Charles Franklin Miller and the Genealogical Proof Standard

By Connie Lenzen, Certified Genealogist

Genealogists typically find several dates for vital events. Some genealogy programs permit multiple “facts,” but it does not make much sense to continually write that a person was born on this date, or that date, or maybe a third date. There comes a time when an analysis of the evidence must be conducted, conflicts resolved, and a conclusion written.


Four different birthdates were found for Charles Franklin Miller (known as Frank), the son of John L. and Nellie (Baker) Miller. All four are direct evidence that answers the research question, “When was Frank Miller born?” However, one person cannot be born on four different dates. Only one is correct, but which one? It's times like this that the Genealogical Proof Standard comes into play.



Left: Frank Miller, ca. 1908, about 21-years-old

The five steps of the GPS are fairly straightforward:

  1. We conduct a reasonably exhaustive search in reliable sources for all information that is or may be pertinent to the identity, relationship, event, or situation in question;
  2. We collect and include in our compilation a complete, accurate citation to the source or sources of each item of information we use;
  3. We analyze and correlate the collected information to access its quality as evidence;
  4. We resolve any conflicts caused by items of evidence that contradict each other or are contrary to a proposed (hypothetical) solution to the question; and
  5. We arrive at a soundly reasoned, coherently written conclusion.

Source: The BCG Genealogical Standards Manual (Orem, UT; Ancestry Publishing, 2001), 1.


Frank Miller was born in 1887 in Charlotte, Michigan. This was either on 12 May, 22 May, 9 July, or 14 July. In order to determine the truth, documents were acquired, the conflicts were resolved, and a conclusion was reached.

The family’s understanding of Frank’s birth date

Tzilla (Titus) Miller, Frank’s wife, recorded her daily activities in a series of pocket diaries from 1935 to 1975. In 1935 and every year until 1952, Frank’s birthday was celebrated on 9 July. In 1953 and every year thereafter, his birthday was celebrated on 14 July.

The diaries are original sources. The information is primary as to the celebration but secondary as to the date of birth because neither Tzilla Miller, the informant, nor Frank Miller, the subject, had first-hand knowledge of Frank’s birth. Presumably Frank’s father who would have first-hand knowledge about his birth told him his birth date.

The informant’s reliability needs to be considered as part of the analysis. Tzilla presumably had no reason to misrepresent the facts; she was merely recording the day’s events. In 1935, she was forty-six years old; in 1954, she was sixty-five years old. Even after putting in a full day of housework and party preparations, her handwriting was firm; indicating she was not suffering from illness that would rob her of mental clarity.

The evidence is direct because it directly answers the research question of when was Frank Miller born.

The entries for 1936 and 1954 are shown.

Tzilla's diary
1936 1954
Tzilla Lilah (Titus) Miller, “Date Book,” July 9, 1936; MS, 1935 (Portland, Oregon; privately held by Connie (Miller) Lenzen, 2012). Tzilla Lila (Titus) Miller, “Date Book,” July 14, 1954; MS, 1954 (Portland, Oregon; privately held by Connie (Miller) Lenzen, 2012).

Birth records. Documents created close in time to an event are often more reliable than ones created long after an event because the informant for this class of record is usually a person who has been a witness to the event.

Birth record number one: A certified copy of birth. A birth record for Charles Frank Miller was found in the family collections. It was created on 25 July 1952, shortly after Frank celebrated his 65th birthday. He was likely applying for pension benefits and needed birth verification.

Frank Miller’s “Certified Copy of Record of Birth” from the Eaton County, Michigan Clerk gives his birth date as 14 July 1887. It is likely the reason for the change in the day that the family celebrated his birthday.

Frank's birth record
Eaton County, Michigan, Certified Copy of Record of Birth, no. 146 (1887), Charles Frank Miller; Eaton County Clerk's Office, Charlotte.

The back of the document contains the following wording,  “This is a True Copy of your Birth Certificate as shown on our records as of date recorded.” The document, as an “official copy” or “record copy,” is a derivative, but it is treated as “best evidence” when the original document is no longer available. As with all derivative documents, it is necessary to search for the original before we can know whether this record copy is the only document available.

The informant was not named; but we can infer that the informant knew Frank’s mother and father and knew his father’s occupation, because those facts are accurate.

The date of record, 30 June 1888, is almost a year after the event. This suggests that the informant reported Frank’s birth on that day or the official courthouse register was created from loose papers on that day.

Frank’s mother died on 9 February 1888, a few months after his birth. Her residence was named as Charlotte on the record, implying that she was alive when the informant reported Frank’s birth. Frank’s father, John Miller, could have been the informant, but we cannot tell. When it is not known if the unknown informant had first-hand or second-hand knowledge of the event, the primary or secondary information classification cannot be made. We can say that this is a derivative source offering information from an unknown informant with an undetermined degree of knowledge.

Frank’s birth date is given as 14 July 1887 on the birth record. It directly answers the research question. Thus, it is direct evidence. However, the document does not answer a second research question, why did the family celebrate 9 July as his birth date?

Birth record number two: The state-level register. Some Michigan state-level birth returns are online. Birth returns are official copies of the original record, and they may contain more information than a certified copy that is created on a form with pre-printed fields. Frank Miller is named as Charles Miller in this version of his birth record. Since his name was given on the certified copy as Charles Frank Miller, the state-level document is not the same as the document that was used to create the certified copy above.

Left side
Right side

"Michigan Births 1867-1902," index and images, FamilySearch ( : accessed 15 July 2009), birth certificate image, Charles Miller, 14 July 1887, p. 128 (of an unidentified register), entry no. 146, Michigan Department of Vital Records; imaged from Family History Library microfilm 2,320,847, item 3.

We can classify this version of Frank’s birth record as a derivative source with information from an unknown informant who provided direct evidence.

Birth record number three: The county-level register. The county-level birth record contains the same information as the state copy except that “Frank” is inserted above the line. That name was likely penned in 1952 when Frank Miller requested a copy of his birth record. The handwriting in the county- and state-level registers is similar, but there are enough differences to show two different people were the transcribers.

Eaton County, Michigan, Record of Births, 3, 1881-1887, p. 480, no. 146, Charles Frank Miller (1887); FHL film 966,584

The same classifications can be made for this record as for the other two birth records. This is a derivative source with information from an unknown informant who provided direct evidence.

All of the entries on the state-level and county-level register were recorded on 30 June 1888, and there were nine other pages with entries recorded on that date; suggesting the register was created after the fact. The entry immediately after Frank Miller’s entry in both registers is for the daughter of Alick and Minnie Racey. Alick Racey and John Miller had the same occupation. The document does not show this, but Nellie Miller and Minnie Racey were sisters. This could indicate that the same person supplied the information about the two children.

The above birth records demonstrate the value of the GPS admonition to conduct an “exhaustive search in reliable records” and to seek out the original. It also demonstrates that we do not always know if the record we are viewing is the true original. In this case, there were probably loose notes that were transcribed into the county-level register by a clerk. That register was transcribed for the state-level register. The county clerk abstracted information from the county-level register for the birth record that Frank received in 1952.

Censuses are often used as sources because of their easy availability. The microfilmed and digitized copies that we view are not the original sources and should be classified as image copies. As with the above birth records, the informant can seldom be determined. The 1900 census provides direct evidence of the birth month and birth year. Frank C. Miller’s birth month and year are given as July 1887. His birthday is not given because it was not required in this census. Besides this omission, there is no discrepancy between the 1900 census and the above birth records regarding his birth month and birth year.

1900 census

1900 U.S. census, Eaton County, Michigan, pop. sch., Charlotte City, E. D. 64, sheet 4A (penned), 68 (stamped), dwelling 72, family 78, John L. Miller; digital image, ProQuest HeritageQuestOnline (access through participating libraries : accessed 27 August 2009); citing NARA microfilm publication T623, roll 709.

Draft registrations

Frank Miller registered for the World War I draft on 5 June 1917. Digitized image copies of the World War I draft registration cards are found online. 

The informant for the card is the person named on the card, but the provided information is both primary (first-hand) and secondary (second-hand).

Surely Frank knew his occupation and where he worked. He knew he was married and had two children. He also knew he had served in the military and was receiving a pension. All of that information is primary.

While Frank was present at his birth, he would not be able to remember those details. Thus, the information about his birth day and place was secondary. Someone, most likely his family members who were present at his birth, told him his birthday and place. These unknown family informants would have no reason to lie. Presumably there was more than one person at Frank's birth, and they would have independently agreed on the birth date.

Frank’s birth date is given as 9 July 1887. The statement provides direct evidence and answers the research question about when was he born. However, it conflicts with the birth records previously located. Additional sources must be located and analyzed.

“WWI Draft Registration Cards, 1917-1918,” digital images, ( : accessed 26 August 2009), Charles Franklin Miller, serial no. 1495, order no. 58, Draft Board 2, Charlotte, Michigan; citing World War I Selective Service System Draft Registration Cards, 1917-1918, NARA microfilm publication M1509; no specific roll cited.


The Selective Service Administration’s Fourth Draft Registration was for men who were born on or after April 28, 1877 and on or before February 16, 1897.  Registration day was 26 April 1942, and Frank Miller dutifully complied with the law. The following image copy of the original document located at the National Archives and Records Administration Pacific Alaska Region facility gives a third birth date, 22 May 1887. The source is original, the information is primary, and the evidence is direct; but is it correct? The only way we can find out is to follow the GPS steps: a reasonably exhaustive research in reliable sources; analysis and correlation of information to assess its quality as evidence; resolution of conflicts; and a soundly reasoned, coherently written conclusion.

Charles Frank Miller registration card, serial no. 619, order no. [none given], Draft Board 10, Multnomah County, Oregon; Records of the Oregon State Headquarters, 4/26/42/4th Registration, Act of 1940; Records of the Selective Service System, 1940-47; National Archives-Pacific Alaska Region, Seattle, Washington.

The 22 May 1887 date can be explained by viewing Frank's discharge and papers in his personal pension file and by placing that information into historical context.

Frank Miller's discharge states he enlisted on 8 June 1905. The image to the right is a copy of the original source. The person who filled out the discharge would have been working with Frank's enlistment papers, another original source. Frank would have been the informant for his enlistment papers, and he would have known what his birthday was.

The discharge also states he was 18-years-old. On 8 June 1905, he was a month short of his eighteenth birthday. In 1905, the minimum enlistment age was eighteen. Frank may have provided reliable information in normal circumstances; but when he enlisted, he apparently did not.

He was living a household with a step-mother and three younger step-siblings. He may have wanted to leave home and see the world. Newspaper reports of fighting in the Philippines may have been carried in the local Charlotte newspaper, and enlistment in the Army would have provided an “out” for a young man.


Note that his birthday is not included on the original form. Rather, someone has printed May-22-1887 on the discharge. This printing resembles samples of Frank's handwriting. Perhaps he wanted to remind himself of the date he had provided.

Frank's discharge

Frank applied for a pension increase in 1930.

The following “Declaration for Increase of Pension” is an original document with primary information and direct evidence about Frank’s birth date. It also has the 22 May 1887 date written on it. While the WWII draft card, the printing on the discharge paper, and the pension paper are all separate documents, they were created by the same person, and his motives need to be considered. The country was in the throes of the “Great Depression,” and every penny was vital to a family’s survival. A man would fib about a birth date if it meant he could feed his family.

“Declaration for Increase of Pension,” Charles F. Miller, MS, 1930 (Portland, Oregon; privately held by Connie (Miller) Lenzen, 2012.

What else could have been created at his birth?

When faced with conflicts in data, the conscientious researcher asks if there is one more source recorded close in time to the event. For a birth, did the local newspaper run a birth announcement?  A line-by-line search of the Charlotte Republican newspaper was conducted for July and August 1887.  There were no birth announcements.

Records at death?

The most unreliable sources for birth information are those records created at a person's death. However, they must be examined.


Frank's obituary, death certificate, and military tombstone provided birth dates. Frank's death certificate gives 14 July 1887 as his birth date.

The informant for the certificate was Tzilla Miller, his wife. While she was not a participant in his birth, she had known him for most of his life. Not only had they been married for almost fifty-four years, they grew up in the same town and attended the same school. Tzilla surely knew Frank's birth date. Yet, as shown by her diaries, she knew his birthday to be 9 July until the official record of birth arrived. After that, his birthday was celebrated on 14 July. She would have no reason to suspect any other date.

Frank's DC
The obituaries that Frank's wife cut out of the newspaper provided his birth date -- again 14 July 1887. The informant for these articles is unknown, but it is likely that Tzilla or another family member provided the information. franksobits

Frank's tombstone at Williamette National Cemetery in Portland provides a birth date—12 May 1887. This is the fourth date for his birth! One wonders where that date came from. Did he perhaps not remember the date that he provided when he enlisted? Perhaps 12 May 1887 is the day that he gave on his enlistment papers. A Form 180, Request Pertaining to Military Records was submitted to the National Archives in 2005, but no records were found.


Direct evidence was found to answer the question “When was Frank Miller born?” Four different dates were found: 12 May 1887, 22 May 1887, 9 July 1887, and 14 July 1887. Frank Miller himself provided the 22 May and 9 July dates; the “official” birth record provided 14 July. One person cannot be born on four different dates; only one is correct, but which one?

Frank provided the 22 May date. It was pencilled in on his military discharge and was used for his pension papers.

The 12 May date is on his military tombstone. That may be the date he provided when he enlisted at age seventeen rather than the required age eighteen. He may have forgotten that date when he applied for his pension and used the 22 May date.

Frank’s family celebrated 9 July as his birthday until he received the official record from the Eaton County Clerk. His immediate family most likely participated in his birth, and they would have had no reason to celebrate Frank’s birth on a day that was not correct.

The 14 July date in the county birth register was recorded a year after Frank’s birth, and it was probably copied from another source. That original source was probably handwritten, and handwriting can be misread. The number “9” can be mistaken for a “14” if the “1” appears to be a continuation of the upward swing of the cursive “y” in July. The state-level register shows distinct columns for month, day, and year, but if there were no columns, it would be easy to read the “y” and the “1” as one letter, and a “9” could look like a closed“4.”


The five steps of the GPS create the guide for our genealogical work even when all the evidence that we collect is direct. In this case, it provides convincing proof that Charles Frank Miller was born on 9 July 1887.

© 2012-2015

Connie Lenzen, CGSM

CG, Certified Genealogist is a service mark of the Board for Certification of Genealogists, used under license by board-certified genealogists after periodic evaluation, and the board name is registered in the US Patent & Trademark Office.



Connie's blog: