Ancestry Magazine

Experienced genealogists place great emphasis on confirming every fact or conclusion from original records, the cornerstones of family history research.

Experienced genealogists place great emphasis on confirming every fact or conclusion from original records, the cornerstones of family history research. Original records are sometimes called primary sources, a tribute to their importance.

Why, you may ask, is it necessary to look at an original record when the same information has already been published in an easy-to-read book with a complete every-name index? The answer lies in the confidence level. Each time information passes from one individual to another, or is transcribed from one document to another, there’s a chance of error in interpreting the source or in producing the output.

This is strikingly illustrated by a demonstration often used in classrooms and sometimes played as a party game: Eight or ten people form a line. The first person writes a one-sentence statement and reads it aloud to the second person out of hearing of the others in line. Each person in turn, without the benefit of the written version, then repeats the statement as heard to the next person in line. The last person writes the message heard and then compares it with the original statement as written by the first person.

The differences are always significant, and occasionally hilarious. The demonstration shows very effectively how information is lost or corrupted by human transmission. Similarly, information recorded long after an event may suffer from fading memories or changed perspectives over time.

Original records contain information for some specific event or theory that was recorded directly from a knowledgeable and unbiased source, at or near the time in question, with intent to preserve an accurate record of the matter.

All other records and published works are derivative-derived or compiled from original records or other derivative records, or set down from memory sometime after an event took place. Sometimes derivative records are called secondary sources, and many invaluable genealogical references fall into this category, among them indexes, abstracts, directories, family histories, published transcripts of records, and, more recently, computer databases. Used properly, they are extremely important research tools. Derivative works are usually found in libraries, while original records are located in archives or in manuscript collections of historical societies or major libraries.

Microfilm copies of millions of original records and derivative works can be used at the Family History Library in Salt Lake City, Utah, or the Family History Centers operated worldwide by local congregations of The Church of Jesus Christ of Latter-day Saints. Centers are usually listed in local telephone directories under “Church of Jesus Christ of Latter-day Saints.”

Records that are original to one event may include derivative information about another event. For instance, a death certificate is clearly an original record for the facts about the death-time, place and cause of death-that are entered by the physician. It’s only a derivative record, however, on the matters of age, date and place of birth, names of the parents, or their nativity. As the accompanying examples show, information may have been guessed at by an uninformed acquaintance or confused by a grieving relative.

U.S. census schedules provide another example of records that are partly original and partly derived. To identify people living at an address on the date of the census, census schedules meet the definition of an original record. Most other information, however, including the age and sex of the individuals, was recalled from past memory, or possibly estimated by a non-family member with limited knowledge of the facts; therefore, for these purposes, the record is derivative or secondary.

Strictly speaking, even some census schedules that were microfilmed, and that are so widely used to great advantage by genealogists, are not original records. They were one of the two “fair copies” (in the sense of “neat and presentable,” rather than “passable”) made in the office by each enumerator from the original made in the field. The enumerators made the copies personally from the original sheets in their own handwriting; nevertheless, each recopying introduced a new chance of error. One copy went to the federal government and was filmed; the other was for the state, and many have been preserved in state archives or historical societies and can be consulted when the microfilm copies are unreadable. The original census schedules remained with the county; some counties still preserve them or have transferred them to state archives.

All other things being equal, original records are more likely to represent the actual state of affairs than ones made long after an event, or from repeated transmission or copying of the information. That’s why the originals, or photocopies of them, should be sought to confirm each link in your genealogy. By no means does that imply th at secondary works are not essential to good family history research. They are the starting point in the research cycle described by Val D. Greenwood. The researcher begins with a study of derivative or secondary sources, including family recollections, to learn about research already performed, conclusions currently accepted, and records that need to be consulted. Next comes an analysis of the problem and the formulation of hypotheses or theories of what the facts might be. Then specific research objectives are selected-for example, “Determine whether Harold Hays of Horsham was the father of Henry Hays of Huntington”-and a plan is developed for the order in which appropriate original records should be consulted.

All the relevant data is gathered and recorded, and then evaluated or weighted for its bearing on the research problem. There are only three possible outcomes:

“Yes, the proposition is true.”

“No, the proposition is untrue.”

“There’s not yet enough evidence to be convincing either way, and more research is needed.”

The cycle then begins again, adding newly proved or disproved propositions to the knowledge base, and starting again with more library research if necessary, or proceeding directly to a new problem analysis.

In weighing the evidence, agreement of two or more independent original records-for example, a birth registration and a baptismal record-is the most convincing. Often, the same data in two or more derivative works does not add weight because it may have ultimately been derived from a single original record.

Sometimes an original record has been lost, but agreement among several records derived from it gives assurance that they accurately reflect the lost original. For example, if family Bibles in different branches of a family, all published about the same time, all agree on the early generations before the books were published, it’s likely that all were correctly copied from a now-lost original.

Genealo gists use “preponderance of the evidence” to describe the weight needed to prove a genealogical conclusion, but they actually require more than-as the legal term implies-just enough to tip the scales in favor of the conclusion.

The confidence level we want is reasonable certainty, not mere probability, and the weight of evidence required is, to use another legal term, “clear and convincing”-not quite as high as “beyond a reasonable doubt,” but still reflecting a high level of confidence-the same level people expect in knowing who their own parents and siblings are.

Original records are the research cornerstones that can provide you with that level of confidence in your own linked genealogy charts and computer databases.

For Further Reading:
Val D. Greenwood, Researcher’s Guide to American Genealogy, 2nd edition, Baltimore, 1990.

Richard S. Lackey, Cite Your Sources, New Orleans, 1980.

Donn Devine, CG, CGI is a genealogical consultant, a practicing attorney, and the archivist for the Catholic Diocese of Wilmington, Deleware. He is active in a number of genealogical organizations.