The Genealogical Proof Standard - International Institute

Introduction
For a long time, serious genealogists have sought to define a standard by which their deduced conclusions—about an ancestral identity, event or relationship—could be measured, to appear cogent and reasonable to other colleagues and the world at large. For the most part, we are talking here about drawing conclusions when we find contradictory or ambiguous information and/or only indirect evidence about the matter or question we want to address. So often our genealogical procedures involve ‘problem-solving’ and not straightforward primary information with direct evidence.

1. For example, say your question is: Who is the father of my ancestor Patrick Flynn? Or you may have a hypothesis: I think Thomas Flynn is Patrick’s father. You have found:


 * Oral family tradition says that Thomas Flynn had four sons: Jack, Patrick, Henry, Seth.
 * Patrick Flynn is in two 19th century census returns, in the household of Thomas and Lucy Flynn (no 'relationship’ columns).
 * Patrick named one of his own sons Thomas.
 * Wills for both Thomas Flynn and a Henry Flynn, who lived in the same general area and were approximately of the same generation, named a son Patrick.

We’ll come back to this genealogical question/problem below.

As the 20th century progressed, along with an explosion of genealogy-related books and Internet sites, so did the popularity of family history and genealogy for millions of people. Relevance for standards was initially found in ‘borrowing’ from academic style and legal criteria. In most classic 19th century books or articles on genealogical methods and problem-solving, you will find reference to the preponderance of evidence principle. It entered the vocabulary of genealogy as the nearest existing principle that could be applied to our genealogical research conclusions.

Preponderance of the evidence is a legal term used to decide issues in American civil courts. Canadians may be more familiar with the ‘balance of probability’ term—same concept. Applied to genealogical issues, it became widely used for ‘proving’ a genealogical conclusion; it only required that the greater weight of the evidence supported your conclusion.

2. In our case about Patrick Flynn, you have four pieces of information that connect him to a Thomas Flynn. Even though an original source (wills/estate files) reveals two men with a son Patrick, the greater weight of your evidence sofar suggests that he is Thomas’ son. But is this enough analysis of the information and the evidence? In fact, is this enough information?

It became apparent to genealogical educators that this principle was not quite strong enough or high enough in genealogical studies. On the other hand, the term used in criminal court cases, ‘proof beyond a reasonable doubt’ is too unequivocal for genealogical application. Perhaps only DNA evidence might be considered absolute proof—only if tissue from the deceased ancestor is included in the testing. Meanwhile we must use all the available sources and information while we are working on a problem, recognizing that future researchers may find new sources or techniques to change our conclusions.

Knowledgeable genealogists began to demand more from themselves—convincing arguments and clearly stated conclusions, including reasoned explanations of any conflicting evidence. Scholarly periodicals (as opposed to society newsletters or commercial magazines) began to encourage articles that displayed meticulous levels of research and presentation. In the last decade of the 20th century the Board for Certification of Genealogists (BCG) became a leader of the movement to describe a standard of research “proof,” among other standards related to genealogical studies, writing, and educating. The result of their joint efforts, from prominent North American genealogists, was the first edition of the BCG Genealogical Standards Manual in 2000.

3. If you conclude that Patrick was the son of Thomas Flynn from the mentioned four pieces of information, we asked: Is this enough analysis of the information and evidence? No, it is not, unless you have asked yourself the typical questions:


 * How reliable are names and relationships in oral family tradition (the first source on our list)? It is usually a derivative source (especially if the ancestors are several generations removed from the informant).
 * The census is an original source, but unless relationships to the head of household are explicitly stated, there could be different reasons for Patrick’s presence in Thomas’ household. Are Patrick’s alleged brothers also recorded at the same time? Are their ages reasonable for a family unit? Could Patrick be a grandson or a nephew of this couple?
 * Naming customs (a son after the father’s father) is one of the weakest pieces of information. Even with cultural groups that doted on such custom, there are always exceptions. It could be supportive only on the demonstration of much stronger evidence.
 * The search for a will/estate file introduced the possibility that Patrick’s father was a Henry Flynn. Can you show that Henry is not the father? What about Henry’s household in various census returns? What other sons, if any, are mentioned in both wills?

4. These kinds of analytical questions raise a doubt about the identity of Patrick’s father, even though the preponderance of evidence ‘looks good’. Therefore, to our second question, ‘Is this enough information?’ we have to answer that not all available sources have been examined. By extending the search to additional sources, not stopping at a ‘slightly greater weight’ of evidence, you will feel much more confident about your eventual conclusion.

After the next sections, we will return to Patrick’s problem.

“Evidence” and “Proof”
A word about these two words! Generally they have been used interchangeably. Lineage societies have traditionally used the word ‘proof’ meaning evidence in support of a certain conclusion. To some, the word ‘Proof’ denotes a mathematical or legal absolute that is not necessarily applicable to our genealogical arguments and conclusions.

Genealogy is more akin to a biological science where much variation occurs and a cluster of attributes is needed to constitute proof. As said before, we strive to do the best we can with the materials at our disposal at any given moment in time. We provide thoughtful examination of the gathered information and its evidence to deduce an identity, relationship, and so on. ‘The best we can’ means pushing ourselves to find every relevant source.

You can state your conclusion by saying something like, ‘I have proved Patrick was the son of Thomas Flynn because of the following ...’ or you can say, ‘The evidence shows that Patrick was the son of Thomas Flynn because ... .’

Genealogical Proof Standard (GPS)
The GPS is almost simplicity itself. It is a logical procedural series that good family historians and genealogists have always followed, even if subconsciously. The five steps are:


 * Making as wide a search as possible for sources that could help establish the identity, event or relationship under investigation.
 * Recording in proper, acceptable format the source citation and/or the provider of the information.
 * Analyzing and correlating the collectedinformation—evaluating the quality of sources and the reliability of information within them.
 * Resolving any conflicting, contradictory evidence with reasoned argument.
 * Stating your conclusion convincingly (more than a “balance of probability”).

5. Where are we with Patrick Flynn? We had four pieces of information that could be ‘argued’ that he is the son of Thomas (and Lucy), by the preponderance of evidence principle. But how far did the ‘as wide a search as possible’ go? Not far enough, if we can pose some additional, generalized (for purposes of our fictitious case) sources that did not appear in our first four sources:


 * Starting with the last event in his personal timeline, was a death certificate located for Patrick? Did it name his parents? Although information about his parents would be secondary, it is new information to add to the collection.
 * Similarly, a potential newspaper obituary might mention his parents, although the information is still secondary.
 * Patrick’s marriage record would be of primary interest. If it occurred at a time period when the names of parents were required, presumably he provided that information himself.
 * Searching for sources can move on to Thomas Flynn and Henry Flynn if Patrick’s father has not been satisfactorily resolved. How many other children are named in their wills, and how do they fit with what you know so far about Patrick?
 * Did the deaths of Thomas and Henry (dates presumably recorded within their estate files) generate newspaper obituaries that detail the names and locations of their children? If both had a son Patrick, which one corresponds to the location for ‘your’ Patrick?
 * What about Lucy, the wife of Thomas in census returns? Did she outlive her husband? Did she leave a probated will? Is there an obituary for her that identifies her children?
 * The same would apply to the wife of Henry Flynn, if necessary at this point.
 * Comparison of the two families/households in additional census returns could fit into the search, depending on the unfolding of results in your progress.
 * Derivative sources such as histories of the town, township, county (not to forget that local churches have also published anniversary histories) should not be ignored for what might be merely corroborative family information, or in difficult cases, give important clues that never appear elsewhere.
 * In a very difficult scenario, when none of the above sources are available or do not yield satisfying answers to the sum of the question/hypothesis, there may be the possibility of working from a different family angle. If you have convinced yourself that your Patrick had a brother called Seth (or some other brother), researching events and records pertaining to Seth may better demonstrate the names of his parents. Of course the brother relationship should also have solid backing.

Sample Proof Argument
Patrick Flynn (1845-1908) was the son of Thomas Flynn (1818-1881) of xxxxxxx town, township, county, province/state. The informant on Patrick’s death certificate named his father as Thomas, and this is supported by his appearance in 1851 and 1861 census returns as a youngster in Thomas’ household. Furthermore, the other children in the same census households are named in Thomas’ will along with Patrick. The names of his siblings also correspond with oral family stories. In addition, Patrick’s newspaper obituary mentions his surviving brothers; once again, the names agree.

This example, or case study, was to show the difference of attitude and perspective between preponderance of the evidence and the Genealogical Proof Standard.

Outlining a genealogical standard of evidence and for the written demonstration of a ‘proof argument’ or ‘proof summary’ has assisted the recognition of genealogy as a valuable and independent field of study in itself, not some tag-along history. The Genealogical Proof Standard stands on its own, without legal or academic crutches.

Research Planning
What we have been describing also involves the research planning you initiate to address a genealogical question or hypothesis. As the research progresses and results are obtained, the research plan to achieve your goal changes and adapts. Sometimes almost in the twinkling of an eye, the plan smoothly adjusts as you evaluate and analyze! You are learning what problem-solving is all about.

Serious researchers are well-advised to read the ‘Evidence Analysis’ chapter by Donn Devine, JD, CG, CGI, in Professional Genealogy, a Manual for Researchers, Writers, Editors, Lecturers, and Librarians.

_________________________________________________________________

Information in this Wiki page is excerpted from the online courses offered by The National Institute for Genealogical Studies. To learn more about these courses or other courses available from the Institute, see our website. We can be contacted at [mailto:wiki@genealogicalstudies.com wiki@genealogicalstudies.com]

We welcome updates and additions to this Wiki page.