A large proportion of all verifiable supercentenarians who have ever lived come from the United States. Approximately 2,000 American supercentenarians born between 1814 and 1913 have been documented so far, and the proportion that are validated is continually increasing.

However, the United States of America is a vast mixture of different states, each with its own systems for recording and keeping documents (Mulcahy & Ford, 2019). This diversity results in the availability of certain documents, such as birth records, in some states but not in others.

While birth records are never the sole form of documentation needed to validate the age of a supercentenarian, they are usually among the strongest forms of documents. This is because they help establish a family relationship between the supercentenarian claimant and their parents, a connection that can only be inferred from other forms of documentation.

In general, civil birth records were not consistently kept in most states until the late 1800s. Some states have church records, such as christening or confirmation records, that can help locate this information. However, for many states, this information is unavailable, necessitating the implementation of alternate strategies.

The U.S. Census

The United States has conducted a census since 1790 (Krieger, 2019), with a decennial schedule. According to the 72-year rule for releasing census records, the most recent publicly available census is the 1950 US Census (United States Census Bureau, n.y.).

The primary purpose of the census is to enumerate the population of the United States and collect essential information about individuals (Winkle, 1994). In the first six census enumerations (1790-1840), only the head of the household was listed, along with notes on age ranges (e.g., 10-14 years old) and status (free white, other free person, or slave).

Starting with the 1850 census, each free person in the household was individually listed with their name and basic descriptors (Winkle, 1994). Over time, the information recorded in census records has expanded to capture key sociodemographic aspects of the population.

Most census records have been well-preserved and are publicly accessible. Unfortunately, a significant portion of the 1890 census was lost in a fire in 1921 (Winkle, 1994; United States Census Bureau, 2021).

Using census records to validate supercentenarians

For many American supercentenarians, an entry in the census is often the earliest available document, and in many cases, it may be the only early-life document stating an actual age for the person. This makes the census crucial for verifying claims with higher certainty.

Delina Filkins (1815-1928) on her 113th birthday. Photo source: FindAGrave

Early supercentenarians like Louisa Thiers (1814-1926) and Delina Filkins (1815-1928) can be proxy-validated using the 1820 and 1830 censuses (both listed as under ten in 1820 and 15-19 in 1830). Although not mentioned by name, family reconstruction has allowed for the identification of these individuals and validation of their age claims. No other validated American supercentenarians born before 1840 have been identified in the pre-1840 census records.

Considering that the number of supercentenarians increased rapidly in cohorts born after 1880, the loss of almost the entire 1890 census poses a significant challenge. For supercentenarians without another document (birth record, state census, etc.), there might be a gap of up to 20 years between their birth and the first time they are documented in the 1900 census (enumerated as of June 1).

Sarah Knauss (1880-1999). Photo source: aginganalytics.com

Sarah Knauss (1880-1999) was for a long time viewed with some skepticism, having been born in September 1880 but first appearing in the Census in 1900, meaning that she was a grown woman at the time. Considering that people can exaggerate their ages for a variety of reasons, especially once they are nearing adulthood (for example to get married or enlist in the army), the Sarah Knauss claim was therefore viewed by some with slight skepticism. I personally located Knauss in an 1894 church book entry where she had taken communion, but no age was given in this entry. Neither was an age given in the index in the church book from about 1888, where Knauss and her family members were listed. As luck would have it however, a recent discovery of a transcription of the 1890 census for the area Knauss and her family lived in that had been made in 1891 stated that Knauss was ten years old at the time.

While some supercentenarian claimants may be fortunate to have their 1890 census documents transcribed, others may only have documents where they were close to adulthood. The question arises: Is such a document sufficient? Arguments exist on both sides, with some advocating for a stricter “10-year rule,” while others are more lenient with a “20-year rule.”

An example of an index to documents that are available for some American supercentenarians. Note the inconcistencies in age reporting in some of the rows.

Complicating the validation process is the fact that some supercentenarians were either missed in their earliest census records or enumerated with an age inconsistent with their age claim. Mid-life census records, particularly for female supercentenarians, often exhibit mild to severe age deflation. An unvalidated (but documented) example involves a woman claiming to be 17 years younger in the 1920 census than indicated by her earliest document (1880 census) and later-life claims. Notably, the 1900 census stands out as the only census record recording a person’s month and year of birth instead of their age. Many supercentenarians also have a month of birth listed in 1900 that is inconsistent with later-life claims.

The question arises: How much credibility should be given to census records? While they are generally useful to gain insight into whether a person was a supercentenarian, caution is necessary in interpretation. When the supercentenarian claimant did not respond directly, minor inconsistencies may arise, such as a relative misremembering the month or year of their relative’s birth. Thorough validation requires constructing an overarching life story using all available documents, interpreting and weighing each document together.

An example of a U.S. census entry, in this case the 1900 census for Katie Hatton (1877-1992). Note that Hatton (Dolly) is listed as being three years older than claimed, which showcases that the census isn’t always 100% reliable.

A unique scenario arises when a census record supports a supercentenarian being older than claimed. My professional opinion is that to upgrade a deflated supercentenarian claim, two consecutive early-life census records (issued before the age of 20) supporting a higher age are needed, without another early document contradicting them. However, it’s important to note that not all researchers share this view.

The SSDI and SS-5

The Social Security Death Index (SSDI) serves as a valuable document for confirming that a person indeed reached the age of 110. However, caution must be exercised in interpreting the SSDI due to instances of ‘ghost cases,’ where a person passed away decades earlier but was only officially registered as deceased at a later date. The index is compiled from a death master file and includes most individuals with a social security number at the time of death (Hill & Rosenwaike, 2001/2002). Covering the time span from 1962 to 2014, recent entries are restricted from public access.

Social Security card application forms (SS-5) also play a crucial role as genealogical tools. These forms often contain information such as the names of a person’s parents and their place of birth. The relatively recent release of SS-5 forms on genealogical websites has facilitated more in-depth research on previously validated supercentenarians. Information from SS-5 forms has helped debunk claims associated with two individuals, Lucy Hannah (1875?/1894?-1993) and Mathew Beard (1870?/1886-1985), who were previously considered supercentenarians.

Other forms of documentation

As mentioned earlier, the United States comprises numerous states, each with its unique set of available documentation. States like New York and New Jersey, for instance, maintained state censuses, while others, like Mississippi, had school censuses. Marriage records spanning various periods from the 19th to the 20th century are accessible in most states. Immigration records are available in the form of passenger lists and naturalization records for some states. Additionally, some states have public address registers for late-life documentation.

Conclusion

In conclusion, although not necessarily as extensive as the documentation found in certain European countries, the United States possesses a diverse array of records that allows for robust age validation of supercentenarians.

 

References

Hill, M. E., & Rosenwaike, I. (2001/2002). The Social Security Administration’s Death Master File: The Completeness of Death Reporting at Older Ages. Social Security Bulletin, 64(1).

Krieger N. (2019). The US Census and the People’s Health: Public Health Engagement From Enslavement and “Indians Not Taxed” to Census Tracts and Health Equity (1790-2018). American journal of public health109(8), 1092–1100. https://doi.org/10.2105/AJPH.2019.305017

Mulcahy, B. L. & Ford, C. (2019). Family History – A Concise Beginner’s Overview. Fort Myers Regional Library. https://www.leegov.com/library/Documents/Family-History-Concise-Beginner-Overview.pdf

United States Census Bureau. (n.y.) The “72-Year Rule.” Retrieved 2023-12-17 from: https://www.census.gov/history/www/genealogy/decennial_census_records/the_72_year_rule_1.html

United States Census Bureau. (2021) January 2021. Retrieved 2023-12-17 from: https://www.census.gov/history/www/homepage_archive/2021/january_2021.html

Winkle, K. J. (1994). The United States Census and Community History. The History Teacher, 28(1), 87-101. https://doi.org/10.2307/494293

When we read in the news that a person has turned 110, we as supercentenarian researchers are naturally thrilled. But there is a need to have a critical mind from the very start because many supercentenarian claims turn out to be false, especially the older the person claims to be.

The reasons for age exaggeration are many; a person can exaggerate their age from a very young age as a result of having married young or having become a parent early, there are also examples of a person exaggerating their age to both get into service, but also to avoid it (Wilmoth & Lundström, 1996; Poulain, 2010). Some people claim to be older in order to start collecting their pension early and others exaggerate their ages later in life in order to be noted by their community and media as especially long-lived individuals. Age exaggeration can also come in the form of identity theft, wherein a person assumes the identity of a parent or older sibling for various reasons (Wilmoth & Lundström, 1996).

It must also be noted that age exaggeration can be accidental, such as a person being mistaken for someone else, which can result in a person being attributed a much longer lifespan than they have actually lived. One example of this is the case of Pierre Joubert (1701? – 1814) where a father and son with the same name were mistaken for one another, resulting in this duo being recognized as the only “validated” supercentenarian for quite some time (Charbonneau, 1990; Desjardins, 1999).

From time to time there will also be examples of age deflation, where a person claims to be younger than they actually are. These cases are often not discovered since few of them actually live to be supercentenarians. One such example is that of Emma Isbell (1859-1969), who claimed to have been born in October of 1860 and died in December 1969, allegedly aged 109. However, she was listed in the 1860 United States Census as being eight months old as of June 1, 1860, meaning that it wasn’t possible for her to have been born in October of that year. Thus, in all likelihood she was actually a supercentenarian, something which was also supported by other early-life documentation.

William Thoms, the pioneer of age validation. Photo courtesy of: The Victorian Web

Age validation of exceptional longevity really originated with William Thoms and his rules for age validation (Thoms, 1873). Over the next century and a half many researchers have attempted to find the “golden rule” for age validation, with many different criteria being proposed (Poulain, 2010; Skytthe & Jeune, 1995; Rosenwaike & Stone, 2003). These criteria have been utilized and adapted for various purposes, with a supercentenarian claim being categorized as having various degrees of validation depending on how thorough their documentation is.

At the most basic level there exists proof that this person exists or existed and claimed to be a supercentenarian. This proof can come in various forms, such as a newspaper report, an obituary or a death record (Skytthe & Jeune, 1995). The next level is reached when there exists some sort of documentation supporting the age claim. Generally, the earlier a document is issued, the better the reliability of it actually supporting the correct age of the person. However, one document is not sufficient if a supercentenarian should be considered to be a validated supercentenarian. For most supercentenarians that are validated by organizations specializing in age validation there exist several documents that are in accordance with the person being a supercentenarian (Jeune & Poulain, 2021).

Some records, however, can be unreliable, especially documents issued when a person was older or when the person themselves wasn’t the respondent. Census records are generally such records that can be used to validate a person’s age with a lower degree of certainty, but they have been proven to be very unreliable, especially for claimants from the United States, where an age can fluctuate by several years from census to census. In order to achieve a higher standard of validation, birth or baptismal records are needed since they will generally contain an actual date of birth, something which is unfortunately lacking in almost all censuses or marriage records.

Still, it shouldn’t be considered sufficient that there are potential document matches for a person that is or was an alleged supercentenarian. Without doing a proper background check it is still possible for inaccurate attributions of identity to occur: especially mistaking a person for their earlier-born sibling is something that has occurred more than once for supercentenarian researchers.

Such is the case of Shigechiyo Izumi (1865? – 1986), who for a long time was recognized as the oldest man of all time (Matsuzaki, 1987; Glenday, 2010). Research later indicated that his extreme age was caused by his name given to him as a necronym for a sibling that had been born 15 years before him, thus making him only 105 years old at the time of his death.

Newspaper coverage of Shigechiyo Izumi’s purported 120th birthday. Photo courtesy of: Newspapers.com

For some supercentenarians a validation with the highest degree of certainty will not be possible. In many areas birth registration was not compulsory when they were born, leading to them not having any birth record. And family bibles, if they existed in the first place, have sometimes been lost. Therefore, there is a need for age validation to also accommodate for these instances. If the documentation is in accordance with the person being a supercentenarian, they should be recognized as such, but it should be noted that the validation was approved with a lower degree of certainty.

To achieve a very high-level of validation, a full family reconstruction is necessary (Skytthe & Jeune, 1995; Poulain, 2010). When performing a family reconstruction, each family member that can be considered relevant is investigated in order to rule out identity-theft, accidental mix-ups, and other potential scenarios that could result in the person not actually being a supercentenarian. Each member is therefore researched, and their vital information is noted in order to paint a complete picture of a supercentenarian claimant’s biography. Often this information can be extracted from the claimant (Poulain, 2010) or their relatives. In other cases, this information can be traced from vital documents such as a death file.

The need for a claim to be objectively reviewed by more than just one researcher to eliminate bias has been addressed in various ways. LongeviQuest, for example, has a Global Validation Commission consisting of researchers that review the research that is presented to them and analyze a claim by how well it adheres to the validation standards that have been postulated by the organization.

To show how age validation works in practice, it should be noted that some countries have better documentation than others. The Nordic countries especially are among the best in the world with their system based in church records and a personal identity number that allows for a person and their family members to be traced throughout their life. This will usually easily rule out wrongful claims.

Carl Mattsson, Swedish supercentenarian, the day he became the oldest Swedish man ever. Photo taken 18 July 2019 by Helena Erlandsson.

One such example of a very high-level validated supercentenarian is Carl Mattsson (1908 – 2019) of Sweden who possesses over 30 documents supporting him being a supercentenarian. Still, even in a country such as Sweden issues can exist in ascertaining how old a person actually is. Mattsson claimed to have been born on March 7, 1908, something which is supported by all documents except for his earliest document, a christening record which states that he was born on March 9, 1908. This discrepancy is, however, minor and doesn’t change the fact that he was a supercentenarian. A full family reconstruction could be performed and confirmed his identity (Appendix 1).

In conclusion, age validation of supercentenarians can take many forms and can vary from being very unreliable to having a high standard of validation, which is something that all researchers should strive to achieve in order for their data to be as accurate as possible.

 

References

Charbonneau, H. (1990) La Rubrique du P.R.D.H.: Pierre Joubert a-t-il vécu 113 ans? Mémoires de la Société généalogique canadienne-française, 41, 45–48.

Desjardins, B. (1999). Validation of extreme longevity cases in the past: The French-Canadian experience. In Jeune, B. & Vaupel, J. W. (Eds.), Validation of exceptional longevity, Odense Monographs on Population Aging. Odense: Odense University Press.

Jeune, B., & Poulain, M. (2021). The First Supercentenarians in History, and Recent 115 + −Year-Old Supercentenarians. An Introduction to the Following Chapters. In H. Maier, J. Vaupel, & B. Jeune (Eds.). Exceptional Lifespans. Springer. Demographic Research Monographs. https://doi.org/10.1007/978-3-030-49970-9_14

Matsuzaki, T. (1987). Characteristic of centenarians. In Transactions of the General Assembly of the Japan Medical Congress.

Poulain, M. (2010). On the age validation of supercentenarians. In Maier, H., Jeune, B., Robine, J-M., & Vaupel, J. W. (Eds.). Supercentenarians. Springer. Demographic Research Monographs. http://dx.doi.org/10.1007/978-3-642-11520-2_1

Rosenwaike, I., & Stone, L. F. (2003). Verification of the Ages of Supercentenarians in the United States: Results of a Matching Study. Demography, 40(4), 727–739. https://doi.org/10.2307/1515205

Skytthe, A., & Jeune, B. (1995). Danish centenarians after 1800. In Jeune, B., & Vaupel J.W. (Eds.), Exceptional longevity: From prehistory to the present, Odense Monographs on Population Aging. Odense: Odense University Press.

Thoms, W. J. (1873). Human longevity, its facts and its fictions. John Murray, London.

Wilmoth, J. R., & Lundström, H. (1996). Extreme longevity in five countries: Presentation of trends with special attention to issues of data quality. European Journal of Population, 12, 63–93. https://doi.org/10.1007/BF01797166

 

 

Appendix 1 – Family tree reconstruction of Carl Mattsson (1908-2019)

Parents
Rutger Mattsson (11.10.1877 – 26.11.1943)
Agda Elisabet Berndtsdotter (7.12.1877 – 7.9.1961) m. 25 Oct 1906
Siblings
Bror Magnus (16.12.1906 – 7.6.1993)
Otto Ragnar (30.7.1909 – 18.10.1985)
Erik Valdemar (12.1.1912 – 24.11.2004)
Ingrid Maria Eriksson (9.4.1915 – 5.12.2007)
Sigrid Alfrida (12.10.1918 – 20.4.1919)
Spouses
Sonja Juleida Margareta Behrendt (22.8.1907 – 20.4.1990) m. 12 Nov 1932, d. 7 Nov 1955
Ella Birgit Viola Andersson (21.10.1929 – 27.3.2010) m. 16 May 1964
Children
Berit Irene (8.3.1939 – 23.3.1984)