusions because many gaps aren't explicitly filled in.

The blue zones project limited the places they would label a blue zone to those that had good data. Many places were rejected because the data wasn't good. Looking at those places, like the UK or France, and stating they have untrustworthy data actually confirms the blue zones work of rejecting claims there due to untrustworthy data.

int_19h2• 3 years ago

That's exactly the point - it means that someone "guessed" the date. But if they guessed the day, the year isn't reliable, either, especially at these time scales.

Here's an example of how a similar principle can be used to observe electoral fraud: https://en.wikipedia.org/wiki/2011_Russian_legislative_elect...

hackeraccount• 3 years ago

You'd have to compare with a control group. If being 100+ was strongly correlated with having a birthdate on the first of the month or such then... maybe be suspicious.

dukeofdoom• 3 years ago

Life expectancy is going down for Americans, despite far fewer smokers.

steveBK1237• 3 years ago

Some stuff is less damning than it sounds for example - "supercentenarian birthdates are concentrated on the first of the month and days divisible by five"

One of my parents comes from a poor rural area of Europe where well into the 60s, it was normal to have a "real" birthdate & an "official" birthdate. This was because the government only issued birth certificates in the nearby city, and being poor & rural, it took time to get there by bus, etc.

So I have aunts & uncles with real-official birthdate deltas of up to 2 weeks.

I'd imagine that going back to 1900s or 1880s, when travel was more difficult, who knows. Further, it is possible that the office in some of these areas for registering births was only open certain days of month/week, the further back you go, etc.

Maybe they only went into the city on market days when they had other business, on the first Monday of the following month.

Don't discount superstition and people registering births on nearest special days like saints days, etc.

Some of these abnormalities get lost to time. Many of my family didn't realize they had a real-official birthdate gap until 30+ years later when their mother told them.

forinti• 3 years ago

That's true. I know some old timers who were born in one place but registered later somewhere else.

This issue was even more pronounced in South America because of the distances.

gifnamething4• 3 years ago

You're just arguing in favour of birth dates being unreliable.

neaden12• 3 years ago

To +- a few weeks, not years.

FeteCommuniste• 3 years ago

Unreliability on the scale of weeks would be rather less significant than years or decades.

steveBK1237• 3 years ago

yes, thats exactly what I am arguing. that is - pointing out weird numeric improbabilities in birth dates is not damning in terms of strictly indicating pension fraud. they are unreliable as recently as 60 years ago, and certainly were even more unreliable 100+ years ago.

01100011• 3 years ago

Wife's grandma recently died in her mid 90s. No one in the family actually knew how old she was. In many developing nations, such recordkeeping was often oral and unreliable.

e40• 3 years ago

My grandma, born circa 1900, had a birthday of Feb 14. She loved Valentines Day. The cakes she would make for that day, and other goodies... so good. My mom and I are sure she chose that day. Her parents both died when she was 8-10. Consumption.

goodcanadian2• 3 years ago

I had a great aunt (died before I was born) who believed her birthday was two months later than it actually was. The date that everyone believed to be her birthday was exactly 9 months after her parents wedding (her real birthday only seven months after).

I had another great aunt (who I did know) whose drivers license said she was ten years younger than she actually was. They just took you at your word without verifying when she got it.

eviks• 3 years ago

Did they not allow reported dates in the certificate in the part of the world? Theoretically you could have issued the certificate in April recording the correct date of birth in January

jrumbut2• 3 years ago

People weren't as concerned with record keeping back then, these things were just gonna grow mold and feed mice in a drawer somewhere unless someone was trying to annul a marriage (for consanguinity) or argue over an inheritance.

Then the modern administrative state sprang up very fast and the generation born in the late 19th or early 20th century was caught in the middle.

steveBK1237• 3 years ago

No, it was government bureaucracy that was serious enough to only issue papers for the date on which you appeared..

but not serious enough to have staff distributed across the rural areas to actually be accessible to people outside the big cities..

It's easy to mandate things that are easily enforceable (date=today only legal birth cert). Doesn't mean its a useful mandate.

darth_avocado3• 3 years ago

My dad is 4 years younger than his age on paper. He had no birth certificate when he was born and when he was finally being put in school, being home schooled so far, was able to answer questions like a 3rd grader. So the school decided to put him in 4th grade and just made up his birth year (dad's family knew the date but that would make him ineligible for 4th grade).

steveBK1237• 3 years ago

While not as extreme, many of the real-official birthdays straddle month end / month starts.

In one case it kicked someones birthdate to a new calendar year and his pension kicks in a year later as a result.

They always joke the donkey was slow and cost him an extra year of work.

jefftk• 3 years ago

The school couldn't tell from looking at him that he was closer to Kindergarten-aged (5) than 4th grade (9) !?

adamc4• 3 years ago

My parents both started college in 1939 at the age of 16. Being promoted to grades beyond your age cohort was apparently not so uncommon then.

schiffern• 3 years ago

My dad is 4 years younger than his age on paper

What a coincidence, me too!!

Going forward, this is now my official excuse...

kortilla• 3 years ago

What would happen if he committed a crime at the paper age of 19 and he was actually only 15? Would they retroactively update the birth record?

docandrew2• 3 years ago

This could be ruled out as the explanation by looking at deceased records as well - if people who died around the age of 30 in the 1950s, 40 in the 1960s etc. also all had curiously “round” birthdays then we could chalk up the fraud to poor recordkeeping instead. I’d be kind of surprised if such a coincidence hadn’t been previously noted though, surrounding discussion of the “birthday paradox” for instance.

int_19h2• 3 years ago

This could describe why birthdays are concentrated on a particular day of the week, but what place would have market days or something similar running on a cadence that consistently falls on days of month divisible by 5?

notjulianjaynes• 3 years ago

Random guess, that works out to 1.4 times per week, annually. So maybe something like Friday 9-5, Sat 10-2 hours could explain it?

chaostheory• 3 years ago

Official and real birthdays is a thing in Asia too. Many boomers base their birthdays on Asian calendars instead of the Gregorian one. Another issue is related to buying immigration related identification.

colonwqbang2• 3 years ago

Interesting, but this should be easy to control for in a statistical study. E.g. suppose 20% of people generally have strange looking official birth dates, but 50% of supercentenarians do.

rovolo2• 3 years ago

This isn't a good control if the record accuracy depends on the age of person.

Imagine if you're trying to find medical fraud and you find out that a lot of supercentenarians have cancer. You try to control by comparing them to the average, and supercentenarians have higher rates of cancer than normal. Is that fraud, or does age lead to higher cancer rates?

elevaet• 3 years ago

Another thing that happens is being born in a place that does birthdays on a lunar calendar, and then immigrating to a place that uses gregorian/solar, and needing to come up with a birthday in the new system.

raldi2• 3 years ago

The key question is, are living people born in 1923 more or less likely to be born on the first of the month than dead people born in 1923?

whyenot• 3 years ago

What do you think that will tell you? For example it seems to me that people born in more rural area a century ago are both less likely to have accurate birth records, and less likely to have good access to health care. There are probably many other covariates that would need to be controlled for.

Spooky232• 3 years ago

That’s not what the abstract said. It said “the first of the month and days divisible by five”. So 1, 5, 10, 15, 20, 25, 30.

Also, human birthdays aren’t evenly distributed. People tend to have sex around holidays and cluster somewhat. Late August through Early October are popular, and April and May less so.

So the statistical significance of the “round number” no doubt exists, but may not be as significant as it sounds to normal modern people in Western society. My grandfather, for example, was reported to the authorities 8 weeks as his country of birth had a poll tax… in his country baptism records are more accurate than civil birth records.

koheripbal• 3 years ago

My great grandmother was from a rural poor area.

She only knew that she was born in the spring in "the year of the big flood" which neither she nor local records could identify.

When we brought her to city hall, the just asked us what birthday she wanted and she picked her favorite number in her favorite month in spring and we guessed a year which was probably +/- 2-3 years correct.

Lots of bad data with lots of human bias.

nomel2• 3 years ago

and being poor & rural, it took time to get there by bus, etc

My naive assumption is that they might be concentrated around Friday, or Saturday, or some other more "convenient" day.

epolanski• 3 years ago

My in law was born on the 29th of February but registered on March 2 because parents thought it was bad luck to be born on Feb 29 and they were busy on March 1.

That was in Italy in 1968.

My polish father, born in 1954 was registered 13 months later, I don't remember the reason though.

vba616• 3 years ago

This was because the government only issued birth certificates in the nearby city, and being poor & rural, it took time to get there by bus, etc.

I guess I don't see the connection.

Why wouldn't it take a week to get the certificate for most people, and why wouldn't the authorities distinguish the birth date from the filing date?

In the US, the sort of birth certificate I have seen - from less than a hundred years ago - has three dates: Born on, Date filed, Date issued. The last two might be the same, but different from "Born on".

People who were born in the 19th century, on the other hand, may not have any documentation at all. I don't know the details, but I remember it being mentioned that nobody knew exactly how old my grandmother was, least of all herself.

I'm not sure it was a matter of transportation or poverty necessarily - just that everything wasn't controlled by documentation back then. Somewhere I think I read that passports weren't a thing until some time in the early 20th century. Maybe it changed because of WWI?

NeoTar• 3 years ago

I think the implied missing detail is that there was, or was perceived to be, a requirement to register a birth in a timely manner - currently in the UK there is a requirement to register a birth is within 42 days (6 weeks), but I have in my head that at some point it was two weeks.

Therefore people might be tempted to lie, saying that the babe was born within the registration deadline.

Or it was just that in the time between the birth happening and being registered they forgot the exact date, and either rounded it themselves, or it was done by the official registering the birth due to imprecise data ("When was the baby born?" "Two and a half weeks ago").

mfer3• 3 years ago

Some stuff is less damning than it sounds

This takes the paper at face value and I would suggest the paper is misleading. For example, the paper says...

However, alternative explanations for the distribution of remarkable age records appear to have been overlooked or downplayed.

After reading through the original blue zones work and this paper, they didn't make a case that alternative explanations were overlooked or downplayed.

For example, when they did the blue zones work they had to spend a lot of time making sure the data was good and many places were dismissed for bad data. This paper puts a lot of focus on bad data. No kidding.

The paper puts a lot of time into covering bad data in the US. But, then doesn't take the time to point out any issues with the US blue zone.

In a place like where Sardinia Italy they call out things like the murder rate being high and a low survivability rate to 55. Except it doesn't look at people in different age groups where "blue zones" tend to be going away as younger generations eat a different diet.

There are levels to the analysis that are missing which makes the conclusions a bit misleading.

avazhi• 3 years ago

That isn't less damning than it sounds. Whether it was overt or 'accidentally' (as in your case) the data is still unreliable, which is the point.

kijin2• 3 years ago

There's probably nothing malicious about most cases of missing or strangely uniform birth dates. In many parts of the world over a century ago, people simply didn't bother with accurate records.

For example, both my grandpa and grandma had two birthdays each, in two different calendar systems, that pointed to wildly different points in time. Nobody remembered exactly when they were born. My other grandma was recorded as being four years older than she thought she actually was, and nobody knows the truth, either. My father's birth was filed with the authorities several years late, though the document itself pointed to the correct date. My family's not from some sort of jungle, either. All of this happened in a highly bureaucratized, highly literate society.

Go back a few more decades and one could easily imagine "She was born in the spring, in the year of the great flood" becoming "Let's just say she was born on April 1, and when was the flood? I mean the second one after Steve became king" when modern record-keepers demand a specific date. We're trying to see more precision in the data than anyone ever intended to record. No wonder we find artifacts.

VintageCool• 3 years ago

Malicious no, suspicious yes. There likely were parts of the world a century ago without accurate records. It also appears that supercentenarians are concentrated in places that didn't keep accurate birth records around the time they were born.

There have been people who attempted to adopt the diets of the "blue zones" where many supercentenarians have lived, but maybe in our search for long life we should have been adopting regional poverty, low incomes, and dodgy record-keeping!

ip262• 3 years ago

It is perfectly possible to be correlated. People almost universally eat more meat as they gain wealth. So if a diet of mostly plants really was the way to go, you aren’t going to find it in upper Manhattan.

kijin2• 3 years ago

We need to look for patterns in the birth records of people born around the same time in the same area, but who died at a "normal" age. If the same patterns occur in both cases, it's just lossy record-keeping. If not, there might be something dodgy about the supercentenarians in particular. Without a proper control group, it's all speculation.

pharrington• 3 years ago

Older countries have better records?

westcort• 3 years ago

Taken another way, imagine a future where dramatically longer human lifespans are possible. Assuming there was fraud here, normal-lifespan individuals could pretend to be a single individual for an even greater period of time.

Even with current technology, foundations can carry on a person’s wishes far into the future. Imagine if a personalized large language model were developed to reliably predict an individual’s future verbal utterances. Could a large language model trained on a large enough corpus of data predict the next thing a living person would do or say? If so, could there be an option to transfer personhood to the language model after that person’s death?

Before judging this as impossible, think of how well our voices can be replicated by AI. As Stephen Wolfram has pointed out, this process must necessarily entail modeling the part of the cerebral cortex that produces speech.

hobo_in_library• 3 years ago

Best would be to see such an LLM being put in charge while the person is still alive, and then laugh every time the person is frustrated with a decision the LLM decided to take.

francisofascii• 3 years ago

My wife’s grandmother lived to 110. There was never any doubt about it. She knew what year she was born. Her mother had lived to 99.

HideousKojima• 3 years ago

My great grandmother died a month shy of her 110th birthday. The main reason I don't doubt it was true is that my grandma (her daughter) is in fantastic health despite being the same age as my grandpa who is in fairly poor health, so there's definitely something in my great grandma's genes that seems to aid longevity. Unfortunately it's looking like my dad got a decent amount of my grandpa's genes instead of my grandma's in that regard, so I'm doubtful I'll be so lucky haha.

kepler1• 3 years ago

I wonder if a lot of things in society are going to require some kind of physical in-person proof because of our inability to distinguish fake from real at some point soon.

docandrew2• 3 years ago

It would be funny if all the hype about “Mediterranean Diet” and longevity was just due to pension fraud in those areas.

shaky-carrousel2• 3 years ago

A healthy diet not only makes you live longer. It also makes the experience less miserable when you reach an old age. That reason alone should be good enough.

tsimionescu4• 3 years ago

The question is what constitutes a healthy diet, and particularly if the Mediterranean Diet is a particularly healthy diet or not.

james4k• 3 years ago

It would be funny, but we do have a lot of other data around disease progression that tends to agree.

Kosirich• 3 years ago

There is also the "hill argument". Mediterranean is hilly, promoting short periods of physically intense exercise which has been proven to be vital for health of cardiovascular system (huge UK wearables study)...aka often walking uphill carrying stuff is good for you

initramfs• 3 years ago

The pessimist in me thinks, the longer one lives (in their 90s and 100s), the more this database analysis could discriminate using this algorithm, and flag healthy, retired centenarians from getting benefits, hopefully not putting a freeze to their accounts or causing any stir to their peaceful retirement.

davidgerard• 3 years ago

(2020)

robwwilliams• 3 years ago

This was posted to bioRxiv originally in July 2019. You reasonably ask why the peer-reviewed version hasn’t been published yet and why 3.5 year work with 6 citations in a hot area of research is now distracting us from ChatGPT.

A colleague of mine points out that for most subjects over 110 the gero-forensics seems to be well done (https://www.researchgate.net/profile/Nikolay-Zak-2/publicati... ).

Interesting summary on page 31, table 2, and the inevitable shoutout or shout-down of the venerable Jeanne Calment.

poulpy123• 3 years ago

I have a big problem with that:

Her original validators have demanded that their opponents provide proof of the switch [7]. However, it is a central tenet of longevity validation that the onus of proof is on the validators [2,3].

The validators have proved that Jeanne Calment was Jeanne Calment. They may be have made mistakes or even lied from the beginning but in any case it is not their role to prove that there was no switch but its the role of the accusers to prove that there was one.

It's the basis of science

dbg31415• 3 years ago

My dad was born on an Indian Reservation in South Dakota in the 1940s or maybe 30s. He has an age for the military, an age on his drivers license, and a bunch of ages he told people over the years. Mostly bankers. For probably 20 years whenever anyone asked him how old he was he would just say 39. He got a kick out of telling people the wrong age. He still does when he meets a new doctor.

Before his older sister died, I asked her how old Dad actually was, and she said he was “about” 5 years younger than her. And her birth certificate just said “September 1933” - it’s possible the date had rubbed off but it wasn’t there.

For anyone who has ever done any genealogy work, a lot of the official census dates were just guessed at. Spellings were made up. Birthdays were made up. Kids were made up… hired hand? Count him as a kid. Second wife? Count her as a kid. It’s just some guy going door-to-door asking you how many kids you have and when were they born and he’s getting paid by the number of houses he completes that day. Does that sound prone to accuracy?

I have some Mormon relatives and it’s always really interesting how hard it is to read that old scribble from the census and family trees, the official census takers handwriting was often sloppy cursive. And the 80 year-old genealogy worker in 1974 didn’t have much better handwriting. And especially in poor areas the government officials just didn’t care that much about getting correct numbers.

mfer3• 3 years ago

Interesting that the publication is from 2020 and doesn't appear to have ever been peer reviewed. Reading through it, the data doesn't support some of the conclusions. I would take this with a grain of salt.

Vibgyor5• 3 years ago

I myself have a difference of exactly one year from my real birthdate and official one.

Why? Because in 80s/90s India, it was common to have 'official' birthdate extended with expectation that this will result in one additional year of government benefits. That logic does not make sense to me but was prevalent at the time of my parents/grandparents etc.

It irks me a bit at times, however, there is nothing I can really do about it at this point - all my records, passport etc. are based on 'official' birthdate. To avoid any confusion or misunderstanding, I quote official birthdate to friends and other people. Realistically, only people who're aware of my actual birthdate are my family.

wolverine876• 3 years ago

(2020)

gpderetta• 3 years ago

This is the comment I posted last time the paper was here on HN (and once before): https://news.ycombinator.com/item?id=32149322

Tl;dr: while fraud or bad record keeping might indeed be involved, I think the article is flawed.

projektfu• 3 years ago

Did this article never go to print?

MagicMoonlight• 3 years ago

Any time you get someone significantly better that everyone else then it’s because they’re cheating.

It’s like with Usain Bolt, he was miles ahead of the other sprinters. And they had all been confirmed to have been cheating.

Supercentenarian records show patterns indicative of errors and pension fraud (2020)

Related Stories