Genealogical Methods/Verification and Validation

Return to The Tapestry Homepage
Return to The Tapestry Homepage
Return to Methods
Return to Methods
This is one of a series of articles on Genealogical Methods, prepared in association with The Tapestry. See Index for a list of related articles.

Verification and Validation

Reference Verification and Validation (otherwise known as "V&V") is an important, but often overlooked component of genealogy. Both processes deal with references cited by an author concerning his sources of information. In reference verification the only objective is to make sure that the informaiton attributed to a given source, is actually found in that source. For example, someone might cite the National Aeronautic and Space Administration (NASA) as a source of the statement "The moon is made of green cheese."

Verification asks the question "Did NASA say 'the moon is made of green cheese?' "
Validation, asks the question "Is it true that the moon is made of green cheese?"

Neither question is trivial, though the significance of reference verification is often not understood.

It sometimes happens that information is inaccurately attributed to a certain source. Sometimes the 'fact' is misunderstood, or otherwise misrepresented. Sometimes the cited source does not present the 'fact' at all. People can and do make mistakes, and when you rely on other sources for information used in your own work, it is important to be sure that those sources say what you are being told they say. You do that through reference verification.

When someone tells us that Fleming, 1971:363 says that "Major John Cowan...was killed by Indians on the Clinch River between 1778 and 1780" it is important that that we check to see that he really does make that statement. People make mistakes, and sometimes cite sources that do not actually contain the information they are said to contain. In reference verfication we attempt to assure ourselves that a source does contain the information it is said to contain. In the present case, we need to get a copy of "The Cowans from County Down", and check to see if there really is such a statement on page 363. [Such a statemet is in fact made on that page, but don't take my word for it, verify this point for yourself. That's what I did when I first ran into this statement.]

Once you've assured yourself that Fleming, 1971 really did make this statement, you can move on to the next step in the V&V process, Validation. Here the question isn't "Was this statement really made", but rather "Is this statement true?" When we seek to validate Fleming's statement the first thing we'd do would be to check to see what he tells us about the basis for the conclusion. Perhaps this statement was based on a contemporary account of the killing of John Cowan, or a perhaps on a descendants recollection.

Either type of source would be helpful in understanding the basis for the statement, and Fleming uses both kinds of evidence in his work. In this particular case, however, he uses neither. Instead, he bases the statement on a family history by "Mrs. Dunavant", a genealogist active in the 1930's whose work he specifically quotes and cites, and indeed seems to be the basis for his insight into the family history. Up front, the presumption is that Mrs. Dunavant "did good work". But irrespective of her qualifications as a genealogist, Mrs. Dunavant was working in the 1930's, and did not know of her own personal knowledge, what happened to Major John Cowan. She presumably learned his fate from a particular source, and presumably that source (or chain of sources) was ultimately grounded in a contemporary account of the killing of John Cowan. Perhaps that account was oral, and passed down by word of mouth. Perhaps it was written down as a letter by a contemporary of the events. We'd like to know exactly what that statement is based on, because the nature of her sources would greatly influence our acceptance of the conclusions of Mrs. Dunavant and echoed by Fleming.

Unfortunately, the excerpts from Mrs. Dunavants work do not include an explanation of "how she knew what she knew". This means we have another problem that has to be researched. While we've verified that Fleming does state that "Major John Cowan...was killed by Indians....", we still don't know that this is a true statement. To validate the statement we now need a) to verify that Mrs. Dunvant (whom Fleming cites as his source) does indeed make this statement, and b) see what she uses as the basis for making it. Perhaps she had specific historical records that conclusively showed that Major John was killed, and wife Mary Walker was captured by Indians. Perhaps she came across an account left by Mary Walker. Contemporaries of Mary, in similar circumstances did indeed leave such records of their experiences (see Margaret Handley's Captivity Account). Or perhaps she simply says "this is what happened" without ever offering any basis for her statements. If the former is the case, then we may want to pursue whatever documents she cites to first verify and then validate the information she provides. Perhaps if her treatment is extensive enough we may not need to track the actual documents; but probably we will want to go through further V&V with her work. But it may also be that she just makes a statement "John Cowan was killed...", and says absolutely nothing about the basis for this statement. We then run into the problem of deciding if she is sufficiently credible to warrant accepting her information even provisionally.

And accepting something like this, even provisionally, when no explanation is given as to what its based on, is probably not a wise choice. In cases such as this, there's no way to distinguish between a work that has underlying (but unspecified) supporting information, and a work that includes a made-up story. We'd like to think that no one would simply make things up whole-cloth---but the unfortunate truth is people do misunderstand what they find, and errors creep in, and the result is indistinguishable in some cases from something made-up. The best one can do in those circumstances is consider the story suspect.

It is unfortunate that in many cases genealogists fail to indicate the sources of their information. There seems to be a preference for simply saying "this is so, take my word for it". Even when an attempt is made at giving sources, the sources cited are often described only as "various records". Statements like this do little good for others, as it is impossible to know what specific records were meant. Court records? Land records? Family bibles? Which court, land, and bible records are meant? It's impossible to say what the authors of such statements had in mind, though at least we know they had something to go on. We just don't know what. And because we don't know what records were used, we can't see for ourselves that they said what someone said they did. We can neither verify the reference, nor validate it. Ultimately, it is our ability, or lack thereof, to verify and validate records and statements that governs whether they are useful or not.

That last statement may be a bit of an overreach, as there are ways of validateing statements even when we don't know what they are based on. For example, we can conduct our own search of available primary sources and see if we can find something that could be used to confirm whatever statement we are checking. It might not be the same source of information used the original author, but if we can find a primary source that confirms what they said, then we can consider their statement as validated. Alternatively, you may find a primary source that refutes the statements being considered. Either way, you can reach a conclusion on the validity of the statement. Perhaps the worst outcome would be one in which you can find nothing to confirm or deny the statement. Without the critical knowledge of what the statement is based on you are left unable to reach a conclusion as to the statements validity. Might be true, might not. What you do then is a matter of personal judgement, but if a diligent and comprehensive search for supporting evidence turns up nothing, then there is probably a bias to consider the statement as unfounded (that is, not validated).

Finally, the validation process is similar to the process of "proving" something. However, in genealogy, they are not the same thing. In general the standards of proof for genealogy are much more rigorous than what is required for validating something. For a proof, the supporting evidence has to be sufficiently vigorous that a statement can be broadly accepted as conclusive. For validation, the evidence need not be conclusive. It need rise only to the level where a reasonable person would consider the statement likely to be true. A discussion of the standards of proof is found Standards of Proof.