Ungeneralizing Generalizations Always Helps

Generalizations and assumptions are necessary to sometimes give our research some focus or a point at which to start. But it is important to remember that generalizations (if they appear to be accurate) are not necessarily true all the time–that’s why they are called “generalizations” and not absolute truths.  Researchers can create headaches for themselves when they apply generalizations without giving any thought to how accurate the generalization may be or whether it applies in their specific situation.

One often encounters generalization veiled as “helpful hints.” These generalizations often state that something was always true about a certain group of people, a certain type of record, how children were named, who the godparents were, etc. I’ve also seen statements (without any substantiation) such as “eighty percent of households in the 19th century contained individuals other than parents and their children,” “no immigrant appeared in a city directory for the first five years they lived in the United States,” “witnesses on a naturalization were always the ‘sponsor’ for the immigrant,” etc.

Always take care when applying such generalizations to your own research. Take extreme care when the generalization has no source or study to support it. Even when a researcher says “based upon my experience” it’s true, one still has to be careful. Sometimes that research, which may be extensive, could be limited to a specific time period, geographic area, ethnic group, etc. Conclusions based upon narrow expertise may not be applicable to a larger group of people.

If you think the generalization may have some merit–then it’s simple:

do some research into the generalization

Sometimes there may be a legal reason that the generalization could be true. There may be academic studies that have been done that address the situation and may provide some more specific insight (along with some sources).

Early in my research, I made generalizations about German immigrants based upon my own research. Later I learned that those generalizations were generally true–about my immigrant ancestors from Ostfriesland. Those generalizations caused me to be slightly confused when I began researching my German immigrants from other areas of Germany. The fault was with me–not with my ancestors. My generalizations about Germans were the problem. They were not true about German immigrants in general. At that point in time my research and experience, while extensive, was fairly narrow. Sometimes we realize that and sometimes we do not. Now I know better.

Assumptions can be just as problematic. Assumptions are statements or concepts that we accept without some direct evidence to support them. Making an assumption about an ancestor because of the ethnic group of which they were a member, the time period in which they lived, the area in which they lived, etc. can also create research headaches. Sometimes we need to make assumptions to get our research started.

We just need to take ownership of our assumptions and dump them when the research suggests that they may not be true.

Lastly…

Researchers need to be wary of any statement that uses the word “always.” Many of us have ancestors who always lived in the periphery of social mores and expectations <grin>. That always makes them a little more difficult to research <double grin>.

FacebookTwitterGoogle+PinterestLinkedInShare

One thought on “Ungeneralizing Generalizations Always Helps

  1. I know exactly what you mean! In the last year I recently discovered a family about an hour away from me to the south. Until this discovery I thought the families were about 4 hours to the north. They are in both places and the information that I received puts all my family history together.what a wonderful surprise!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>