Evaluer la fiabilité d'une source

Evaluer la fiabilité

Barbara Krasner-Khait demande à Elizabeth Shown Mills quelles sont les meilleures pratiques pour effectuer des recherches de grande fiabilité (qualité).

Dans la période actuelle, où les généalogistes peuvent avoir accès à une somme incroyable d'enregistrements à travers des arbres familiaux disponibles sur le web et sur des supports électroniques, comment saurez-vous que l'information à laquelle vous aurez accès aura les exigences de fiabilité imposées par la recherche généalogique? Avez-vous pris le temps lors de vos propres recherches pour citer vos sources? Etes-vous déjà convaincu de vos informations sans vous être documenté pour savoir comment et pourquoi vous avez tiré telle conclusion ?

Depuis plus de 20 ans, Elizabeth Shown Mills a formé des familles d'historiens sur les techniques permettant d'obtenir de la documentation fiable. Son premier article sur la fiabilité des informations, intitulé "Comment documenter correctement vos notes de recherche" (“How to Properly Document Your Research Notes”), a été publié dans le numéro de Septembre/Octobre 1979 de Everton’s Genealogical Helper. Avec l'aide de son mari, elle a travaillé sur des modèles pour évaluer la fiabilité des informations, en établissant 13 indicateurs permettant d'analyser la fiabilité ainsi que 13 autres indicateurs pour la documentation, qu'elle a publié dans Evidence! Citation and Analysis for Family Historian (Genealogical Publishing Co., 1997).

Mills dit que, “Si nous ne savons pas quelle est l'origine d'une information, nous ne connaissons pas sa valeur. Nous n'avons aucun moyen de savoir ce qu'il faut croire ou comment évaluer la fiabilité d'une information par rapport à une autre.”

Définir des indices

Les indices sont les informations que nous cherchons à utiliser afin de tirer des conclusions sur nos recherches. Nous commençons avec des données “brutes”, nous les analysons pour déterminer si elles correspondent à la réalité des faits, et en quoi elles concernent la personne sur laquelle nous effectuons des recherches. Ce processus est celui qui conduit à la preuve — la meilleure information que nous ayons et qui nous conduit à tirer un certain nombre de conclusions, tandis que nous sommes conduits à écarter d'autres types d'information. Tom Jones, qui est un généalogiste professionnel de Virginie et également président du bureau de certification pour ls généalogistes (Board of Certification for Genealogists), dit que “En règle générale, les chercheurs commencent avec une question, par exemple, qui était la mère de John Smith, ou bien où est né Jane Johnson. Ils recherchent des sources et des indices qui peuvent apporter un éclairage sur la question, qui permettent d'en évaluer la fiabilité et, si les découvertes sont en accord avec la Preuve Généalogique — en viennent à une conclusion sur la réponse à leur question.”

Attacher chacun des détails à une source

Chaque morceau d'information doit porter son propre poids. Selon Mills, le critère le plus important est celui qu'elle cite en premier dans le “Guidelines for Documentation”: “Tout relevé ou fait qui n'est pas de notoriété publique doit être accompagné de son propre relevé de source.” Elle ajoute, “Les détails les plus infimes sont essentiels pour nos besoins.”

Les sources peuvent être d'origine ou non. Et c'est cette dernière catégorie qui est problématique. Si une source non originale contient une erreur, cette erreur peut être amplifiée à chaque fois que la source est copiée. Vérifieux rigoureusement les citations afin de déterminer le degré de fiabilité de votre information par rapport à la source originale.

Information can be either primary or secondary. Primary information represents first-hand knowledge of the recorder or the informant. Secondary information is everything else. Let’s say you have an ancestor’s death record. This can be considered a primary source on the physician, because he provided the information himself. The death record is also a secondary source for information on the decedent’s parentage and birth information, since the informant was naturally someone other than the person who died.

Not all sources are created equal. Jones warns, “Sources can be right or wrong and the evidence gleaned from them can be of excellent or poor quality. Analysis of the evidence helps the genealogist determine whether the source is correct or not. Analysis should be very careful, because sources of apparent high quality can be wrong or vice versa.” When you thoroughly analyze the evidence and place it in the context of all the available evidence you have regarding the particular research question you’re trying to answer, you can confidently state your conclusion.

Assessing An Ancestor’s Identity

Three situations may arise in evaluating an ancestor’s identity: Multiple sources agree — All available evidence from sources such as death records and marriage records agree, such as indicating an ancestor’s parentage. Fortunately, this situation applies to most cases.

Evidence conflicts — Evidence from different sources do not agree, making the case more complicated and needing careful evaluation. Says Jones, “I found evidence of four different men as having been the father of my ancestor, Amzi Leach. Evaluating the evidence helped me conclude which of the men was the actual father.”

No direct source directly answers the question — A thorough search yields no source directly identifying an ancestor or ancestral fact. You can correlate all available evidence and draw a conclusion that meets the Genealogical Proof Standard. Jones says, “This was the case with identifying the parents of a client’s ancestor. While no record identifies the ancestor’s father directly, a combination of marriage, probate, and tax records in combination with an undocumented publication that by itself wouldn’t carry much weight, yielded an incontrovertible conclusion identifying the father.”

As you trace your ancestry, it’s likely you’ll run into the second and third situation at one time or another. I’ve run into the “Evidence conflicts” situation myself in the case of Chaim Ber Dvorkin and his relationship to my great-grandmother, Breina Dvorkin Krasner. Chaim Ber’s parents were Elias Meyer and Chaya according to his 1915 death certificate. Breina’s parents were Hillel and Michla according to her 1937 death certificate. Were Elias Meyer and Hillel Meyer the same person? I’ve evaluated naming traditions (following the eastern European practice of naming after the deceased) and tax records (no vital records exist for their ancestral home in Belarus) and came to the conclusion that Elias Meyer and Hillel Meyer are indeed the same person and father to Chaim Ber and Breina. I have not found enough evidence to draw any conclusions about the connection between mothers Chaya and Michla.

Negative Evidence

“As Sherlock Holmes famously pointed out, the key to a case can be the fact that a dog doesn’t bark,” wrote Mills in her June 2000 NGS Quarterly editorial. We have to know what we didn’t find but maybe should have. Donn Devine, cg, cgi of Wilmington, Delaware adds, “When you have negative evidence, you have to account for it. All too often, evidence conflicts. You need to weigh and evaluate the information. That’s a process that calls for experience.”

One example comes from Jones. He couldn’t find any Overtons in Spotsylvania Country land tax lists, even though deed records showed they owned land. His application of negative evidence resulted in: (1) the conclusion that the surname had evolved recently from Howerton; (2) the organization of seemingly unrelated individuals into family groups; and (3) the identification of earlier generations of the family.

Importance of Citation

I’ve always believed that citation should allow someone else to replicate your research. I’ve relied on this many times to check out some detail in my original sources. Citation also helps to organize what evidence we’ve accumulated. Says Jones, “With many sources bearing on every genealogical question, citation helps the researcher to ‘keep them straight,’ that is, to maintain a record of where each item of information and evidence comes from.”

Source citation helps evaluate the quality of the information used. For instance, a source created in the last few years about a genealogical fact from the 1820s would not seem to have much credibility.

Detailed citation can lead us to additional information. This was the case when a citation led Jones to uncatalogued manuscript material at the Cincinnati Historical Society, far from where his ancestors lived. It held a key to identifying one of his ancestors and even included letters another ancestor had written.

Question, Question, Question

“Don’t be afraid to question everything,” says Mills. “If no source is given, it’s obvious something’s wrong. If it doesn’t seem to hang together, find out why.” It’s all too easy to accept information as a given, without checking out the sources. It’s also helpful to learn the environmental context around your ancestors — migrations, neighbors, economic situation, etc. The amount of available secondary information has skyrocketed with the explosion of the Internet. And a lot of it has no source information. Devine says, “You have to dig deeper and replicate what someone else did. Sometimes you don’t even know where to start. If images are involved, you don’t know it they’ve been altered with a photoediting program.” Adds Jones, “If the Internet is to fulfill its potential as a valuable resource for genealogists, individuals and businesses posting information for genealogists need to provide clear and detailed source citations. Only with such citations will consumers know whether the information is useful or whether it is bunk.”

Put It On Paper

“It’s always good to put your reasoning on paper, especially if you’re dealing with a complicated problem,” says Devine. In addition, he often shares his thinking with fellow genealogists. “Sometimes you can convince yourself, but others say otherwise.” It’s also a good idea to photocopy information whenever possible and cite the source directly on it, so if it’s copied again, the source is there.

Devine also recommends naming the source first, before beginning to write down the information. “Otherwise it’s easy to overlook recording the citation,” he says, “and it’s also the natural way to attribute information: ‘Aunt Minnie told me…’ rather than ‘The foregoing information was provided by Aunt Minnie.’”

Jones agrees and says, “Be extremely careful to note in great detail the source for every scrap of information concerning your ancestors. While it may seem time-consuming to do this, it will save you time in the long run. The detailed source citations will enable you to evaluate and re-evaluate your evidence and come to accurate conclusions without having to search for the sources again and again.”

Review Your Information

I was examining my grandfather’s ship passenger record one day and noticed another passenger on the manifest from the same town. This passenger bore a surname that seemed to be linked to mine. As you gain experience, you can — and should — go back to your sources. You may make additional observations that can aid your research. And citations can make it easier to make the most of the evidence you have. Says Jones, “Typically with thorny problems, genealogists may have to go over the same ground more than once. The citations make it easier to do that.”

So take the time now to cite your sources and continue to do so as you research. Document your conclusions according to the Genealogical Proof Standard. The methodology will help you make the most of your research, while allowing other researchers to understand and replicate your work.

The Genealogical Proof Standard: A Five-Step Process

1. We conduct a reasonably exhaustive search 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 assess its quality as evidence;

4. We resolve any conflicts caused by items of evidence that contradict each other or are contrary to a proposed (hypothetically) solution to the question; and

5. We arrive at a soundly reasoned, coherently written conclusion.

Source: The BCG Standards Manual, Millennium Edition (Ancestry, 2000)

This article originally appeared in the January/February 2001 issue of Family Chronicle. Last Updated ( Friday, 11 November 2005 )