You can't always trust 'eyewitness' accounts

We'd all like to believe that the family stories we heard at Grandma's knee were sacred truths. None of us wants to believe that our sainted grandmothers would have lied to us.

But could they have embellished, skipped over some unpleasant details, or gotten confused by our constant quizzing? Could we have inadvertently tainted the interview by planting ideas in Grandma's head?

One of the distinctions we make in genealogy is the difference between original and derivative sources. Generally, we give more weight to original sources. The reason for this, of course, is because each time a record or an account of an event is retold, copied or summarized, errors inevitably creep into the product.

Grandma would be an original source if she were telling us about something she saw or experienced herself. Traditionally, we've convinced ourselves we could take her version of an event as the gospel truth.

But it's critical that genealogists keep in mind that the "original" label on something doesn't necessarily mean accurate or truthful. Even though an individual was present at an event, an eye witness to something that happened, can we trust what she believes she saw?

A recent ruling from the New Jersey Supreme Court can serve as a valuable training aid for all genealogists. The case tells us that we must analyze a number of things before accepting an eyewitness account as accurate.

The court said that memory isn't as reliable as initially believed, noting that stress, focusing on a weapon rather than a face during the commission of a crime, distance, lighting, intoxication and racial bias can affect the accuracy of an identification.

We probably aren't going to be asking Grandma about some criminal activity she witnessed, but we'll likely be seeking her memories of a specific family member or event.

So a good genealogist can tweak the court's advice and apply it to a family scenario. Study the question you're asking Grandma. Are you asking her to remember something that causes her stress now or would have at the time she experienced it? Did she play an active role in the event, or did she view it from the sidelines? How long ago did this incident occur, and how old was she at the time? Does Grandma generally have a good memory? Is she alert and well grounded, or does she have dementia? We must ask ourselves as many questions as we ask her. All this is necessary in order to assess the truthfulness or accuracy of her statements.

The Jersey court's decision came as no surprise to many legal and psychological experts. Hundreds of studies on eye witnesses over the past decade have shown that recall of events can be affected by time and other outside factors.

That is an important point for genealogists to study: it means that Grandma's memories likely have dimmed as she aged, and if we aren't skilled in our interview techniques, her "memories" could be mere reactions to suggestions from us.

The New Jersey court ruling resulted from a 2004 case, The State vs. Henderson, in which an eyewitness admitted a detective "nudged" him to identify the defendant. Studies have shown that police officers give inadvertent cues to witnesses with mere nods or subtle encouragement.

But before we condemn law enforcement, stop and think how many times we family historians nudged or encouraged Grandma. In the process of our "interrogation" did we "suggest" an answer to her? Perhaps she had a hard time remembering a date, or even the name of a cousin or other relative. Questions such as "Did this happen in 1948?" or "Did Uncle George tell you that?" suggest an answer. So if we later write in our family history that Grandma confirmed the event occurred in 1948, we're really misleading our readers and ourselves.

Like the good criminal investigator, we must ask open-ended questions that don't suggest an answer. Only then can we be sure we haven't tainted the information Grandma gives us.

Elizabeth Loftus, a University of Washington psychologist who has studied memory issues for years, wrote that "Misinformation has the potential for invading our memories when we talk to other people, when we are suggestively interrogated or when we read or view media coverage about some event that we may have experienced ourselves."

Writing in the Stanford Journal of Legal Studies, psychology professor Barbara Tversky and law professor George Fisher discussed "original memory" and explained that interpretation occurs at the formation of memory — so it is distorted from the beginning. They pointed out that "rarely do we tell a story or recount events without a purpose . . . and that every act of telling or retelling is tailored to a particular listener."

So what is a poor, beleaguered but conscientious genealogist to do? We must apply another cardinal — but often ignored — rule of genealogy: Make an exhaustive research of every "fact" presented in written family histories.

That means we take the burden of proof off poor Grandma's shoulders. We keep in mind that nothing is proven by a single source, no matter how official or reliable the source appears to be. We look in every nook and cranny to find bits and pieces that we can link together to prove or disprove what Grandma told us.

If Grandma embellished or got confused, bless her heart. If we purport that her word alone proves something, shame on us.