When Official Records Disagree: What Genealogists Can and Cannot Tell a Jury

Home > Blog > When Official Records Disagree: What Genealogists Can and Cannot Tell a Jury

A federal judge in Manhattan denied summary judgment in a $10 million life insurance dispute because the conflicting historical evidence about Catherine Cohen’s birth date was too tangled for any court to resolve without a jury, while simultaneously ruling that the competing genealogists could not tell that jury which records to believe or what her actual birth date was.

The case of Wells Fargo Bank, N.A. v. The United States Life Insurance Company illustrates how federal courts grapple with historical facts when official records point in opposite directions. Catherine Cohen, who was raised in the St. Stanislaus Orphanage in Pennsylvania, died on October 21, 2021, while insured under a life insurance policy worth millions of dollars. The central question was whether she was born on May 10, 1921, the date she consistently used throughout her life, or on January 29, 1920, as her daughters later claimed and as a delayed birth certificate suggested. If she was born in 1920, her policy matured on November 8, 2019, nearly two years before her death, meaning the insurer, US Life, owed nothing. But if she was born in 1921, the policy was still in force, and Wells Fargo, as the policy’s owner and beneficiary, was entitled to the full death benefit. With no contemporaneous birth certificate available, both sides turned to genealogists to help unravel Catherine’s true date of birth, and Judge John P. Cronan’s 71-page opinion provides essential guidance on how far such experts may go in telling a federal jury which historical records to believe.

The Competing Historical Records

Judge Cronan’s opinion meticulously reconstructs the evidentiary record, and it is a genuine mess. The evidence supporting May 10, 1921, came from Catherine herself and from numerous official documents she had completed over the course of her long life. Her 1940 Social Security application listed that date, as did a 1944 record. Her 1970 and 2007 State Department passport applications similarly reflected May 10, 1921, as did her voter registration records, DMV records, and a Florida ID and driver’s license. Most significantly, Catherine submitted a sworn affidavit in 2010 in which she attested under penalty of perjury that she was born on May 10, 1921. Her siblings also submitted affidavits confirming that date, and these were not casual recollections, they were sworn statements from family members who had been alive at the time of Catherine’s birth, albeit at a young age. US Life did not contest that throughout her life, Catherine consistently identified May 10, 1921, as her date of birth.

The insurer, however, put together its own substantial case. Catherine’s daughters, Carol Cohen and Marie Levitt, swore in affidavits that their mother was actually born on January 29, 1920. According to them, Catherine and her siblings had never known their true birthdays because they had been separated from their parents and placed in the orphanage at a young age. Catherine’s older sister simply picked May 10, 1921, because it fell on Mother’s Day, and Catherine used that date for the rest of her life. Marie further testified at her deposition that her mother, Catherine, was the person who told her the story about the birthday being assigned in the orphanage by an older sibling. US Life also pointed to a baptismal certificate from St. Stanislaus Orphanage and orphanage records containing the January 29, 1920 date.

Yet those orphanage records were far from reliable. Judge Cronan noted they contained “multiple internal inconsistencies” and “varying dates of birth” for Catherine and her siblings, with some records reflecting 1920, others 1922, and even 1930. The orphanage records also contained correspondence from the State of Pennsylvania dated October 1936 through March 1937, which suggested that the orphanage had asked the state to search for a birth certificate for Catherine. The State of Pennsylvania apparently searched records for the years 1920, 1921, and 1922, but failed to locate a birth certificate. This further added to the ambiguity surrounding the year of Catherine’s birth. While US Life argued that these letters could not prove what the orphanage wrote to the state, Judge Cronan concluded it was for the jury to decide what inferences to draw from the state’s correspondence.

The 1943 Delayed Birth Certificate

US Life also pointed to Catherine’s delayed birth certificate, which she obtained from the Pennsylvania Department of Health in 1943 and which recorded her birthday as January 29, 1920. The certificate reflected that its information was based on a school record, the baptismal record mentioned above, and a notarized affidavit signed by “Albenia Deminski,” who was noted as being Catherine’s aunt. But Catherine’s daughter Marie testified that she had never heard of Albenia Deminski, casting at least some doubt on Deminski’s relationship to Catherine and the basis for her apparent assertion that Catherine was born in 1920. And while Catherine may have filed an affidavit with the Pennsylvania Department of Health at some point to obtain that delayed birth certificate, that affidavit was not submitted to the court. Wells Fargo therefore appeared to have grounds to challenge the reliability of Catherine’s delayed birth certificate. Judge Cronan acknowledged that the delayed birth certificate carried probative value, but emphasized that it was “just one piece of evidence in an otherwise conflicted record.” On summary judgment, he wrote, it was not appropriate for the court to weigh that delayed birth certificate against a similarly strong piece of evidence like an affidavit signed by Catherine herself. Likewise, even assuming Catherine swore an affidavit attesting that she was born on January 29, 1920, in order to receive the delayed birth certificate, that only further muddled the record with contradiction on a material factual issue.

The Social Security Puzzle

Even Social Security records contradicted themselves. US Life pointed to Social Security Administration records which listed Catherine’s date of birth as January 29, 1920, and noted that Catherine’s eligibility for benefits began in 1985, when she would have been just sixty-four years old if she truly had been born in May 1921. But Wells Fargo identified inconsistencies, including that Catherine first applied for benefits on May 15, 1986, which is more consistent with a May 1921 birthday. Even with US Life’s counterarguments, including that Catherine’s application for a Social Security number recorded May 5, 1921, rather than May 10, 1921, as her birthday, Judge Cronan concluded that the court could not determine as a matter of law that Catherine’s birthday was January 29, 1920, particularly given the exacting standards governing US Life’s motion for summary judgment.

The Court’s Holding: A Genuine Dispute

After reviewing the record and the parties’ arguments, Judge Cronan concluded that US Life may have a story to tell a jury. May 10, 1921, was simply a birthday assigned to Catherine by her older sister after she was taken in at the St. Stanislaus orphanage, and she in fact was born over fifteen months earlier on January 29, 1920. US Life had evidence to support that story, and it might ultimately prove persuasive. But the record had too much conflicting evidence as to Catherine’s date of birth for the court to resolve that factual question on summary judgment. While US Life urged the court to view Wells Fargo’s evidence through the lens of the uncontested evidence that Catherine’s sister selected that date to serve as Catherine’s birthday, that evidence was very much contested. Indeed, the evidence presented by the parties also supported the plausible inference that no one truly knows when Catherine was born. Mindful that in the present posture the court must resolve all ambiguities and draw all reasonable inferences in favor of the nonmoving party, as well as the presumption in favor of the accuracy of the birthdate listed on the policy, Judge Cronan concluded that a genuine dispute of material fact persisted on the true date Catherine was born. As US Life’s ability to prevail on its defenses hinged on establishing that Catherine was born on January 29, 1920, or otherwise on some other date which would have resulted in an attained age of 100 years or older upon her death, its summary judgment motion was denied.

The Role of Genealogical Experts

Beyond the historical dispute, however, Judge Cronan’s opinion offers crucial guidance on the role of genealogical experts in federal court. Both sides retained certified genealogists. US Life brought Roger Joslyn, a certified genealogist since 1981, a Fellow of the American Society of Genealogists, and a specialist in the Mid-Atlantic region, who opined “to a reasonable degree of certainty, that Catherine Cohen was born on January 29, 1920.” Joslyn reached this conclusion after reviewing records produced during the litigation and publicly available materials like census results, social security indices, and news articles.

His approach illustrates the broader challenges involved in genealogy and historical records research, particularly when surviving records contain conflicting dates or information. He explained in his report that the genealogical research process starts with identifying relevant records about a person, then resolving any conflicting information by considering when, where, and why the conflicting records were created, assessing the reliability of a record by considering whether it is a primary, secondary, or derivative source, and considering how the information fits contextually with other information about related topics. Wells Fargo moved to exclude Joslyn’s testimony on several grounds, principally arguing that his proffered testimony would overlap with the jury’s tasks of weighing the various pieces of evidence and reaching a conclusion about Catherine’s true date of birth.

Wells Fargo countered with Kathleen Hinckley, a certified genealogist since 1984 and former Executive Director of the Association of Professional Genealogists. Hinckley’s report concluded that Joslyn’s analysis was flawed because he relied heavily on orphanage records and the delayed birth certificate, which in her view were not reliable sources. She also opined that Joslyn improperly ignored Catherine’s use of the 1921 year of birth beginning in 1940 and throughout her life, and she ultimately concluded that absent a contemporary birth record or infant baptism, Catherine’s date of birth could not be determined with any reasonable degree of certainty because there were too many inconsistencies and conflicts. US Life moved to exclude Hinckley’s testimony as well.

What Genealogists Could Not Do

Judge Cronan held that most of what both genealogists sought to say was inadmissible. Joslyn could not teach the jury a methodology for deciding which historical records were more reliable, could not rank the competing records or tell the jury which documents were more reliable, and could not tell the jury Catherine’s actual birth date. That ultimate conclusion was for the jury to reach. The court explained that expert testimony may not usurp the province of the judge to instruct on the law, or of the jury to make factual determinations, nor may an expert act as a vehicle to present a factual narrative in effect simply accumulating and putting together one party’s story. The court specifically cited Joslyn’s opinions that certain records “are obviously incorrect,” that a date in one document “was likely an error,” that he did not consider Catherine’s Social Security application to be reliable proof because it contained other information inconsistent with the total body of information about Catherine’s family, that the delayed birth certificate was more reliable than the documents reflecting a May 1921 birthday, and that certain other documents should not be accepted as accurate. By opining not just on Catherine’s date of birth but also on the reliability and consistency of the documents presented by the parties in support of their respective views, Joslyn’s testimony would usurp the fact-finding function of the jury. The court applied the same boundary to Hinckley: she could not simply counter Joslyn by telling the jury which records were reliable or what Catherine’s birth date was. Her conclusion that Catherine’s date of birth could not be determined with any reasonable degree of certainty was also excluded, as that too was for the jury to decide.

What Genealogists Could Do

Yet the court did allow something important: expert testimony explaining historical context surrounding unfamiliar records. Joslyn could explain such matters as early-twentieth-century immigrant recordkeeping practices, including why names were often recorded using numerous different English spellings; the historical absence of birth registration in Pennsylvania, where statewide recording was not implemented until 1906 and births often went unrecorded even after that date; and Pennsylvania’s procedures for obtaining delayed birth certificates in the 1940s. This historical-context testimony was based on specialized knowledge not possessed by a lay juror and would assist jurors in evaluating and weighing various records at trial. Judge Cronan noted that Joslyn was qualified to offer such testimony because of his genealogical research experience in the Mid-Atlantic region, his familiarity with Pennsylvania’s historical practice for contemporaneous recording of births, and his experience with what delayed birth certificate forms were in use in Pennsylvania counties in the 1940s. Unlike the testimony in the first category, this testimony would be helpful to the trier of fact by providing background context on documents the jury was likely to see at trial, without opining on the reliability of any particular document. The court permitted Hinckley to offer similar historical context and background testimony, but with the same limitations.

The Takeaway for Forensic Genealogists

Judge Cronan’s opinion provides a clear framework for genealogists testifying in federal court. Permissible testimony includes explaining historical context and background for unfamiliar document types, explaining recordkeeping practices in specific time periods and regions, and explaining statutory frameworks that governed the creation of certain records, limited to background context rather than legal interpretation. Impermissible testimony includes teaching the jury a methodology for assessing record reliability, ranking competing records or telling the jury which documents are more reliable, stating the expert’s conclusion as to the ultimate factual question such as the person’s actual date of birth, acting as a narrator who simply tells the jury one party’s story, and offering personal evaluations of the credibility of witnesses or documents.

The distinction is critical: a genealogist can explain what a delayed birth certificate is and how it was issued, but cannot tell the jury whether this particular delayed birth certificate is more reliable than other evidence. That task belongs to the jury. As Judge Cronan explained, by opining not just on Catherine’s date of birth but also on the reliability and consistency of the documents presented by the parties, the experts would usurp the fact-finding function of the jury. The case thus stands as an essential guide for forensic genealogists and litigants alike: experts may illuminate the historical landscape, but they may not tell the jury which path to take through it. The jury, armed with historical context from qualified experts but left to make its own credibility determinations and weigh the evidence for itself, must ultimately decide when Catherine Cohen was born – a factual question that, after nearly a century, remains genuinely uncertain.

For attorneys and other professionals dealing with disputed identity, age, kinship or family-history evidence, genealogy and historical records research may require correlating vital records, church and institutional records, government files, Social Security records, immigration records and other archival sources created at different times and for different purposes.

Case Discussed:
Wells Fargo Bank N.A. v. The United States Life Insurance Company in the City of New York, No. 22-cv-8606 (JPC) (S.D.N.Y., Aug. 4, 2025).

Up Next: The Long Shadow of Secrecy: What Novak v. CIA Reveals About FOIA Requests for Intelligence Records