Review of Longevity Validations at Extreme Ages

Abstract

The verification of human longevity is a hazardous process with many errors, false claims and myths among the true records. In the light of past withdrawn and disputed cases we conduct a sceptical reappraisal of validations from the top of the lists of world's oldest people. Longevity data is used increasingly by scientists with applications in research programs trying to understand and combat the processes of aging. Our goal is to evaluate the robustness and accuracy of the information being referenced. We identify weaknesses with the standards and practices in use but we also find significant new evidence for some important cases including Sarah Knauss. We propose ambitious new standards so that the process of validation can continue to improve with the benefits of new technology, including DNA testing.

Full text

1
Review of Longevity Validations at Extreme Ages
Philip Gibbs and Nikolay Zak
Abstract: The verification of human longevity is a hazardous process with many errors,
false claims and myths among the true records. In the light of past withdrawn and
disputed cases we conduct a sceptical reappraisal of validations from the top of the lists
of world’s oldest people. Longevity data is used increasingly by scientists with
applications in research programs trying to understand and combat the processes of
aging. Our goal is to evaluate the robustness and accuracy of the information being
referenced. We identify weaknesses with the standards and practices in use but we also
find significant new evidence for some important cases including Sarah Knauss. We
propose ambitious new standards so that the process of validation can continue to
improve with the benefits of new technology, including DNA testing.
Introduction
In recent years the validation of the ages of supercentenarians has been undertaken by
organisations such as the Gerontological Research Group (GRG), the International Database
on Longevity (IDL), Guinness World Records (GWR), individual scientific collaborations and
some independent longevity enthusiasts. There is significant overlap between these groups.
Towards the end of the 20th century, formal longevity validation was conducted mostly by the
Guinness Book of World Record (now Guinness World Records). Although they aimed for a
consistent standard, their purpose was commercial entertainment rather than science. The
case of Jeanne Calment drew increased interest from gerontologists and demographers,
accelerating the development of a more scientific approach.
In 2002 an international project to compile a database of supercentenarians was initiated by
these scientists. Individual teams were formed to cover the countries with the highest
numbers of longevity cases. Studies assisted by government agencies were used to put
together lists that were hoped to be reliable, complete and unbiased. The validation criteria
used for the IDL were variable. Database entries were labelled with an “A” where a minimum
of two documents were used to verify birth and death. IDL is anonymous and relies on the
assumed competence of government records and studies [31].
The GRG was originally formed in 1990 with the goal of seeking medical means to slow and
ultimately reverse aging [32], but following the foundation of the IDL, the task of maintaining
lists of supercentenarians became its most visible activity. They have since become the
principle organisation for longevity validation. Their initial network of country correspondents
had a strong overlap with the teams that produced the IDL but motivation and the validation
procedures used were not the same.

2
Calment’s Validators defined five criteria for validating longevity claims but still left much
room for interpretation [19]. The declared criteria used for validation by the GRG [33] are a
weak interpretation of this standard, yet still an improvement on the IDL. The GRG require a
minimum of three documents covering beginning, middle and end of life. The birth record is
only required to originate from the first twenty years of life, leaving the risk that an early age
exaggeration or identity switch could be overlooked. It is one of our aims here to determine
whether this and other weaknesses have had any negative impact on the reliability of
validation.
The IDL was aimed at scientists and produced anonymous demographic lists, whereas the
GRG published names and worked also with GWR. This may have given the GRG a strong bias
towards validation at record ages. A comparison of their Table B list from 2015 [15] with the
latest list [17,42,43] shows that in a space of five years the number of validated women who
died in their 115th year increased by only 3 from 22 to 25, while the number at 116 increased
by 5 from 6 to 11 and the number at 117 years old increased by 5 from 2 to 7. This is
significantly different from what would be predicted by any mortality model.
It has long been recognised that longevity records are plagued by erroneous and fraudulent
claims so that the validation process must be rigorous, yet it cannot be made so strict that no
one can pass the test [1,2,3]. Even with due care taken, a number of cases that were
previously validated have been subsequently removed from lists, such as those of Shigechiyo
Izumi [3,4,29,39,41], Carrie C. White [5], and Kamato Hongo [2]. Recently the investigations of
Nikolay Zak have cast doubt on the validation of Jeanne Calment, the longevity record holder
at 122 years and 164 days [6]. Although two of the original validators (Robine, Allard) and
their collaborators (Herrmann, Jeune) have reaffirmed their work [7], they have not
addressed the conclusions of our study applying Bayesian inference to the question of her
authenticity [8]. There are other disputed cases that remain in the validation lists, including
the longevity of Lucy Hannah and Mathew Beard, while the evidence for the validations of
Nabi Tajima, Violet Brown and the majority further down the list have not been subjected to
full scientific review and publication.
Scientific interest in longevity records is increasing. There is no consensus on the correct
model of mortality rates at high ages. Some demographers believe that there is a mortality
plateau with the annual rate of death limited to a value around 50% above 105 years [9].
Others have suggested that the annual survival rate continues to diminish towards zero,
imposing a soft limit on the natural human lifespan at around 115 years [10]. However,
mortality statistics might be plagued by false data, so that conclusions could be misleading
[2,11,12]. These are fundamental issues in the field of gerontology with implications for the
nature of the aging process and potential applications to elderly care and life extension. Even
single cases such as those of Jeanne Calment and Sarah Knauss (119 years and 97 days) can
be statistically significant because they appear as outliers which seem to refute the hypothesis
of a maximum age. The emerging field of biomolecular studies exploring genetic [13] and

3
environmental [14] causes of exceptional longevity, also depends increasingly on the
reliability and accuracy of longevity validation.
There is a human side to longevity studies and it is understandable that validators want to
give the benefit of doubt in individual cases where personal stories and national pride are at
stake. This may be acceptable for the entertainment side of world records but it could hinder
the attainment of scientific goals. In view of the increasing scientific importance and the
significant rate of validation failures, we propose to critically reappraise the accepted cases
from the top of the record lists to see what lessons can be learnt. The validation of Jeanne
Calment is often cited as a gold standard. Most other cases have been subjected to far less
scrutiny.
It may be impossible to upgrade the validation status of past longevity claimants due to a lack
of vital data from the times and places of their early life. The absence of genetic samples, or
the lack of access to them where they have been taken and stored, is also a hindrance. DNA
testing could be a powerful tool for improving the quality of the validation process going
forward. This would require samples from living relatives as well as the main subject. Where
samples have been taken in the past they have been for studies which assume the correctness
of the age validation rather than a part of it.
We aim to outline a stronger “cast-iron” standard for the purposes of future scientific studies.
A questionnaire is offered to validators to help improve consistency and completeness of their
reporting. In order to inform this standard we looked at individual cases of validation,
including those that have been withdrawn or disputed. The aim is to understand their
weaknesses and how better records and DNA tests could help with the validation of future
similar claims.
Validation Reviews
Our first requirement in order to undertake this study was to find an official list of validated
supercentenarians over 110 years of age, past and present. Guinness World Records certifies
record holders but we also need to include runners up. It is possible that some of these are
overlooked because there is less incentive to validate them. The GRG has a list of 1739
supercentenarians as of 1 January 2015 [15]. Lists of living and recently deceased
supercentenarians have been published annually in Rejuvenation Research [16] but these
would need to be combined to provide an all-time list. The best sources that do this are the
Gerontology Wiki and Wikipedia, which provides lists of the all-time oldest verified hundred
men and women [17,42,43]. The IDL provides only anonymised lists which do not suit our
purpose.
For the archicentenarian cases that follow we have used published validation reports where
available, to conduct our review. For a few, mostly in the US and Canada we have been able
to do our own searches on ancestry.com, findmypast.com, familysearch.org,

4
newspapers.com and elsewhere. We were also able to research Danish newspapers and
archives with the help of Per Hagemann. In some cases we located important new documents.
It is likely that additional relevant details are held by validators and others. We have not tried
to contact them systematically because we wish to evaluate what is now publically available,
but we have had some helpful private communications about a few cases. We encourage
validators and other researchers to publish all available information that they have in more
complete form. Where privacy is a real concern a restricted repository of documents may be
required. Any data that is not published is at risk of being lost making any future re-
evaluations in the light of new information hard to carry out.
Summaries for top listed women, validated or formerly validated.
Jeanne Calment - 122 years, 164 days, France
French supercentenarian Jeanne Calment holds the world record for the oldest validated
human [18,19,20]. Her validation has been disputed by Valery Novoselov, Nikolay Zak and
others. We reaffirm here that we continue to dispute her case due to an ambiguity in her
record which could allow for a mother-daughter identity switch [6,8]. 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 dispute over Jeanne Calment’s age is not based on the “impossibility” of her longevity. It
may well be possible for someone to live beyond 122 years of age. The argument against her
is that a specific identity switch scenario is found to be significantly more probable in a
Bayesian analysis of the known evidence. In brief it is thought that both mother and daughter
contracted tuberculosis which was considered shameful for people of their social standing.
They tried to hide it and became trapped by signature fraud which would expose a scandal if
the identity switch was not hidden. Financial advantages of the switch may also have been a
motivating factor. Claims that the daughter was seriously ill from 1928 to 1934 are refuted
by later photos of her at a wedding and a costume festival, while similar photos of the mother
are conspicuously missing. We believe it was the mother who died in 1934 and not the
daughter as claimed.
After extensive investigations it seems unlikely that documentary evidence including public
records, photographs, testimony and news reports will be sufficient to resolve Jeanne
Calment’s case to the point of forming a scientific consensus. DNA testing on the other hand
is capable of settling the matter. It is known that Jeanne’s daughter Yvonne was the offspring
of double second cousins. This would be identifiable in her DNA via an autozygosity test and
would distinguish her from her mother with a high degree of certainty. Consanguinity tests
with suitably chosen living relatives would also be deterministic.
DNA samples for Mme Calment were taken and kept by the Jean-Dausset foundation in Paris.
Unfortunately, requests for access or even confirmation of their continued existence have so

5
far not been answered. A principle lesson to be learnt from this case is therefore that bio-
samples taken should be held in trust under terms of use that are made public. It must be
appreciated that DNA sequences hold enormous amounts of information about a person and
their close family. It may not be appropriate or even lawful to publish the whole genome
sequence, but it could be made available for legitimate scientific investigations under suitable
conditions imposed according to family wishes.
Sarah Knauss – 119 years, 97 days, United States
Born as Sarah Deremer Clark, she is the only other supercentenarian currently validated to
have lived beyond her 118th birthday. Her case was originally researched in 1997 by
genealogist Edith Rodgers Mayer for Phoebe Ministries [21] but the original documentation
is no longer available. The GWR announced her record that year, and then in November 1998
a team of gerontologists (Perls, Robine, Wilmoth, Jeune) visited her and announced that “Her
age will now be thoroughly validated” [22]. Some details of the validation were given in 2002
(Robine, Vaupel) [23], and a further article was published in 2010 [18]. The most complete
report available on the evidence for her validation was provided by Robert Young, also in 2010
[5].
According to her longevity claim, Sarah was born in Hollywood, Luzerne, Pennsylvania on 24
Sept 1880, but civil birth records were not taken there until 1906. Furthermore the crucial
1890 census returns were destroyed after a fire (but see below) and her church baptism
records have not been found. According to the validation reports, the earliest records
confirming her year and month of birth are the 1900 census and her marriage in 1901, i.e.
nearly two decades later. A 20 year gap in the record from birth makes the validation of such
an extreme longevity very unconvincing. Taking a sceptical stand, the possibility that she
could have been a few years younger than claimed must be considered. Perhaps her age could
have been exaggerated to make an underage marriage respectable or for other reasons. A
picture of her on her wedding day shows her to have been very young looking.
An opposing argument advanced by Young is that she had younger brothers whose baptisms
were recorded in the church records after the family moved to Bethlehem, Pennsylvania. If
she was younger, she had to be at least 9 months separated from their births, but even
assuming that she was born in the month of September there are slots in 1881, 1882, 1883,
1886 and 1887 where she could be born. If her birth place was Hollywood as claimed then we
can understand the lack of baptism record because not all the registers from churches in the
area have been found for that time. A better argument for her age is therefore that if she
were born after they moved to Bethlehem, then her baptism would have been recorded
there, whereas a baptism record from before the move could easily be lost.
We know from an article we found in a news archive search that they still lived in Hollywood
on 21st December 1881 when the Hazleton Sentinel reported that “Bertie” the five year old
son of Walter Clark died of membranous croup (diphtheria). A report on the 115th birthday of

6
Knauss in The Morning Call said they moved when Sarah was 18 months old. If Sarah was
much younger, her baptism would have been recorded in the same church records as her
younger brothers Walter and Edward. This constraint would mean that she could only be one
or two years younger than claimed, but this is based on testimonial evidence for the date of
their move which may be unreliable. Further corroboration is needed.
Sarah Clark’s middle name Deremer is the married name of her aunt who was also born as
Sarah Clark. Sarah and her husband Abile Deremer lived beyond their 65th wedding
anniversary celebrated in 1928, but no children were recorded from their union. After they
both died in 1931 a notice in the Wilkes-Barre Record of 17 June 1931 indicates that Sarah
Knauss was given a 20th share of their inheritance, as were Knauss’s daughter and
grandchildren. Less was left to other surviving members of Walter Clark’s family. This
favouritism can easily be explained by a special friendship, or if Sarah Deremer was her
godmother, but another possibility that cannot be ignored is that Sarah Deremer was in fact
the real mother of Sarah Knauss. She could have given her into the keeping of her brother’s
family due to hardship. Such informal adoptions could be hidden.
There is a mystery surrounding the Clark family’s sixth child who first appeared with the name
Earl on the 1900 census. There is no record of his baptism in the church register despite the
baptism of his older brothers and younger sister being found. From 1910 his name changes
to Foster Earl. Was he given the nickname Foster because he was adopted? In carrying out
our validation review we could see that more evidence was needed to refute adoption
scenarios.
Our searches have uncovered an extensive genealogy for Sarah Knauss’s family. Her father
Walter Clark had a brother and two sisters, plus two additional half-sisters from his step father
Benjamin Miller. Local news reports show that in adulthood, some of their families led
troubled lives. Crime rates were increasing and record keeping was poor, a common feature
for areas with misrepresentation of longevity [11]. Notable cases are Emma J Deremer (b.
1865), niece of Abile Deremer who married Dr. S. L. Good in 1888 and died in 1891 and Mary
Alice Miller (b. 1864), half-sister of Walter Clark who married G. A. Ueberroth in 1882, and
then Charles M. Hinkle in 1891 after the death of her first husband in 1890. Sarah’s mother
Amelia also had numerous siblings and there were other relatives who could have had
children at this time. Informal adoptions within families were historically common and can
often only be uncovered by DNA testing. If this were the origin for Sarah Clark then there
might be less certainty over her birth date. She could have been born to a different mother
at the same time as her brothers and could have been baptised anywhere.
While trying to find more information on Sarah Clark, we became aware of reports on
supercentenarian internet forums of a claim that there is a church record of her religious
confirmation in 1895, but with no age or birth date recorded. Confirmations in most Christian
churches are made from age 14 onwards, so this would be a good indication that her birth
date is roughly correct, but in some churches the sacrament can take place at a younger age.

7
We searched for this record in the registers of the Cathedral Church of the Nativity in
Bethlehem where some of the later Clark children had been baptised. We found a record from
Easter 1895 for Sarah Clark. It is most likely the right person because she is listed after a Mrs
A Clark who would be her mother (page 266 of the register for 1888-1912). The entry is listed
under communions rather than confirmations but communion can happen at the same time
as confirmation. Although Sarah’s age is not given, earlier records from just a few years earlier
did record age and showed that confirmation in this church could be conducted at an age as
young as 11. The conclusion from this evidence is therefore not as strong as others were
claiming but it does suggest that she was probably born no later than 1884.
Figure 1 – Earliest known record of Sarah Clark, c. 1888
We looked again at the church registers for Bethlehem and we found another helpful record
that other researchers appear to have missed. In the index of the volume for the period 1862
to 1888 the entire Clark family as it had been at that time is listed with parents first, followed
by the children in age order: Walter, Amelia, Charles, Sarah, Walter and Edward (Figure 1).
The last two died as infants and had been recorded as baptised and buried, but this index
gives us a place where Sarah is mentioned before her communion. It records that she had
been baptised. The record for this baptism must be elsewhere (possibly a church near
Hollywood.) It is plausible that the rector in Bethlehem would have checked it but we can’t
be sure. The exact date when this index was written is not noted. If it was after 1893 the
younger sister Emily would be included. Edward was buried in 1890 and the record for this
seems to have been added. We think the index was probably written when a new volume
was started in 1888. This then is the earliest known record of Sarah’s name and it confirms
that she was born before Walter in July 1884. Since she is not in the 1880 US census, this
implies that she was born in the interval from 1880 to 1883.

8
On the 28th August 1901 Sarah Clark married Abraham Knauss in the same church where she
had been confirmed six years earlier. She was a few days short of her 21st birthday if her
claimed birth date is correct. The Rev Gilbert H Sterling presided and entered the age of 20 in
the parish register which he had been signing since at least 1892. He had arrived there as the
new rector in June of that year from a previous calling in New York. He would probably have
known the family well by the time of Sarah’s marriage. He had baptised her youngest sister
in 1893 and could have encountered Sarah and her mother at bible classes in preparation for
their confirmation. He may have checked that she was baptised in order to ensure that
confirmation could go ahead, but birth date is not always recorded on the baptism record so
there is no guarantee that her birth year was correct. He would not have known the family
before 1892 when he was elsewhere so he might not spot an age discrepancy of one to three
years.
Later identity switch opportunities for Knauss must also be considered. The families of her
daughter Katherine, brother Charles and sister Emily would be expected to be aware of any
switch. There are photos of Sarah starting from 1897 where some resemblance is apparent.
There is enough information on Sarah’s social activity in newspaper reports to give confidence
that she did not switch at least until her husband Abraham’s death in 1965. Later the presence
of her family makes a switch improbable. We therefore do not consider a post-marriage
switch to be a viable option on current evidence.
Before completing our review we made one last search online for missed early records of
Sarah Clark, and we had a lucky break. As stated earlier, the 1890 census returns were lost
after being damaged by fire. Before that in 1891, Joseph H. Werner made an index of the
returns for the area of Northampton County including Bethlehem where the Clark family
lived. This has been preserved and a copy of its contents are available at
http://www.bethlehempaonline.com/beth1890/abc.html Here we found a copy of the
returns for the Clark family as follows [24]: Clark Walter 41, carpenter--Amelia 33, Charles H
12, Sarah D 10, Foster E 1. 524 Broad. This clearly shows the members of the family living at
the time, including Sarah D Clark with an age of 10. The 1890 census was taken in the month
of June so her correct age should have been 9, but this is good enough to rule out a later age
exaggeration for the purposes of marriage. It is worth mentioning that archive.org shows this
web page to have been in existence since at least 2008. It would be worthwhile to find a copy
of the original publication to verify the authenticity of this record.
We approached the age claim of Sarah Knauss with some scepticism due to her being a year
and a half older than the next placed supercentenarian. The previous validations were weak
because of the lack of early birth records. Alternative scenarios were considered and could
not be unambiguously ruled out. The new records that we found for Sarah herself show that
she reached the age of at least 116 and there is a good probability she did indeed live to over
119 years of age. These discoveries have strengthened the validation. The high level of
consistency in her story and records is impressive, yet because of the missing baptism record

9
and no confirmation of birth date before 1890, we could not completely rule out the error of
up to three years. It has to be taken into account that she is an outlier in the longevity lists.
For a cast-iron standard we consider it necessary to have evidence of date of birth from the
earliest years of life. Without such evidence there is no solid argument against alternative
birth scenarios where origins are uncertain, or age exaggeration for early school entry or
other motives.
There is still hope that her baptism record might be found. Some church records from near
her place and time of birth have not been scanned or indexed. In Bethlehem they attended
an Episcopal church which suggests that St. Peter's Episcopal Church in Hazleton is one place
where the family could have gone to worship when they lived in Hollywood. Unlike other
churches in the town, its records have not been published. The original parish registers are
held at the University of Pennsylvania library requiring permission from the church for access.
However, we made enquiries through a church official who kindly checked the registers for
us and informed us that the Clarks were not recorded there. Another possibility is St. Paul
United Methodist Church in Hazleton where Emma Deremer married in 1888. The records for
this church are held at the Historic St. George Museum and Archives to which we have made
a request and hope to get the information soon.
The most likely candidate for Sarah’s baptism venue seems to be St. Paul United Methodist
church in Drums, a small town near Hollywood. According to the obituary of Walter’s
stepfather Benjamin Miller, the family bought Tinney Farm near the Methodist church in
Drums in 1880. It seems likely that the Clarks living nearby would have joined Walter’s step-
family for worship. We have made contact with the church and have had helpful responses
even from the Methodist bishop. Unfortunately, the church informed us that they have not
preserved the old records. There is still a hope that the records could be found elsewhere.
As for other validations, once again DNA evidence could have been used to rule out the
alternative adoption hypothesis and other potential identity switch scenarios. This could still
be possible using the DNA of her descendants and relatives if they are willing to participate in
tests.
Nabi Tajima - 117 years, 260 days, Japan
Nabi Tajima is listed as the third oldest woman ever in GRG records, yet questions about the
accuracy of her birth record have been raised. Her case is marked as “disputable” on
Gerontology Wiki pages [42]. Tajima is said to have been born in the Kagoshima region of
Japan in 1900. Despite her high placing, no scientific record of her validation has been
published yet and detailed information is hard to come by.
A number of potential problems with the Japanese system of Koseki records have been raised
[3,34,38,39]. In particular, there was a custom in Kagoshima of using the registration of a
deceased first child for its younger siblings [38,39]. Is it possible that the birth record of Nabi

10
Tajima has been compromised? A full report is needed from Japanese validators to clarify the
real situation.
Tajima had a son born in 1919 when she would have been about 19 years old. This is
sometimes cited as evidence that she could have been at most two years younger than her
claimed age. In fact it is possible to give birth from thirteen years or younger, and an underage
pregnancy is a strong motivation for a woman to exaggerate her age.
Her validators may have the evidence necessary to rule out these possibilities as well as
potential identity switches and we are looking forward to assessing the validation report if it
appears in [35].
Lucy Hannah - 117 years, 248 days, United States
Although Lucy Hannah still remains on the oldest women’s lists in fourth place [17] and her
age is present in the IDL database, her former validation is now widely assumed to be
incorrect and her case is marked as “disputable” [42]. Investigations are said to reveal that
her true birth date was probably ten to twenty years later than claimed, but no invalidation
report has yet been published to document the evidence.
It is said that at her death in 1993 her family claimed that she was 118 years of age. This does
not appear to have received any press attention at the time. We have been unable to find any
news reports. There are no confirmed photographs of her at any age and no family that
outlived her have been identified.
In 2003 the US Social Security Administration validated her age as 117 in a large study of US
supercentenarians [25]. They appear to have based their conclusions on an unpublished
record from 1972. There was also a marriage certificate from 1943 which gave her parent’s
names as Solomon Terrell and Sallie Edwards. If this is correct then she was a widow from an
earlier marriage with surname Brown. A family with parents Sol and Sallie Terrill, her brother
with surname Edward and 4 children was found in Tallassee, Alabama in the 1880 census, but
Lucy is not present. Instead she has been identified as a Lucy Terrell age 4 living with
grandparents Thomas and Sarah Terrell in Linden, Alabama 130 miles away. This validation
was accepted by other validation bodies including the GRG, GWR and IDL, and she still remains
on official lists [16].
A subsequent investigation by enthusiasts noted that the date of birth according to the
marriage certificate would be 1895, much later than the claim. It is not uncommon for an
older woman to give a younger age on a marriage certificate to match her husband’s age, so
this alone is not conclusive. A 1930 census record for a widowed Lucy Brown living with two
Terrell nieces appears to confirm the later birthdate. A marriage record for Solomon Terrell
and Sallie Edwards in 1871 suggests that the family in the 1880 census could be the right one,
but Lucy could have been born later when her mother was about 40 years old. This is
sufficient to cast serious doubt on her validation.

11
Note that some sources claimed she married a John Hannah in 1901 and had 8 children two
of whom outlived her [40]. This appears to have originated from a made-up story edited into
Wikipedia that was subsequently removed due to it being unsourced. A photo sometimes said
to be of her has also been debunked.
In our opinion, very little is established concerning her identity. We are not yet convinced that
any of the records including the marriage certificate are connected to her. There is a distinct
lack of records linking events in her life. Even her race is uncertain. It is possible that further
census records that would cast more light on this case can be found.
Based on our observations we dispute the Lucy Hannah validation and deem it to have been
invalidated by a severe lack of evidence supporting her claim. We are expecting a GRG
endorsed report on the Lucy Hannah case to appear soon.
This example demonstrates that government agencies should not be blindly trusted for
validation. More generally, no validation should be accepted without a published list of
sources for evidence used, and preferably copies of documents where possible. We are
surprised that she has not yet been removed from validation lists. Validation groups seem to
work on the principle that once validated, a case can only be invalidated where there is proof
that the original conclusion was wrong [7]. We do not accept this protocol. A validation should
only stand if the available evidence would confirm it using current validation standards. This
is one reason why it is important for all validation evidence to be published. The case of Lucy
Hannah attracted widespread attention because of her extreme age but other cases from USA
in IDL list were validated with similar methods which raises questions on validity of these data.
Marie-Louise Meilleur - 117 years, 230 days, Canada
At fifth place on the women’s list, Marie-Louise Meilleur is another verified supercentenarian
alongside Sarah Knauss for whom we believe that a strong validation may be achievable with
the available evidence. A validation report is available (Desjardins) [26] and additional
biographical notes were published [18]. Her baptism record from 1880 in Kamouraska,
Quebec can be readily checked online. It took place just two days after her birth. She had two
sisters whose lives are accounted for. She does however appear to have one opportunity in
her family history where an identity switch could have been possible. Between 1910 and 1912
her first husband Étienne Leclerc died followed by both her parents. The death record of
Étienne says he died in February 1911 but he was later recorded with his family in the census
of that year. In 1913 she left her home town of Quebec to live with her sister Albertine in the
country. We can confirm that her three young children moved with her, or to join her, as they
were recorded on the 1921 Canadian census together.
Testimonies of these children reduce the probability of an identity switch but they were very
young when it could have happened. She is said to have returned home only once for a visit
many years later. She remarried to Hector Meilleur in 1915. Her age was recorded as 32 in

12
the marriage certificate and her father was listed as a farmer (not deceased as it should have
been). Hector had children from a previous marriage, and they had further children together.
According to later newspaper reports, all the children called her “Nanny” which suggests she
could have worked as a childminder for the family before her second marriage.
Where there are dramatic events of this nature in someone’s life, there is an opportunity for
an identity switch. Bereavement and estrangement from many of her relatives and friends
means that this could go undetected. We do not consider such a switch to be likely in this
case, but in a Bayesian analysis of longevity claims, conditional probability can be higher even
for unlikely switch scenarios because of mortality evidence (see ref [8]). Such events should
be carefully scrutinised. To improve certainty we would like to see a documented history of
those times with sourced evidence. Her eldest daughter also named Marie-Louise (1901-
1940) needs to be accounted for. She was married to Joseph Bossé and her last child André
Bossé was probably born 25 Dec 1939 so more detailed evidence is likely to rule out this
switch. A swap with her sister Albertine seems to be unlikely because Albertine was married
to Jules Chuinard (1872-1952). They had four children recorded in the 1911 census in
Kamouraska and seven children in 1921 census in Des Joachims village. Assuming that a sibling
swap can be ruled out, it might be possible to eliminate other switch scenarios such as with
another relative or nanny by comparing DNA from descendants of her first marriage with that
of those from her second.
Violet Brown - 117 years, 189 days, Jamaica
At this time, no scientific report has provided details of the evidence used to validate the case
of Violet Brown. Official records, news reports and family testimony help fill some of the gaps,
but there is still too much we do not know.
Civil registration records her birth on 10 Mar 1900 in Duanvale, Trelawny, Jamaica. She had
at least 7 siblings including a sister 3 years younger who could be a potential swap partner.
Four children are also recorded; each one with just the mother’s name Violet Moss, but the
fathers were known and their family names were used later. The oldest was Harland
Fairweather born 1920 who immigrated to England. He returned to his homeland in his later
years and was himself notable in the record books as the oldest person with a living mother.
His brother Irving Adolph Russell born 1923 immigrated to the United States where he died
in 1989. Both these brothers have living descendants in their adopted country. Irving had a
twin sister Elsie Estelle Russell who outlived her mother. A third younger brother Beresford
1926-1927 died as an infant.
There are four other younger children of Violet Brown whose birth records we are unable to
confirm online; Barrington Russell b. 1932, Sylvera B Russell b. 1935, Hydeline G Brown b.
1938, and Strickland Maurice Davis (birth year unknown). Five of her children were still alive

13
at the time of her validation. For privacy reasons we will not give personal details beyond
what is already known publicly.
As her children’s names suggest, Violet Mosse Brown had several partners during her life. She
reportedly married Augustus Brown who was the father of her youngest daughter, but we are
unable to confirm this with a marriage record and the date of marriage is unknown. Augustus
was a farmer who also took care of the village graveyard and kept its records. Some
biographical reports give his year of death as 1997 but a death certificate from 1978 showing
his daughter, H G Brown, as the informant, contradicts this.
Violet lived her life in a small village where anything unusual would be noticed. There are
strong family resemblances between Violet and her sons where pictures are available. Most
potential switch scenarios would have to defy the accepted history at several points.
Testimony could go a long way towards completing a basic validation but until a report is
available we cannot conclude a review.
A full cast-iron validation would have required extensive DNA testing for multiple relatives.
Violet Brown’s body was donated to the University of West Indies for medical research.
Emma Morano - 117 years, 137 days, Italy
Emma Morano born in 1899 was the eldest of eight siblings including her four younger sisters.
According to criteria set for an Italian study of supercentenarians, she should have a birth
certificate [9] but we would like to see this confirmed directly with published details. The
same applies to other Italian claims in the top lists.
There are opportunities for sibling identity switches that need to be ruled out. This could be
done by checking records of birth, death and marriage for each sister. Where viable options
for switches remain, a Bayesian test needs to be applied. In her favour, Morano did not travel
far from her birthplace and she had siblings who also lived to a great age.
We have no reason to think that any potential identity switch scenario is possible for Emma
Morano, but insufficient details for siblings and other family history have been published for
the validation. At this time no specific scientific report has provided details of the evidence
used to support her case. Records for Italy from this time are generally very good so the
analysis should be possible but without online access, that exercise is beyond the scope of
this review.
We understand that a validation will be published soon, and we are looking forward to
assessing the report when it appears in [36].
Chiyo Miyako - 117 years, 81 days, Misao Okawa - 117 years, 27 days, Japan
The next two women who exceeded 117 years were from Japan and were validated for
Guinness World Records around 2015. Although we have no cause to doubt their longevity,

14
we have not seen published records of birth or other vital details that would allow us to
appraise these cases. Some other trusted validations have proven unreliable, so it is our
opinion that full publication is essential if they are to be used in demographic studies or other
scientific research. We are looking forward to assessing the validation reports if they appear
in [35].
María Capovilla - 116 years, 347 days, Ecuador
Maria Capovilla is another claimant who could potentially attain a high standard of validation
if her validators would provide more details of her records. A team of gerontologists (Young,
Jeune, Robine, Cheung) visited her home country at around the time of her death in 2006 and
were able to obtain some documentation, but only a few biographical details have been
published [18]. She spent her life in Ecuador where her 1889 birth date is said to have been
recorded at her baptism. We would like more details of when and where this was, but the
birth validation can be accepted provisionally on that basis.
Capovilla belonged to the Ecuadorian high society so her life was well documented. She
married an Italian and they had five children, two of whom outlived her along with many
other descendants. If a detailed account of her life can be provided along with sufficient
photographs and official records it should be possible to independently determine whether
any identity switch possibilities are viable, or are ruled out by the evidence.
Susannah Mushatt Jones - 116 years, 311 days, United States
Susie Mushatt appears on the 1900 census in Alabama at 11 months old and she can be traced
in the 1910, 1920, 1930 and 1940 US census returns. Although this gives a good confirmation
of her approximate date of birth there is no record to confirm the exact day on 6th July 1899.
When Susannah applied for Social Security benefits at age 65 no birth record for her could be
found, so she had to secure notarized statements from three individuals who knew when she
was born [27]. There are some anomalies in censuses. In the 1910 census her year of birth is
1900, in 1920 census her name is Louisa and in the 1930 census where she lives in Mount
Vernon, New York with her husband John H. Jones (whom she married 18 June 1929) she is
28 years old and married at 27. In 1940 she is recorded as a 40 year old maid, living with the
Cokells family in Beverly Hills Judicial Township, Los Angeles, California.
She divorced her husband and had no children but she had many relatives (including her
brother Callie (1909-2004) and friends who knew her well. There is a lot of photographs in a
book which was written by her niece Lavilla Mushatt Watson in tribute to her achievements
throughout her life [27]. Susie had younger sisters, so potential options for early identity
switches need to be addressed.
The book is a useful document for the purposes of validation. It illustrates a general point that
the official record does not always tell the whole story. Record keeping was not very rigorous
in rural Alabama in early 1900s. Susannah had a sister Georgia born in 1900 who died age 9.

15
She fell between the two census records and was never recorded in any official document
[27]. As far as we know the family testimony in the book is the only record of Georgia’s
existence, apart from birth counts on the 1900 and 1910 census records which show that
there were two children born and dead between 1900 and 1910. Susie’s maternal
grandmother Cindi Bell was born in 1825 according to the 1900 census and died in 1940 at an
alleged age of 115. However, her death certificate states that she was born in 1875 which is
the year of birth of her daughter Mary according to the 1900 census.
Susie’s older brother Hardy had to change his name and date of birth and moved to another
state while Susie was at Calhoun school where she graduated in 1922. The reason for her
being 23 years old while graduating from school needs to be addressed in the validation. Susie
was described as a gifted and bright student. After graduation she left her home and visited
her family only in 1940.
Without a validation report we are unable to go further but it is possible that a high standard
of validation would be within reach given more details.
Gertrude Weaver - 116 years, 276 days, United States
Born Gertrude Gaines, she gave birth to her first son in 1916 before she turned 18. Her parent
family appears in both the 1910 and 1900 census returns. She was recorded as Girtrude age
10 in 1910 but in the 1900 returns the youngest child in the family is Tonpon aged 2 years. On
both occasions it was recorded that the mother had had 7 children and one had died. If this
is accepted as fact it implies that Tonpon was Gertrude by a different name, and her age is
confirmed. Changes of name for infants are a normal occurrence, but why should we believe
the count of children rather than their stated names and ages?
If she was born in 1900 after the census then she had her first child at age 16. This would be
a motive to exaggerate her age. There is no confirmation of her date of birth and indeed her
birth date of 4th July is often a sign that the real date had been forgotten and replaced with
the date of American Independence Day. The census returns for 1920 to 1940 were
inconsistent and put her birth year at 1900 to 1904.
Unless there is better evidence that has not been published we do not consider the validation
of her birth date to be sufficiently robust.
Tane Ikai - 116 years, 175 days, Japan
Ikai had five siblings and four children so there should be sufficient records to check her birth
date. The brief report so far available [18] does not give sufficient detail about sources for us
to evaluate the validity of her birth or claimed age.
Elizabeth Bolden - 116 years, 118 days, United States

16
Elizabeth Bolden appears on all the released census returns from 1900 to 1940. In the 1900
census in Tennessee she is Lizzie Jones, born August 1890 and age 9. She has a younger sister
Josie born in 1896. By 1910 she is married at 19 years with an infant child. From 1920 to 1940
her age is recorded as 26, 37 and 44. There is no earlier birth date record.
Her validators admit she could be a year younger than assumed [5,18]. Because of age
inconsistency in later census returns and the possibility of age exaggeration and even a sibling
swap for an early marriage we allow for a larger uncertainty of four years.
Besse Cooper - 116 years, 100 days, United States
Besse Brown is first recorded in the 1900 census in Tennessee where her birth month is given
as August 1896. She appears again in 1910 age 13, 1920 age 23 (with parents in Tennessee),
1930 age 33 (with husband Luther H. Cooper and one year old daughter Helen) and 1940 (age
43 with daughter Angie 11 years old and two sons). She has two younger sisters age 2 and 5
in the 1910 census and one younger sister age 15 in the 1920 census where she is counted as
a teacher. She graduated East Tennessee Normal School (now East Tennessee State
University) in 1916. In 1940 census both her sisters were alive and lived with her brother in
Tennessee. This makes a sibling swap unlikely.
In 2009, 92 years after her graduation, she returned to her alma mater for a celebration in
her honour. She brought with her a class ring. A matriculation card was also found. There are
pictures of her available. She was socially active and family testimonies make identity switch
scenarios unlikely. In October 2013, Cooper's grandson Paul Cooper founded the Besse
Brown Cooper Foundation, a non-profit organization dedicated to "providing financial, legal,
medical and public relations support" for supercentenarians worldwide. Unfortunately, it
seems that this foundation no longer exists.
Although we would like to see a full validation report to confirm our conclusions we believe
that her case can form a high standard of longevity verification provided more information
can be gathered to confirm her date of birth.
Maria Giuseppa Robucci 116 years, 90 days, Italy
In accordance with Italian standards of record keeping we assume that there are birth and/or
baptism records for Maria Giuseppa Robucci born in 1903. She married Nicola Nargiso in 1928
and had five children.
If a validation report could be produced giving details of her records, siblings and family a high
validation standard should be possible.

17
Carrie C. White - 116 years, 88 days, United States
Carrie C. White was verified as a 116 year old by the Guinness World Records in 1989 before
the modern era of longevity validations which led to her case being debunked. She is
mentioned here because her case teaches us a valuable lesson about replication of errors [5].
White had been institutionalised in mental health care for most of her life. Her age was
validated because records beginning from the time of her internment consistently gave her
year of birth as 1874. This included hospital records and the census returns for 1920 and 1930.
Once more historical records had been indexed, it became easier to find her parent’s names
and look back at earlier records from before she was taken into care. The census records for
1880 (where she was absent) and 1900 proved that she had in fact been born in about 1888.
It is now clear that her incorrect age must have been copied repeatedly from a poor estimate
recorded around the time she became institutionalised for life. From this we learn that
multiple records that give the same answer do not substantially increase the certainty of the
information shown. In the case of White, the apparent certainty was debunked by going back
to an earlier time, but even this would not be sufficient if the mistake is to not recognise an
identity switch, rather than a made-up date. This is relevant to cases such as the longevity
claim of Jeanne Calment where we are often told that the multiplicity of records confirms
their correctness. As Carrie C White’s example teaches us, this is not the case.
Ana María Vela Rubio - 116 years, 47 days, Spain
There is no public validation record or details of records available for Ana María Vela Rubio.
Only a few biographical details are available. It is not clear whether the standards for civil
registration or church records in the region make reliable birth date verification possible.
Kamato Hongo - 116 years, 45 days, Japan
Kamato Hongo was originally validated at 116 years of age. Her case was subsequently
debunked and removed from the validation lists [1,2,18]. Family records were observed to be
inconsistent and it is now thought that she was a case of age inflation due to a teenage
pregnancy. There are different opinions on her true age with gerontologists placing it in the
range of 109 to 112 years.
Hongo died in 2003 and she is the only case to be debunked who died after Jeanne Calment.
Once again this demonstrates that full transparency in every validated case should be an
essential requirement.
Giuseppina Projetto - 116 years, 37 days
Projetto is one of four Italian-born longevity claims that qualify for our review, marking a
significant over-representation for Italy near the top of the validation list compared to other

18
European states. Is this due to legitimate circumstances such as the moderate climate, a good
diet, a favourable genetic pool, better records, superior elderly care, or is it due to sibling
identity switches, age exaggeration to hide early pregnancies, or is it even just a statistical
fluke or the result of more validation effort? Important questions such as this cannot even
begin to be addressed without validation reports including details such as the birth, marriage
and death records for the claimant and her siblings. At this time very little has been published
by the validators and Italian records are hard to access remotely.
Jeralean Talley - 116 years, 25 days, United States
Jeralean Talley was born into the Curtz family in Georgia in 1899. The 1900 census records
her birth month as May 1899 which is consistent with her age claim. She appears with
changing members of her family on all the census returns until 1940. Although the family is
recognisable, the recorded ages and spellings of their names varied wildly from one census to
another. This might be a consequence of poor education. Even the count of children born
and living is recorded inconsistently between 1900 and 1910. Other records include her
marriage in 1936.
It has been reported that she was one of 12 siblings and had one child of her own. Her family
includes a younger sister Louverta born 1905 who lived until 2003. Only 9 of these children
are accounted for on the census so the possibility that there was another girl born after 1900
cannot be ruled out on the evidence. This means that an early sibling identity switch due to
confusion and poor record keeping is a potential possibility.
In view of these possibilities and the age inconsistencies we do not consider the validation of
this claim to be sufficiently safe for scientific purposes.
Maggie Barnes - 115 years, 319 days, United States
Maggie Hinnant was born sometime in the 1880s in North Carolina. According to the official
validation, her birth date was 6 March 1882 [5,18,25]. This date is said to be recorded in the
family bible but it is not clear when that would have been written or even if it has been
verified. Other dates both earlier and later are possible.
Hinnant married William Orangie Barnes on 29 October 1899 (unconfirmed) and gave her age
as 19. In the 1880 census her parents were not yet married, so this age was probably an
exaggeration. In 1900 the census confirms that Maggie and William were married. There were
no children but their oldest daughter, aged 10 in the 1910 census, could have been born
shortly after. It is possible that she was in early pregnancy at the time of her marriage. In later
censuses her birth year settles down at 1882.
The case of Maggie Barnes is therefore very similar to that of Elizabeth Bolden and Gertrude
Weaver where age exaggeration may have been used to make a wedding at a young age look
more respectable. Once potential exaggeration is recognised the question is how much, and

19
why should the declared information be trusted at all when there is no birth registration or
baptism record?
In these three cases the claimants do seem to have at least reached the age of
supercentenarians. Given how rare that is, we can conclude that age exaggeration at
marriages could have been very common and should be considered as a realistic possibility
whenever it cannot be ruled out.
Dina Manfredini - 115 years, 257 days, Italy and United States
Dina Manfredini, born Dina Gueri in 1897 emigrated from her home in Italy to the United
States in 1920 with her husband Riccardo Manfredini. She was pregnant with her first
daughter when they made the journey. From that date forward, adequate documentation of
her life can be found. Starting with her emigration record from 1920, she then appears in the
1930 and 1940 census with consistent record of her age. We found a picture of her in the Des
Moines Register, Iowa dated 7 Jun 1944 with a report of her son being with the armed forces
in England. Other events of her family’s life were covered in local news stories.
Italian records should include her birth which we assume her validators have obtained. It is
therefore only the lack of a validation report that presents any obstacle to checking whether
her longevity can be validated to a higher standard. A list of her siblings should be provided.
She had made a return visit to Italy with her children where she reunited with her family, so
details of this event would help rule out an identity switch.
Summaries for top listed men, validated or formerly validated.
Shigechiyo Izumi - 120 years, 237 days, Japan
Shigechiyo Izumi who died in 1986 was originally validated by the Guinness Book of Records
and included in the GRG published lists [28]. Doubts about his longevity were raised well
before his death [29]. Some of these doubts were similar to rumors about Jeanne Calment
[6,8]. Although scientific papers disputing his age were published (in Japanese) soon after his
death [39, 41], he was not removed from the record books until around 2011. The Guinness
Book of Records had accepted a letter from the Japanese Prime Minister’s Office confirming
his age based on the family register. It is now believed that his birth was confused with that
of an older brother who died as an infant. It was common practice in the region at the time
to reuse the name and registration of a child who had died. Though there seems to be no
proof that Izumi was not a supercentenarian and his physician said that his age was consistent
in multiple censuses [3], Izumi’s claim had “been rejected by almost all experts who are
familiar with it… and the common belief is that he was in fact "only" 105 years old at the time
of his death” [4]. The original papers describing Izumi’s case are however not readily
available.

20
This case is mentioned here because it shows the importance of exploring the full family
history, especially the list of siblings, in order to complete a validation. Official confirmations
of claimed age, such as the letter from Japanese Prime Minister’s Office in the case of Izumi
[3] or the Court of Tarascon in the case of Calment should not be considered as scientific
proofs while the burden of proof should be on validators and not on sceptics. This case also
illustrates that there is sometimes unnecessary delay in removing dubious claims from the
official lists.
Jiroemon Kimura - 116 years, 54 days, Japan
Jiroemon Kimura is currently the world’s oldest ever man in the validated lists. He was
validated scientifically by a Japanese team [29] and is one of the few other cases which we
consider to have been validated to a standard comparable to the “gold standard” applied to
Jeanne Calment. Since Calment’s case is nevertheless disputed due to an unresolved switch
scenario, it is now also necessary to ask if the case of Kimura is sufficient to eliminate any
ambiguity in his record.
One difficulty with Japanese claims as explained by Kimura’s validators is that vital records
are not public. To access records the validators must approach the family for permission. In
this case the validators were able to obtain birth, marriage and death data for all Kimura’s
siblings and children with only a few missing gaps. These are helpfully listed in the validation
report. This should be a requirement for any scientific validation.
Using these lists and the background information provided, we have the ability to identify any
potential opportunities for an in-family identity switch. Kimura had a brother five years
younger whom he visited in Korea for a period of five months to care for him while he was
sick. The date of this brother’s death is missing but he is assumed to have died at 85. This
provides the first potential opportunity for a swap with the brother. Kimura’s oldest son died
at the age of 76 when Kimura was 103. This offers a second potential opportunity for a switch.
Kimura’s validators draw attention to the possibility of switches in the introduction to their
report. They also identify the specific period around when he visited his brother as one where
a swap might be possible. They dismiss this based on testimonial evidence of Kimura and his
family. However, not many details are given on these crucial points. In general, photographs
can also help rule out switch scenarios, but in some cases more information is needed.
Unfortunately, even the DNA evidence can’t rule out the potential early life sibling swap. Even
with one of the most detailed validation reports available we do not feel that we have
sufficient information yet to know if a cast-iron standard is attainable.
Christian Mortensen - 115 years, 252 days, Denmark and United States
Christian Mortensen’s birth in Denmark in 1882 is well documented in both official and church
records. His validation reports (Wilmoth, Skytthe, Friou, Jeune) can justifiably claim to match
the gold standard of the validations for Calment and Kimura [4,18,30]. However, he led a life

21
of mostly solitude and travel. His record is an outlier in his country where no other male
supercentenarians have been validated. His case therefore requires extensive scrutiny. With
help from Per Hagemann we were able to carry out further searches in Danish archives for
information about his life.
He immigrated to the United States, arriving on 8th Sept 1903 after completing a tailor’s
apprenticeship. The ship’s manifest on his arrival shows that he had arranged to stay with
Niels Hansen of Kenilworth Avenue, Oak Park, Chicago. His brother Anders Peter Julius
Mortensen may also have been in Chicago working as a cobbler at around this time [37], but
he must have returned to Denmark where he married in 1909. Jyllands-Posten articles from
Oct. 3, 1992 and Aug. 10, 1995 say that Christian Mortensen went away unhappy with the
tradition that a tailor apprentice should pay for his training, even though he made good
money home to the tailor master. It annoyed him that he couldn't become a soldier (he was
too small) and that made him travel to the United States, where he initially took a job as a
milkman.
International travel provides both opportunity and motive for an identity switch. In the US,
Mortensen led an itinerant lifestyle for some extended periods of his life. We found the record
of his marriage 16 Nov, 1916 with Mayne Carne, a childless widow of James W Carne whose
birth name in 1886 was Slavins, while her step father’s name was Spaningberg. They were
recorded living together in 1920 when Mortensen was working as a dairy teamster. It was a
rather short and unhappy marriage with no children. After their divorce Mayne remarried to
Carl J. Koepke.
Christian’s nephew Richard Mortensen said (Jyllands-Posten from Aug 15, 1997) that the
family lost contact with him for 25 years until he sent a letter and communication was
resumed. Since then he has visited Denmark four or five times, the last time in 1968. His niece
Grethe Mortensen said (Horsens Folkeblad from Aug 19, 1995) that he has been to Denmark
four times, the first time in 1949. This agrees with his published photos from 1949, 1955 and
1962 and the statement that the visit in 1962 was his third visit to Denmark (Aarhuus
Stiftstidende from 13 April 1962). However, in an interview with the validators, Mortensen
said that he had returned home to see his mother. She died in 1924. After world war two,
Mortensen returned home to Denmark on multiple occasions until at least 1969 and must
have seen family back home. We know he saw his brother Julius and nephews, but it is not
clear how consistently his identity could be checked over time. In 1978 he moved into a
retirement home to live out his last two decades in care.
We have searched the official records and newspapers for the US extensively and found a few
other items that the validators seem to have missed. We can plug a gap in the record between
about 1905 and 1918 with a 1910 census return from Los Angeles Assembly District 75,
California where he was recorded as T. P. T. Cristian Mortensen working as a tailor in a shop
and accommodated as a lodger. The three initials match his first given names at baptism of
Thomas Peter Thorvald Kristian Ferdinand. His recorded age and year of entry are exactly

22
correct and the circumstances are consistent with his testimony that he was in the Western
U.S. at this time. He is also in the Los Angeles City Directory for 1914 as a tailor under the
name Thorwald C Mortensen working for Charles N Van Pelt from Kansas. He later said that
he ceased to work as a tailor in 1915 when mechanisation reduced demand for custom
tailoring.
We discovered three reports in the Chicago Tribune from 1946 to 1947 mentioning his
activities as a Sea Scout. The first reported on 21st July 1946 that he qualified as an adult
leader. His name is given as Chris Mortensen and his address in Normal Avenue matches the
1940 census and WWII draft card. This is also consistent with his testimony that he enjoyed
sailing and built his own boat after retiring to South Texas shortly afterwards in 1950.
To trace his whereabouts between 1950 and 1978 the validators were apparently able to look
at his pension records (presumably from the now defunct Continental Canning Company.) It
would have been helpful if these locations could have been listed. Perhaps it is not too late
for details such as these to be published.
His validators dismissed the switch possibilities because he was able to recall stories from his
youth and also because he was often seen by his Danish family. However, both his parents
had died by 1924, and his brothers by 1967 so the period of testimony from younger relatives
must be shown to overlap. His memory of events was not always good. For example, he could
not remember the name of his wife (Jyllands-Posten, Aug 16, 1997) and the name of the
village where he was born (Jyllands-Posten, July 15, 1997). It is true that he gave a remarkably
detailed account of his life to validators and reporters during interviews at his retirement
home. These match the documentary record well.
Obituary in Horsens Folkeblad from May 9, 1998 states that he exchanged letters with the
family - one niece in Horsens, one niece in Klovborg and two nephews in Aarhus. This evidence
is useful to rule out late life switch but they did not know him before his departure to United
States. Aarhus Stiftstidende of 13 April 1962 reported on his return home for a year-long stay
after his boat was destroyed in hurricane Carla of 1961. He is said to have spent his time
visiting his friends and relatives including his nephew Viggo Mortensen. It was surprising that
he said nothing of his time in Chicago and said that most of the time he lived in San Francisco but
otherwise his interviewed account is consistent with the records we have found. The article
included a picture which despite being unclear did bear a resemblance to the man seen at the
end of his life. Another paper in Jyllands-Posten had a report from the same day with a
different picture. Similar report in Aarhus Stiftstidende for 16 June 1955 tells about an earlier
visit home. This also had a picture that appeared to be the same person in as far as it was
possible to tell.

23
Figure 2 - Christian Mortensen age 67 (left) and his brother Julius age 71 (right) in
Skanderborg, 1949.
A photo from 1949 (Figure 2) where Mortensen is pictured with his brother Julius shortly
before his death was published in Jyllands-Posten of 16 Aug 1995. In this photo Chris looks
somewhat younger. His assumed age of 67 does not seem to match his appearance. The other
brother Carl Emil lived elsewhere than Chris in their youth, and we do not know whether they
met in 1949. The photo was taken at Borgergade 11. Chris was counted in this house in 1901
together with tailor Anders Andersen, his wife and children and another young tailor Christen
Peter Sørensen. The street was named Lillegade back then. One of the other two men on the
photo who are mentioned as old acquaintances from Skanderborg could be Andersen who
was living there from 1901 until 1949.
If an identity switch took place between 1969 and 1978 his replacement must have spent
significant time with him, heard his life story in detail and managed to cheat the family in
Denmark while earlier switch should have been missed or approved by his brother Julius and
several other people, but for the most extreme cases of longevity such as this we must allow
for the possibility of improbable scenarios. Although he had not spoken his native tongue for
much of his life he was still able to speak some rusty Danish when interviewed for the Jyllands-
Posten (Aug 10 1995). Danish reporters confirmed that he spoke with an indelible “East
Jutland accent”. This reduces the probability for a wide range of potential switch scenarios
but it does not rule out the possibility of a swap with a family member.

24
Table 1 – Evidence timeline for Christian Mortensen
Date
Event or occupation
place
Source
16-Aug-1882
Birth
Skaarup
Birth register
26-Dec-1882
Baptism
Fruering
Church register
1890
7 years old
Skaarup
Census
27-Sep-1896
Confirmation
Fruering
Church register
1 Jan 1900
Certificate issued
Skaarup
Conscript register
17-Apr-1900
death of brother Hans Johan
Skaarup
Death record
1901
Apprentice tailor (res. from 1898)
Skanderborg
Census
3 Mar 1903
Issued, tailor
Immigration record
26 Aug 1903
Emigration
Copenhagen
8 Sep 1903
Immigration
New York
Passenger list
27 May 1904
death of father
26 Mar 1905
Danish brotherhood
Chicago
Document
1908
Travel west
Testimony
1910
Tailor
Los Angeles
Census
1914
Tailor
Los Angeles
Directory
1915
Ceased to be tailor
Danish newspaper
16 Nov 1916
Marriage to Mayne Carne (Slavin)
Indiana
Marriage registration
12 Sep 1918
Milkman
Chicago
WWI Draft card
1920
Teamster in Dairy
6034 Union Ave, Chicago
Census
15 Jun 1922
Became US citizen
Chicago
Naturalisation record
1923
Joined Masonic Lodge
Chicago
Testimony
Before 1924
Visited Denmark
Testimony
10 Jun 1924
Death of mother
Death register
23 Mar 1925
Master Mason
Chicago
Masonic transfer
1926
Ex-wife remarried
Marriage registration
About 1928
Wrote to family
Testimony of nephew Richard Mortensen
7 June 1929
Hired by Can Company
Clearing Illinois
Pension record
Oct 1929
Wall Street crash
Historical
23 Dec 1936
Issued social security
SSA card
1940
Laborer, tin can factory
Normal Blvd, Chicago
Census
1942
Continental Canning Co.
6236 Normal Blvd
WWII draft card
21 July 1946
Qualified as Sea Scout leader
Normal Ave
Chicago Tribune
25 Mar 1949
Departs for Denmark
Passenger list
1949
Pictured with Julius and Andersen
Skanderborg
Danish newspaper
31 July 1950
Retired from canning
Clearing Illinois
Pension record
1950
Moved to South Texas
South Texas
Testimony
Mar 1951
Death of brother Anders Julius
Testimony
16 June 1955
Visiting Denmark
Denmark
Danish newspaper
7 Sep 1955
Arrived back in US
New York
Passenger list
11 Sep 1961
Boat destroyed by hurricane
Texas
Testimony
13 April 1962
Visiting Denmark
Denmark
Danish newspaper
Before 1962
Lived mostly in San Francisco
Danish newspaper
14 Nov 1962
Returned to US
New York
Passenger list
9 Nov 1967
Death of brother Carl Emil
Testimony
Aug 1968
Issued Operator License (car?)
Florida
After 1968
Exchanged letters with the family
Testimony
1969
A month in Denmark
Passport
14 Jul 1970
Freemason transfer
Texas
Dimit
1978
Entered retirement home
California
Testimony
25 Apr 1998
Death
California
Death record

25
Mortensen is said to have lived and worked in 28 states (Jyllands-Posten from Aug 16, 1997)
so our timeline (Table 1) remains very incomplete. There are indications from his interviews
that he worked as tram repairman in San Francisco during the Great Depression. One positive
reason to suspect a switch at some point is that his height barely changed between his
twenties and his final years [4,30]. It is typical for people to lose about 8cm in stature over
the course of a long adult life. This is not conclusive, but it does make additional checks
necessary.
He was the youngest brother in his family, so sibling switches can be ruled out. He or his
brother could have had a son whose presence is not visible and he had cousins [4], so the
possibility of an in-family swap remains. A DNA comparison with living relatives would have
been completely deterministic by ruling out plausible switch options. Since he had no
identified children of his own this test is no longer possible unless samples have already been
taken and kept, or if a forensic study to find traces of his DNA is performed. It would be useful
to have a more extended record of his family history to identify named family contacts.
Publication of transcripts of interviews with validators and other evidence would be helpful
for this important longevity case.
Christian Mortensen said that he was a scout when he was a boy and was restless since then,
travelling to the places with a climate which suits him best. He attributed his longevity to
clean lifestyle, drinking 18-20 glasses of boiled water a day and following vegetarian diet
(Jyllands-Posten, August 1995). His lifestyle was probably influenced by famous doctor
Hindhede, who was a physician in Skanderborg, Denmark, where CM lived at that time [30].
He also smoked cigars as mentioned first in Aarhuus Stiftstidende from 13 April 1962 and
most of the later sources about him. His poor vision was mentioned in his WWI draft card and
he seems to wear glasses in the 1955 photo as well as in later photos from the nursery home.
In summary we find his validation reports to be good compared to many others. The wealth
of additional evidence including photographs, testimony and official records brings his
validation very close to the standard that we hope can be achieved for future lists.
Mortensen’s case underlines the rule that DNA of supercentenarians should always be
preserved and sequenced for the purposes of validation. If his DNA from after 1978 matched
correctly to relatives in Denmark he would achieve a cast-iron validation.
Emiliano Mercado del Toro - 115 years, 156 days, Puerto Rico
The birth date of Emiliano Mercado del Toro in 1891 is reliably recorded on an original birth
certificate and a baptismal record from 1892 as reported by the validator. He lived his entire
life in the same area except during US military service. In his brief validation report it is said
that many of the documents used are confidential [5,18]. This is a significant hindrance to
independent checking. However, some of the most important records are public. In the 1930
census his age is underreported by 6 years.

26
Identity switch needs to be ruled out. DNA tests could have been used to eliminate potential
swaps other than an unlikely early sibling switch.
Mathew Beard - 114 years, 222 days, United States
Mathew Beard was validated in 2003 by the same SSA Kestenbaum study that led wrongly to
the validation of Lucy Hannah [25]. His case is also marked as “disputable” on the Gerontology
Wiki He had been a preacher and farmer in Wildwood, Florida with his wife Angerina Brooks
and their large family. He died in 1985.
According to his own testimony recorded in newspaper articles from around 1975 onwards,
he was born in Norfolk Virginia on 9th July 1870 and moved with his family to Missouri at age
three where he worked in a saw mill at age 12. He first arrived in Wildwood Florida in 1887
working on laying the tracks for the Seaboard extension of the Florida Central and Peninsular
Railroad, but he did not settle there until later. His next twenty or thirty years are described
as a series of adventures; fighting in Cuba during the Spanish-American War of 1898, boxing
in St Louis Missouri, and playing black league baseball in Richmond Virginia. He does not give
many details of his family but he says they had a farm in Nelson Missouri. He also reported
that he had a grandfather who lived to nearly 146, but did not supply a name. He returned to
Wildwood where he became a farmer. He was known as “Preacher” and claims to have been
ordained in 1919, about the same time as he met his wife. He even spoke of visiting Scotland.
Unfortunately there are no reliable early records that confirm his story. We found the family
in the US census for 1930 and 1940 and the Florida State census for 1935 and 1945 (for 1935
his name is wrongly transcribed as Mathew Beat). His birth year according to these entries
could be 1870, 1872, 1879 or 1898. The 1930 census records that he was born in North
Carolina and the 1945 census says Georgia. The others say he was born in Florida. None say
Virginia as he reported later. His many children recorded in the census were born between
1911 and 1940, so the youngest were born before he said he met his wife and therefore may
not be his own.
No confirmed record of him or his children earlier than 1930 can be found. However, his wife
appears as a six year old child in the 1900 census with the name Angere Brooks, confirming
that her birth year of 1894 - which appears on her gravestone - is probably correct even
though it too is inconsistently reported later. She also appears with her parents in the 1910
census as Tangerine Brooks age 16. This would be about a year before her eldest child is
thought to have been born. Her parents can be found in the 1920 census but she is then
absent.
It is thought that the SSA validation used a census entry from 1880 in Tennessee where there
is a record of a Mathew Baird born about 1868, a son of Anderson and Martha. There is no
reason to believe that this was him. This same child also appears in the 1870 census at age

27
four as Mathis Baird. If this is correct then his age at death would be an unlikely 117 to 119
years, but there are no details that confirm it.
Another Mathew Beard born October 1886 can be found in the 1900 census for Georgia. This
would make him only 98 when he died, but again there are no matched details apart from his
name. The name Beard is sufficiently common in South Eastern states that a few near hits are
to be expected. The 1880 census has a Matthew Beard born 1870 in Illinois and the 1900
records someone with the same name born 1890 in Texas. In Missouri, Florida or Virginia
where we expect to find him, there is nothing found before 1930 even allowing broad
matches to his name.
The lack of records to confirm his story before 1930 is clearly problematical for his validation.
With the details he gave we would expect to find him in some census, war veterans list, WWI
draft or in newspaper reports or church records. It is also a concern that his original family
cannot be identified. His testimony of his early life could be true, but it reads more like an
imaginative fantasy. There is no credible record of his birth and there is nothing to refute the
suggestion that his name and identity were made up to hide his true early past life. In the
light of this we dispute his validation.
Given these findings it is difficult to understand why he is still listed as reliably validated in the
GRG tables with no indication that he might be removed. His age is also present in the IDL
database.
Walter Breuning - 114 years, 205 days, United States
Walter Breuning does not have the benefit of a published scientific validation. Unfortunately,
there is no birth or baptism record from his early years that confirm his full birth date. For
that we need to look at his World War I draft card at age 21 which gives the date as 21
September 1896 in Minnesota. The earliest record that lends support for his birth validation
is the 1900 US census when he was three years old. In 1900 the census also recorded birth
month which is given correctly as September 1896. His marriage certificate from 14 Oct, 1922
records his age as 28, which is about three years older than expected, but this is the only
inconsistency we found. His birth year is almost perfectly consistent with the age claim
throughout census returns for 1900, 1910, 1930 and 1940. We consider the birth date records
sufficient for our standards in this case.
Breuning’s parents died in 1915 and 1917, but he had four siblings who lived to between
about 80 and 100 years. Their testimony could have strengthened his case but we don’t know
if they were interviewed. His brother Frank was about eight years younger and therefore
provides a potential opportunity for a switch. A sibling swap during childhood or early
adulthood where both remain alive would be unusual if not unheard of. It would also mean
Walter claiming to be 28 when his actual age was 17 at his marriage.

28
Walter reported that he started working for the Great Northern Railroad in his teenage years.
His draft card confirms his employer at age 21 and a city directory lists him as a clerk there in
1923. Company employment records may be able to prove that he worked there earlier. This
makes an early switch with his brother almost impossible but we should take into account
their difficult financial situation and the minimal age of workers which could make his brother
unable to work officially. Breuning said that he would have to hide from owner James J. Hill,
as Hill did not want any railroad employees under the age of 18.
In later life, newspaper reports follow his activities as a Masonic leader often with pictures.
The Great Falls Tribune of 9 February 1984 has a photograph of him which shows that his
appearance changed very little between then and his final years. Late life pictures of his
brother from his family seen on ancestry.com show that there was a strong family
resemblance between the two. There is also a picture of Walter in a school group at age 9 and
a picture of his younger brother at 16 (unverified). A more detailed publication of pictures,
records and family testimony is required for greater certainty.
His first wife Agnes C. Toohey died in 1957. He remarried although he is said to have lied
about that in testimonies. The marriage record from 3 Oct, 1958 was only discovered after his
death. His second wife Margaret Vanest died in 1975.
Apart from the missing birth record which hinders the verification of his exact birth date, we
think that there must be sufficient documentary evidence to provide a strong validation, but
we cannot confirm that this has been done without a full report. DNA could have been used
to rule out most identity switch possibilities by comparing his DNA with nieces and nephews,
but the potential early swap with his brother cannot be ruled out this way.
Yukichi Chuganji - 114 years, 189 days, Japan
This is another case from Japan which regretfully we are unable to check because full details
of birth, siblings and children are not provided by his validators and are not accessible to
public search.
Joan Riudavets - 114 years, 81 days, Spain
There is a brief validation report for Joan Riudavets born in Menorca in 1889, which provides
at least some of the details we require (Poulain) [1]. We are not told what suitable birth
records were available, but there is a list of siblings and children.
Among these, the only potential identity switch opportunity is between Joan and his brother
Jaime who was five years younger. Jaime’s death was recorded in 1915 when Joan was 25.
We cannot rule out the sibling switch at that time. We are told that Joan was head of a local
youth group in 1912. If we knew this continued until his marriage in 1917 we could be more
sure that a switch was not possible.

29
The validator interviewed Joan himself and his younger daughter. She was not alive at the
critical moment but could have useful information. The validation does not explicitly mention
the possible switch so we are not given sufficient information to determine the case with
confidence, but we accept that a basic level of validation was recorded.
Fred Hale - 113 years, 354 days, United States
There is no validation report for Fred Hale but good records can be found online. His age was
recorded with perfect consistency in all five available census returns from 1900 to 1940, and
at his marriage. The 1900 census confirms that he was born in December 1890. There is no
birth or baptism record, but his draft card from 1918 gives his exact birth date correctly.
Although he married at 19 after the birth of his first daughter, his age was recorded earlier
and consistently so age exaggeration is not a factor.
He had only one brother and he was older. Two of his sons outlived him and the third died of
influenza as an infant. There are therefore no opportunities for switches within the family
circle. Family testimony confirms his identity [40].
His parents lived to 91 and he had many living descendants. He was married for 69 years and
only travelled extensively after his wife’s death. It is unlikely that a late-life switch would be
possible given the size of his family, but a DNA test would have eliminated any possibilities of
out-of-family switches.
Although a written validation report and some photos would have helped to confirm that
nothing has been missed, we can conclude from our own searches that the fortunate
circumstances make this validation reasonably robust.
Israel Kristal - 113 years, 330 days
No validation report is available at this time for Israel Kristal and his is a case where a clear
explanation for the validation would be required. He was a holocaust survivor and emigrated
from Poland to Israel [40]. Such upheavals inevitably make validation more difficult. We do
not know if he had any siblings.
The most significant problem for us is that his birth date was validated using a document from
1918 when he was 15 years old. The full provenance of the document has not been published.
Kristal’s name has been added to the document in a different handwriting and with an
undeciphered note. This needs to be explained. It was considered sufficient for the GWR
certification but it will not meet the requirements for a cast-iron validation because it is too
late in his early life, and too disconnected from his actual birth.
We are informed that a validation report for this case will appear soon so we reserve
judgment until seeing what other evidence may be available.

30
Summary of the Reappraisals
When we began this review exercise, our expectation was that the main issue would be the
potential possibilities for identity switches similar to the one uncovered for Jeanne Calment.
Our main goal was to determine whether DNA tests could be useful to strengthen or even
refute past or future validations. What we found was that more basic issues need to be
addressed first. Many of the validations are not sufficiently well documented for our
purposes, and some not at all. We can accept that this is not possible to do for every individual
supercentenarian, but proper reporting should at least be undertaken for anyone over 115,
anyone who held the World’s Oldest Person title, and those people who are included into
scientific studies of ageing and longevity. In some cases the validation can be reconstructed
or even improved using online genealogical and news archives. To summarize our findings
and for future useful information we created the Archicentenarian Wiki. [45]
Where documentation is available, or where we have been able to find records ourselves, we
have found that the standard used for validation of the birth record is sometimes lower than
we would like, especially for births in the U.S. The standard currently in place accepts a record
from within 20 years of birth but sometimes we felt this was not sufficient to give confidence
in the correctness of the birth date. In the case of Sarah Knauss we were able to significantly
improve the quality of the validation by discovering new early records, while in other
examples our findings cast doubt.
Demographic data is produced from validations by the International Database for Longevity.
We are aware that measures are taken to remove bias but some older cases that we do not
trust are likely to be represented in the statistics. Since IDL is anonymous it is impossible to
check the validation of most data-points starting from the age of 105. Users of this data are
required to trust that it is correct. The book “Supercentenarians” details the procedures by
country used to construct the database and was a useful source for some individual validation
reports [1,5,18]. We understand that a delayed follow-up book entitled “Exceptional
Lifespans” is now due in the Spring of 2020 and this will add new validation reports for some
cases mentioned here.
While validation standards have improved with time, some old cases that are in doubt are
kept and the extent to which the validation process is documented appears to have become
less. In the women’s lists, almost all verified claims for supercentenarians older than Charlotte
Hughes (115 years, 228 days) who died before Jeanne Calment in 1997, have been or should
be withdrawn. The only exception is Tane Ikai for whom we have insufficient information to
determine the reliability of the validation. In the men’s list everyone older than Frederick
Frazier (113 years, 138 days) who is not disputed died after Calment. Lucy Hannah, Carrie C.
White, Shigechiyo Izumi, and Mathew Beard fall into this category.
For more recent validations in the list where sufficient information is available, we still find
cases where we have significant doubts, even without invoking the possibility of identity

31
switches. These include Gertrude Weaver, Elizabeth Bolden, Jeralean Talley and Maggie
Barnes. A number of other cases can be confirmed to pass the basic level of validation
currently in use, but potential identity switch scenarios cannot be ruled out due to a lack of
DNA testing. For the rest, the problem is simply that there are insufficient validation reports
or online records to be able to determine how the validation was done.
Name
Claimed age
Year
Refs
Issues
Age
Oldest ever validated women
Jeanne Calment
122 years, 164 days
1997
[18,19,20]
MCT
99,122
D
Sarah Knauss
119 years, 97 days
1999
[5,23,18]
BY
116-119
R
Nabi Tajima
117 years, 260 days
2018
VIBY
113-117
N
Lucy Hannah
117 years, 248 days
1993
VIBAML
107, 117
D
Marie-Louise Meilleur
117 years, 230 days
1998
[18,26]
MT
117
R
Violet Brown
117 years, 189 days
2017
VITS
114, 117
R
Emma Morano
117 years, 137 days
2017
[9]
VIES
?
N
Chiyo Miyako
117 years, 81 days
2018
VI
?
N
Misao Okawa
117 years, 27 days
2015
VI
?
N
María Capovilla
116 years, 347 days
2006
[18]
I
116
R
Susannah Mushatt Jones
116 years, 311 days
2016
[27]
VET
116
R
Gertrude Weaver
116 years, 276 days
2015
VBYA
111-116
U
Tane Ikai
116 years, 175 days
1995
[18]
VI
?
N
Elizabeth Bolden
116 years, 118 days
2006
[5,18]
BYASE
112-116
U
Besse Cooper
116 years, 100 days
2012
VB
116
R
Maria Giuseppa Robucci
116 years, 90 days
2019
VI
?
N
Carrie C. White
116 years, 88 days
1991
[5]
B
102
W
Ana María Vela Rubio
116 years, 47 days
2017
VIB
?
N
Kamato Hongo
116 years, 45 days
2003
[1,2,18]
BYA
109-112
W
Giuseppina Projetto
116 years, 37 days
2018
VI
?
N
Jeralean Talley
116 years, 25 days
2015
VAES
113-116
U
Maggie Barnes
115 years, 319 days
1998
[5,18]
BYA
113-117
U
Dina Manfredini
115 years, 257 days
2012
VITE
115
R
Oldest ever validated men
Shigechiyo Izumi
120 years, 237 days
1986
[29,38,39,41]
VIES
105
W
Jiroemon Kimura
116 years, 54 days
2013
[29]
MLSC
111,116
R
Christian Mortensen
115 years, 252 days
1998
[4,18,30]
TM
115
R
Emiliano Mercado del Toro
115 years, 156 days
2007
[5,18]
I
115
R
Mathew Beard
114 years, 222 days
1985
VBYAM
90-119
D
Walter Breuning
114 years, 205 days
2011
VESY
106, 114
R
Yukichi Chuganji
114 years, 189 days
2003
VI
?
N
Joan Riudavets
114 years, 81 days
2004
[1]
IES
109,114
R
Fred Hale
113 years, 354 days
2004
VT
113
R
Israel Kristal
113 years, 330 days
2017
VIBT
?
N
Table 2 – Summary of longevity validation reviews

32
Table 2 summarises the results of our reappraisal with references. The likely ages and age
ranges that we have determined are also given. Ages in italics are for less likely options
including potential identity switches. Where we have more confidence the age is in bold.
When references are not given it means we could not find any published validation report
with specific details of the case. Generic switch opportunities are not included and exist to
some extent in almost every case. A question mark indicates that we have not been able to
find sufficient documentation or records online to make a determination. Some of the main
issues that we identified are labelled with letter codes as follows:
V - no validation report as of December 2019
I - insufficient information
B - inadequate record for birth
Y - possibility of age exaggeration for underage school, marriage, pregnancy,
occupation or military reason
C - potential parent-child identity swap
S - potential sibling identity swap
E - potential early-life identity swap
M - potential mid-life identity swap
L - potential late-life identity swap
A - inconsistent age reporting
T - migration or travel
In the final column we provide our conclusion on the current status for each claim as follows
W – Claim has been withdrawn. Do not use these cases.
D – These cases are sufficiently problematical for us to dispute the validation and we
therefore consider them invalidated. It is our opinion that they should not be assumed
to be correct unless further strong evidence in their favour is found.
U – These passed the weak validation standard required by GWR, but for scientific
use the validation is unsafe due to possible age exaggeration or identity switches.
R – There are some doubts about these cases which mean they should be treated
with caution. The high claimed age mean that they require extra scrutiny. Scientific
users are advised to review the evidence themselves if they consider referencing
them.
N – No conclusion. Information publically available is insufficient to determine these
claims.
Those who find our analysis too strict are advised to look at validations from 19th century [44]
and see how confident were scientists of that time in correctness of validations of ages well
above 120 years. One of the arguments used was the low rate of mistakes after double
checking. This remind arguments of some validators today.

33
It is not our purpose to denigrate the seminal work of the many longevity validation pioneers
who have initiated the scientific study of demographics above 110 years of age. They have set
up networks to identify potential supercentenarians around the world and have made useful
preliminary lists by eliminating the majority of errors and false claims, especially over the last
twenty years. It is only recently that more reliable records from the time and place of birth
for many supercentenarians have become more available and easier to search. It is also only
now that the cost of whole genome sequencing has reached the point where extensive
application to supercentenarians and their relatives may be economically viable with the low
levels of funding available. Our aim is only to encourage further evolution of standards in line
with technology developments. For a discussion of validation standards and their history with
references see e.g. ref [1,2]
However, we do think that the required standard for reliable validation at the highest ages on
these lists has been underestimated. Both the need and opportunity to improve the standard
going forward now exists and we appeal to the longevity validation community to put into
place a plan of action to accomplish it. In particular, the fundamental scientific question of
whether or not there is a maximum natural limit to longevity needs to be answered and this
can be achieved over the next two decades, but only if validation standards are raised.
We have taken on the task of defining some principles for an improved “cast-iron” standard
which we outline below. We do not underestimate the enormity of the task ahead for
validators if they are to achieve this standard. We understand that realistically, most friends
and family of any longevity claimant will not be interested in achieving a higher scientific
standard of validation. For them what matters most is to be awarded the coveted world
record holder certificate. Once this has been done they have little obligation to assist further
with the provision of information that could potentially invalidate their claim. We do not think
that the cast-iron standard should necessarily be required for the recognition of any record,
but it is important to gather evidential information before any title is awarded where possible,
so that the best scientific standard can also be achieved.
We also appreciate that questions of bio-ethics and privacy standards must be addressed.
These are beyond the scope of this review.
As a first step towards improving standards, all details of evidence used to validate the birth
date of claimants on the published lists of supercentenarians should be made public. This
relatively small and reasonable advance would make it possible for scientists to reconstruct
the supercentenarian lists on a more reliable basis. In an effort to improve consistency of
reporting we offer a questionnaire for validators below.
Towards a Cast-Iron Validation Standard
The following conditions outline the cast-iron standard for longevity validation that we hope
can be achieved in future.

34
Scientific validation should not be concluded during a claimant’s lifetime unless a cast-iron
validation is already possible. To do so does not allow for a Bayesian test including final age.
All relevant details of evidence and reasoning used in the validation should be made public
within a year of the validation being declared. Evidence that is confidential should not be
taken into account. This is to allow scientific users of the validation to make their own
judgements on the robustness of the validation. It is also necessary in case a later re-
evaluation in the light of new information is required. Families are entitled to their privacy,
but if they are not willing to allow relevant information to be made public then the validation
of their claim cannot be replicated and has no scientific value.
Date of birth should be established with a birth certificate or other record (such as infant
baptism) from near the time of birth. In rare cases, later evidence of date of birth may be
accepted provided there is also a record such as a census within the first ten years that gives
the correct year for the birth, and that there is sufficient consistent dating evidence
throughout life to leave no reasonable doubt. It is important that consistency goes all the way
back to near birth to avoid the possibility that there has been replication of error. Inconsistent
age reporting does not necessarily mean that a claim cannot be validated, but it makes the
requirement for an unambiguous record from the time of birth essential. Once a date of birth
has been established in this way it does not prove the date of birth for the person who died
at a great age. The potential for Identity switches must also be examined and ruled out. All
relevant details of evidence for date of birth should be made public at the time that the
validation is declared.
A genealogy for the claimant should be established with the help of family and public records
and early newspaper reports. This should include where possible the recent ancestry, all
siblings, and parent’s siblings as well as children, sibling’s children and any deceased
descendants or close relatives. This genealogy including all sources should be made public as
part of the validation.
As many early photographs as possible of the claimant and any potential switch candidates
should be gathered and published with the best possible electronic scans. Where copyright is
a barrier to publication, e.g. for pictures from newspaper reports, the details of the source
should be given as reference.
As much testimony as possible should be gathered and published along with sources for any
relevant newspaper reports from earlier life. In particular all friends and relatives should be
asked the dates when they were in regular contact with the claimant, either in person, or by
remote means. This can be used to rule out potential identity swap scenarios.
If the claimant or their family owned any documents such as birth certificate, family bible,
passport and photographs please discuss with them how these will be conserved and make
copies at the highest possible resolution. Archival documents should be sought from as many

35
relevant sources as possible including civil and religious records, census returns, city
directories and newspaper archives. Validators should also seek nuclear and mitochondrial
DNA samples from the claimant and as many relatives as possible, and should carry out whole
genome sequencing in order to confirm that the claimant has the expected autozygosity ratios
and consanguinity levels with relatives. The DNA sequences should be held in trust with
published conditions for further use on request according to the wishes of the claimant and
relatives.
Where the quantity of information is too much to publish in full it should be archived in a
form where researchers can request access.
Once birth and death date are established, all potential identity switch opportunities during
the claimant’s lifetime should be identified and ruled out using the evidence gathered.
Where there is any doubt, a Bayesian inference test should be performed. See reference [8]
for the method.
Other details of the claimant’s lifestyle and health that would be of interest to science should
also be made public or held in trust according to permissions obtained.
Where these tests are passed beyond a reasonable doubt and all the required publications
have been made, the cast-iron validation standard for scientific use can be declared. The cast-
iron validation does not make the claim true but it gives a much higher level of assurance that
can be applied for scientific research. Any validation will always be sensitive to new evidence.
Validator’s Questionnaire
This questionnaire should be filled in by longevity validators and the answers should be made
public. If possible, the claimant and as many friends and relatives as possible should be
interviewed to help get answers to these questions. This will aid quality, transparency, and
consistency of validations.
Please provide all answers where known with as much relevant supplementary details as
possible. In each case reference the full source of the information. Where based on original
testimony, give the name, location, date and relation to the claimant of each witness. Where
details of people are asked for, give their names used, gender, locations, dates or birth, death,
marriage and occupations where known and applicable. For documents indicate the type,
where it can be found, where and when it was created, and provide a transcript or translation.
These questions should be considered as a minimum. Any additional information which is
relevant should be included in the validation report.
Even where full evidence gathering is not possible, it is useful to provide answers to the
following questions where applicable:

36
(Q1) What are the names and relevant affiliations of the validators?
(Q2) When was the validation carried out?
(Q3) What are the known given names of the claimant? Include both formal names and
informal nicknames that are known to have been used.
(Q4) What are the family names used by the claimant? Include any married names and
maiden name used.
(Q5) What is the claimed birth date and death date (if deceased)?
(Q6) Does the claimant have any distinguishing features?
(Q7) Where and when has the claimant lived or possibly lived during their life? Give as
much detail as possible. Include any significant travel away from home.
(Q8) What were the claimant’s occupations and major interests throughout their life?
(Q9) Who were the claimant’s parents?
(Q10) Who were their spouses and other long term partners?
(Q11) Who were their siblings?
(Q12) Who were their children?
(Q13) If possible, provide a genealogical family tree in GEDCOM form.
(Q14) What archival sources have been searched for documents?
(Q15) What are the earliest known documents where the claimant is mentioned?
(Q16) List all written documents that help confirm the claimed date of birth the date,
giving the date, month or year of birth or the age recorded as applicable.
(Q17) What other relevant documents were found?
(Q18) Which anticipated documents could not be found?
(Q19) Were there any cases or early school entry, marriage, pregnancy, military service or
occupation where there may have been an incentive to exaggerate age?
(Q20) Were there any instances of extended travel, isolation, illness or estrangement
when an identity switch may have been less likely to be noticed?
(Q21) List all younger relatives and close friends of the same gender who did not outlive
the claimant or whose fate is unknown and who lived to an age where an identity swap
with the claimant would be plausibly possible.
(Q22) List witnesses who have provided testimonial evidence, either through other
sources or directly to the validator.
(Q23) For each witness give ranges of dates when they were in contact with the claimant
either directly or indirectly. How well can they confirm that the person did not change
identity during that time?
(Q24) Who else is known to have been in contact with the claimant, and for which dates?
(Q25) Provide a chronological and geographical timeline of events in the claimant’s life.
(Q26) Itemise the claimant’s recollections where details can be checked against historical
facts. Indicate how well they matched.
(Q27) Were there any events that they could not recall or recalled incorrectly?
(Q28) What photographs are available of the claimant and potential identity switch
partners, especially from before their longevity was known?

37
(Q29) Have any DNA samples been gathered from the claimant and relative?
(Q30) What were the results of consanguinity tests between each pair of samples, and
autozigosity tests of samples where relevant?
(Q31) What other evidence helps to rule out potential identity switch scenarios or age
exaggeration?
Conclusions
We have reviewed the longevity validations for 23 oldest women and 10 oldest men from the
lists of oldest verified supercentenarians, including three cases that were recently withdrawn
(W). We dispute (D) a further three validations that we consider invalid because there is
insufficient evidence that they were supercentenarians (Calment, Hannah and Beard.) In four
cases (Weaver, Bolden, Talley, Barnes) the validation is unsafe due to likely age exaggeration
of a few years although they probably were supercentenarians. Any scientific research that
relies on these cases should be called into question.
We reinvestigated the case of Sarah Knauss in great detail and with due scepticism because
she was the all-time oldest undisputed verified person. Two new previously unknown records
of significance from her early life were located. These support her longevity claim, but there
is still some doubt left because of her missing baptism record. Likewise we studied the case
of Christian Mortensen and found new records that filled gaps in his life. There is not much to
contradict the verity of his longevity claim but his often solitary and mobile lifestyle means
that an identity switch cannot be ruled out without a DNA test.
Similar levels of uncertainty were attributed to the cases of Meilleur, Kimura and Breuning
and other cases where a potential for identity swaps or incorrect birth date remains (R). From
these 12 cases only 7 have published validation reports as of December 2019 and our
confidence in reliability of these cases varies. We advise scientists who are going to use them
in their research to evaluate the presented evidence. In the remaining 10 cases, a lack of
validation reports and inaccessible records mean that there was insufficient evidence to
review. We are looking forward for evaluating new validation reports when they are
published.
We highlight the following general conclusions
- There is inadequate documentation of validations
In 20 of the 33 cases we looked at there is no published validation report available. Where
there are reports, only a few cases gave sufficient details needed to review the case.
Important unpublished documents have almost certainly been lost because they have not
been stored in any public archives. We understand that validation reports for some recent
cases may be published soon (e.g. refs [35,36]).

38
We accept that reports cannot be written for every supercentenarian validated, but given the
nature of the subject, it is essential that validation reports for record cases be written up and
published promptly by validators, with as much detail as possible. There is always a possibility
that new information will be found that requires a validated case to be re-examined. If the
validation has been done carefully this should not be an onerous task. Indeed the process of
writing a report is the best way to ensure that no gaps have been left.
Validation documents including photos and testimonies collected should be preserved. This
can be done by submitting them to a local university library for example. While privacy is a
genuine legal and moral concern it should not be an obstacle for preservation. Where
personal details of living people form part of the dataset the appropriate permissions should
be sought.
We hope that our sample questionnaire will help prompt for more necessary details to be
included in reports.
- Validation standards are often too lax.
In many cases validations are completed with only three or even two records. This is
insufficient to rule out many possible sources of error. Birth records are accepted up to twenty
years after birth even when further dating evidence is inconsistent. This allows common cases
of early age exaggeration to slip through.
Extraordinary claims require extraordinary evidence, so the most extreme cases of longevity
require a higher level of scrutiny and a higher standard for validation. It is therefore
impossible to set one standard that suffices for all cases. Publication of the standard used and
evidence found in each case is therefore important. We hope that our cast-iron standard will
lead to better practice.
- The potential for identity switch scenarios is underestimated
Although some validators documented that they considered the possibility of identity swaps,
there was rarely any systematic analysis of the options. We felt that they were dismissive of
the potential for fraud or error and accepted weak evidence or hand-waving arguments to
rule them out. A systematic examination of the claimant’s timeline would be appropriate. The
time periods when family members knew them and would witness a switch need to cover the
full lifespan. Where there are gaps, photographs and other evidence can be used to fill in. Any
younger close relatives of the same gender who did not outlive the claimant must be regarded
as potential switch partners at times when they might be indistinguishable by age difference.
DNA testing should always be used where it can rule out any plausible switch scenarios.
- There is too much reliance on the authority of official records
Civil and church records for birth, marriage and death are essential records for validation.
Census returns often fill in missing gaps. However, it should not be assumed that these
records are reliable or authoritative because they are official. In fact there has been very little

39
checking historically against fraud or errors in the public record. Census records are
particularly unreliable because they are gathered mainly for statistical purposes and the
returning officers can guestimate ages or mishear the information provided.
We often hear that the case of Jeanne Calment is robust because there are over 30 official
records documented. The cases of Carrie C. White and Shigechiyo Izumi shows that a single
official error can be replicated to provide a convincing trail of false evidence. Official
announcements by government agencies are also often taken as authoritative when in fact
little evidence has been considered and the officials concerned may not have much expertise.
Izumi was a clear case of this mistake. Other cases wrongly validated in the Kestenbaum study
using SSA records provide other examples of how unreliable official determinations can be
given undue weight.
The official record needs to be supported with other evidence including photographs,
testimony, newspaper reports and DNA tests. The whole should be analysed rationally with
Bayesian inference rather than as a tick-box counting exercise.
- Record setting can compromise quality control
Guinness World Records plays an important role in longevity validation because they
encourage families to come forward with information in order that their elderly relative can
be proudly recognised with a world-record certificate. However, the desire for a record to be
awarded puts pressure on validators from family, care-homes and even governments who see
it as a source of regional of national pride. This could potentially lead to claimants being given
the benefit of the doubt so that the record can be conferred before a full investigation can be
completed. In this way the scientific integrity of the validation may be compromised. We are
not implying that this has happened.
There is no easy solution to this problem but one possibility is to allow for different levels of
validation standard so that a record can be granted where the evidence is good while some
reservations may prevent a higher standard of validation being used for scientific purposes.
The danger in this case is that once a record certificate is awarded the family may be reluctant
to provide further evidence which might compromise the record. All evidence such as
testimony and photographs therefore needs to be gathered before any record is recognised.
- The community is too slow to invalidate obviously false cases once validated.
The longevity validation groups often work on the principle that once a claim has been
validated it can only be overturned where there is direct proof that the validation was wrong.
This is an unwritten rule, but its explicit use can be observed in forum discussions and some
official statements and scientific papers whenever someone challenges a validated case. In
this way the tenet that the onus of proof is always on the validators is turned on its head. In
our opinion this is bad practice.

40
In the case of Izumi it took many years before invalidation was accepted even though the
authenticity of the validation claim was questioned early on. Lucy Hannah remains in the
validated lists even though few people consider her case valid and people who question the
case of Mathew Beard and Jeanne Calment are quickly rebuffed. This is because there is no
clear proof that they were younger than claimed so they remain validated even though the
case for their claim is without substance or highly disputed.
In our opinion any claim should be removed from the verified lists unless the validation can
be replicated at any time using current standards of validation. This is another reason why
validation evidence should always be documented and published.
- DNA testing needs to be used for longevity validation
Finally we note that a high level of validation can sometimes only be achieved using DNA
testing to eliminate cases of identity switching. If concerns are raised at a later date after the
death of the claimant it may be impossible to obtain samples and usage permissions. Even
when someone is alive they may not be sufficiently sound of mind in late life. There should
therefore be a systematic process of gathering samples from anyone who passes the age of
110 while still in good health of mind. Close relatives also need to be asked to provide
samples.
Cold storage of bio-samples would require ongoing funding, but whole genome sequence
results can be stored cheaply on hard disk along with mitochondrial data, methylation
patterns and any other digital information that biologists consider relevant.
Whole genome sequencing is necessary for best quality tests. This is no longer an expensive
process. The DNA data must be held in trust under conditions for common use that meet the
requirements of the family. It is important to take measures that ensure that control of the
information cannot be lost to future public scientific use for the purposes of validation.
- The IDL needs rejuvenating
The International Database on Longevity was an ambitious project with worthy goals. It has
failed to have the impact on scientific research that it should have had. One reason for this
was that it was too early to benefit from good quality records of birth globally. The quality of
the validations was therefore too low to provide conclusive information on demographics at
extreme longevity. There could be too many cases like Beard, Hannah and Izumi present in
the database. The recent update which is still incomplete has been disappointing although for
some countries such as France there is now good data.
In our opinion it would be worthwhile relaunching the project with a new effort. The number
of supercentenarians has increased dramatically. It may now be possible for more countries
to participate. Above all, the quality of validation at the highest ages needs to improve.
Avoiding age dependent bias is difficult but this should not be achieved by reducing to the
lowest common standard. Instead weightings can be provided to reflect the level of reliability

41
and completeness in each age and region sector. Possibility of identity swaps should never be
ignored.
Acknowledgments
In addition to the many people who helped us with the Calment case we are grateful to Per
Hagemann for a huge amount of help searching Danish records for Christian Mortensen. We
would also like to pay our respects to the supercentenarians featured in this study and their
families who we believe have always acted with integrity and in good faith.
Declaration
There have been rumours appearing in the press, on TV and even spread openly in conference
talks that our collaboration is secretly manipulated either by government agencies with a goal
to undermine the integrity of scientific research, or by “GAFA-funded” life-extension groups
seeking access to DNA from supercentenarians. We wish to make it absolutely clear that these
conspiracy theories are entirely baseless. We are unfunded and are motivated only by
scientific curiosity and our aim is to bring integrity back to science. We have not received any
payments or fees for our gerontology work beyond normal conference expenses and do not
expect any remuneration in the future. We have no business or financial interests of any kind
that would benefit from this research. We are also grateful to have been aided with material
found by many other amateur enthusiasts.
We consider these rumours to be malicious, defamatory, slanderous and damaging but we
do not believe legal action to be a suitable process for scientific debate. We ask only that
where such accusations are being made, we are also invited to participate and are given the
rightful opportunity to present our side of the argument. To anybody who witnesses anything
being said against us in public meetings without us present we request that you ask why we
are not being accorded a right to reply – Nikolay Zak and Philip Gibbs.
References
[1] Michel Poulain “On the age validation of supercentenarians” In “Supercentenarians”,
Heidelberg, Springer (2010)
[2] Michel Poulain. “Individual Longevity Versus Population Longevity” in “Centenarians”,
Caruso [et al.], Springer (2019)
[3] John R. Wilmoth and Hans Lundstrom. “Extreme Longevity in Five Countries,
Presentation of Trends with Special Attention to Issues of Data Quality”, European Journal
of Population 12: 63-93 (1996)

42
[4] Wilmoth J, Skytthe A, Friou D, Jeune B. “The oldest man ever? A case study of
exceptional longevity”, The Gerontologist 36(6) (1996)
[5] Robert Young, “Age 115 or more in the United States: Fact or fiction?” In
“Supercentenarians”, Heidelberg [et al.], Springer (2010)
[6] Nikolay Zak, “Evidence That Jeanne Calment Died in 1934—Not 1997”, Rejuvenation
Research Vol. 22, No. 1 (2019)
[7] Jean-Marie Robine, Michel Allard, François R Herrmann, Bernard Jeune, “The real facts
supporting Jeanne Calment as the oldest ever human”, The Journals of Gerontology: Series
A, glz198 (2019)
[8] Nikolay Zak, Philip Gibbs, “A Bayesian Assessment of the Longevity of Jeanne Calment”,
Rejuvenation Research, to appear.
[9] Elisabetta Barbi, Francesco Lagona, Marco Marsili, James W. Vaupel, Kenneth W.
Wachter, “The plateau of human mortality: Demography of longevity pioneers”, Science 29
Jun 2018: Vol. 360, Issue 6396, pp. 1459-1461
[10] Dong X, Milholland B, Vijg J. “Evidence for a limit to human lifespan”. Nature 538, 257-
259 (2016)
[11] Saul Justin Newman, “Supercentenarians and the oldest-old are concentrated into
regions with no birth certificates and short lifespans”, bioRxiv,
https://doi.org/10.1101/704080
[12] Saul Justin Newman, “Errors as a primary cause of late-life mortality deceleration and
plateaus”, PLoS Biol 16(12)
[13] Gierman HJ et al. Whole-Genome Sequencing of the World’s Oldest People. PLoS One.
9(11) (2014)
[14] Robine JM, Herrmann FR, Arai Y, Willcox DC, Gondo Y, Hirose N, Suzuki M, Saito Y.
“Accuracy of the centenarian numbers in Okinawa and the role of the Okinawan diet on
longevity: responses to Le Bourg about the article "Exploring the impact of climate on
human longevity".”, Exp Gerontol. 2013 Aug;48(8):840-2 (2013)
[15] Gerontological Research Group, Table B, http://www.grg.org/Adams/B.HTM, retrieved
27 Oct 2019.
[16] Robert D. Young, Waclaw Jan Kroczek, “Validated Living Worldwide Supercentenarians
112+, Living and Recently Deceased: February 2019”, Rejuvenation Research, Vol. 22, No. 1
(2019)

43
[17] Wikipedia “List of the verified oldest people”
https://en.wikipedia.org/wiki/List_of_the_verified_oldest_people , retrieved 27 Oct 2019.
[18] Bernard Jeune, Jean-Marie Robine, Robert Young, Bertrand Desjardins, Axel Skytthe,
James W. Vaupel “Jeanne Calment and her successors. Biographical notes on the longest
living humans” In “Supercentenarians”, Heidelberg [et al.], Springer (2010) ”
[19] J.-M. Robine and M. Allard “Jeanne Calment: Validation of the Duration of Her Life” in
“Validation of Exceptional Longevity” (Odense Monographs on Population Aging Vol. 6)
[20] Robine JM, Allard M, “The oldest human”. Science 279:1834–1835 (1998)
[21] Moyer, E.R., “Sarah Clark Knauss. Age 117 years”, Phoebe Ministries (1997)
[22] B. Jeune and J.W. Vaupel, “Species of Evidence of Exceptional Longevity”, in “Validation
of Exceptional Longevity” (Odense Monographs on Population Aging Vol. 6) (1999)
[23] Robine, J.M. and Vaupel, J.W. “Emergence of supercentenarians in low mortality
countries”, North American Actuarial Journal, 6(3):54–63, (2002).
[24] Joseph H. Werner, "Inhabitants of Northampton County as furnished by the Census
Bureau at Washington, D.C. in the Eleventh U. S. Census, 1890”, published 1891, relevant
part of transcript available online at
http://www.bethlehempaonline.com/beth1890/abc.html, accessed 9 Nov 2019
[25] Bert Kestenbaum, B. Renee Ferguson, “Supercentenarians in the United States”, In
“Supercentenarians”, Heidelberg [et al.], Springer (2010)
[26] Bertrand Desjardins, “Did Marie Louise Meilleur Become the Oldest Person in the
World?” in “Validation of Exceptional Longevity” (Odense Monographs on Population Aging
Vol. 6) (1999)
[27] Lavilla Mushatt Watson, “Susannah Our Incredible 114-Year-Old Aunt”, The Peppertree
Press.
[28] Los Angeles Gerontology Research Group, “Validated Supercentenarian Cases Aged 114
and Above”, Rejuvenation Research, Volume 10, Number 4, 2007
[29] Gondo, Y., Hirose, N., Yasumoto, S., Arai, Y., and Saito, Y. “Age verification of the
longest lived man in the world.” Exp Gerontol. 2017; 99: 7–17
[30] A. Skytthe, B. Jeune, and J.R. Wilmoth, “Age Validation of the Oldest Man”, in
“Validation of Exceptional Longevity” (Odense Monographs on Population Aging Vol. 6),
(1999)

44
[31] Amandine Cournil, Jean-Marie Robine, Heiner Maier, Jutta Gampe, James W. Vaupel,
“The International Database on Longevity: Structure and contents” in Supercentenarians”,
Heidelberg [et al.], Springer (2010).
[32] “Gerontology Research Group”, Organizational Profile, Journal of Anti-Aging Medicine
VOL. 6, NO. 4 (2003)
[33] L. Stephen Coles, “Aging: The Reality: Demography of Human Supercentenarians”, The
Journals of Gerontology: Series A, Volume 59, Issue 6, June 2004, Pages B579–B586
[34] D. Craig Willcox, Bradley J. Willcox, Qimei He, Nien-chiang Wang, Makoto Suzuki, “They
Really Are That Old: A Validation Study of Centenarian Prevalence in Okinawa”, The Journals
of Gerontology: Series A, Volume 63, Issue 4, April 2008, Pages 338–349
[35] Gondo Y, Hirose N, Yasumoto S, et al. “Age verification of three Japanese
supercentenarians who reach Age 115+.” In: Maier H, Jeune B, Vaupel JW, eds. Exceptional
Lifespans . Heidelberg, Germany: Springer Verlag; to appear 2020.
[36] Jeune B, Poulain M. “Emma Morano – 117 years and 137 days.” In: Maier H, Jeune B,
Vaupel JW, eds. Exceptional Lifespans . Heidelberg, Germany: Springer Verlag; to appear
2020.
[37] Lilian Kristensen, “The oldest Dane”, 1998. Available online at https://dis-
danmark.dk/bibliotek/907990.pdf, accessed 03.12.2019.
[38] Yasuhiko Saito, “Supercentenarians in Japan”, In “Supercentenarians”, Heidelberg,
Springer (2010)
[39] Matsuzaki, T, “Examination of centenarians and factors affecting longevity in Japan”, in
S. Hishinuma (ed.), “Why do the Japanese Live Long?”, (Dobun, Tokyo), t20-123. 1988.
[40] John Withington, “Secrets of the Centenarians: What is it Like to Live for a Century and
Which of Us Will Survive to Find Out?”, Reaktion Books, London, 2017
[41] Matsuzaki, T, “Characteristic of centenarians”, in Transactions of the General Assembly
of the Japan Medical Congress, 582–583 (1987)
[42] GRG, “Oldest Validated supercentenarians All-Time”
https://gerontology.wikia.org/wiki/Oldest_Validated_supercentenarians_All-Time, accessed
20 Dec 2019
[43] GRG, “List of male supercentenarians”
https://gerontology.wikia.org/wiki/List_of_male_supercentenarians, accessed 20 Dec 2019

45
[44] G. Duncan Gibb, “Ultra-Centenarian Longevity”, The Journal of the Anthropological
Institute of Great Britain and Ireland, Vol. 5 (1876), pp. 82-101
[45] https://archicentenarians.fandom.com/wiki/Archicentenarians_Wiki