Volume 5, Number 11 - Spring 1976

Your Thousand Names
by Dr. Lyle Owen

A person is obviously just as much descended from his or her mother, and the mother of that mother, as from father and father’s father. But considerable family history research has been confined to "the name line" largely just to make it more manageable in size. Anyone who has tried to search out all his ancestors impartially knows how overwhelming the task soon seems. The forebears multiply, and the surnames increase almost as fast.

It is of course only a traditional and legal reason that causes us to bear our father’s surname, rather than our mother’s; likely this custom in origin was one more aspect of the famous male dominance. In some cultures the naming scheme is otherwise,


and even here the women’s lib rumblings portend questionings to come.

On both maternal and paternal sides, as one investigates back, the added names revealed, and the vistas they open up to the researcher of new family branches (with all the human interest implied) intrigue greatly. When one runs across a new surname, ancestral to the researcher but unknown before (the birth name of the woman a known male ancestor married, perhaps), there is generally a feeling of pleasant surprise at the new knowledge gained, and a thought like "Why, I know people with that name; I wonder if we are related?"

And occasionally one runs across, a second time, a surname already known to be ancestral --such as a second Smith or Brown, but not known to be related to the first one. Smith, the most common family name in the United States (borne by about 1 Out of each 100 people) is on a probability basis the most likely candidate for finding, then again finding (though I haven’t for myself discovered a second Smith ancestral line -- yet!) I’ve sometimes thought that everybody must have Smith ancestry if only he knew all his family names. And the thought always occurs to the researcher:

Are my 2 Smiths (or whatever) related? No evidence of it -- but perhaps? This is a part of the adventure of genealogical research, and one always hopes that someday he will find out their connection if any.

As an example, my own children have Douglas as such a double-shot ancestral surname (the 2 not known to be connected), pretty far back on their father’s side but only 2 generations from them in the mother line. This double occurrence is in spite of the fact that according to the Social Security Administration’s enormous list of surnames (the biggest list of convenient access), Douglas is only 228th in frequency of surname occurrence in the United States (to compare with Smith as 1 and Johnson as 2).

Such occasional reoccurrence, in one’s complex total ancestry, of the same surname reduces the total of his ancestral names somewhat, though not very fast. The farther back one can go, of course, the more probable such a number-reducing result. But the number of different family names used in the United States is almost unbelievable. In 1964 the Social Security Administration published its Report of Distribution of Surnames in the Social Security Account Number File, with about 166 million records on hand then. These fell under 1,091,522 differently spelled surnames -

really an astonishing number of names that someone has, or has had, since Social Security records started in the 1930s. (These records include many individuals now deceased, and more than one name for many who changed name, most commonly by marriage.) Relating the 2 figures shows an average of only about 150 uses per surname, which seems surprisingly small, but very many of the names, with their exact spelling, are exceedingly rare. For instance (and believe it or not!) a few people have family names consisting of a single letter. These enormous Social Security records, as of 1964, show at least 1 person having such a I-letter surname for every letter in the alphabet except Q.

The most common surnames in the Social Security records, starting with the famous Smith as number 1, are listed in the report -- all those having 10,000 or more occurrences in those total records to 1964. This list of most common names, so defined, runs to 2,183 different surnames. Really this is a rather surprisingly small number of surnames, this list of the most common, in view of more than a million different family names being used in this country.

By way of a curious contrast, Charles Berlitz in the summer 1972 Horizon magazine writes that there are not more than a hundred Chinese family names, though they get variously spelled when done into English, like Wang or Wong, Wu or Woo. And elsewhere one reads that the most common family name in the world is Wong, in such great numbers in China as to put to shame our Smiths, of whom there are about 2 million in the United States. Even in the United States, Wong is common enough to be 881 in the Social Security list of surname frequency -- between Bird, 880th, and Lange, 882nd. (Byrd is 296 and Lang 429, but Wang in not in this "most common" list.)

Individuals bearing the 2,183 "most common surnames" in the United States comprise just a little over one-half of our people, judging from the vast Social Security Sample. Of the nearly 166 million individual records to 1964, something over 87 million (or 53%) had surnames on the "most common" list.

As an additional small check as to whether this roughly half-and-half division between "most common" and less common surnames holds for one’s own family names (those found in the whole inverted pyramid of his ancestors), I took a 3-generation sample of the nearest ones to me (my 14 parents, grandparents, great-grandparents), and


found them a neat half and half. One guesses that this will hold generally true for the reader, if he cares to make a similar check against the Social Security name list citied above. With of course no assumption that anyone else is interested in my family names, but to make the example concrete enough to be helpful, these 4 out of the 8 birth names of my 8 great-grandparents are in the "most common" list, here given in the order of their frequency in this country: Smith, Thompson, Owen, Hannah. The other 4, here listed alphabetically because the Social Security report does not give frequency of occurrence for such less common names, will illustrate examples of names not on the "most common" list: Judkins, Losey, McKinney, Tibbets. Probably, and especially if the reader’s folks have been in this country enough generations for the melting pot to work, he also will find a roughly half-and-half division in his ancestral names between widely-used and not-often-used surnames.

The number of ancestral names one has is really amazing. As we go back, the number of names doubles each generation as does the number of ancestors. Name doubling assumes that there is no name duplication like that of the Douglas example given above. But while the number of names doubles each generation backward, only half of them are new names, these being the maiden surnames of the wives.

Thus the number of surnames accumulated by research into the whole inverted pyramid of one’s ancestry is the total number of individuals in the farthest generation one gets back to -- for example, 128 forebears and surnames in the 7th generation back, the time of the American Revolution. If there are no name duplications (like 2 Smiths or any other repeated surname) in that generation, one has uncovered 127 different ancestral names in addition to one’s own present surname - 128 ancestral surnames in total! The 7 generations take us back about 200 years, 30 years or a little less being a generation in the present sense of age overlap of mother and daughter, or father and son.

At 10 generations back, or around 300 years, the number of your ancestors in that single 10th prior generation passes the thousand line. You are descended from 1,024 men and women (half of each) who lived in that generation, and each of them may have had a different surname at birth -or 1,024 names in total, all equally ancestral to you, and 1 of which carried down to become your own last name. To get back to Mayflower and early Boston days, 1620-1630 or thereabouts, as some of us can do for maybe 1 or 2 of our many lines, extends the time reach to about 350 years, or around 12 generations. Here the number of one’s ancestors becomes really astonishing, in view of the inevitable doubling each generation if no intermarriage of kin (like cousins, near or distant) occurred. Starting with 2 parents, then 4 grandparents, and on back 10 more generations, yields 4,096 individuals (half being wives) in that 12th generation, of about Mayflower times, that you are descended from. And the number of names, including maiden names bf the wives, is the same unmanageable 4,096, unless there was a little reduction by name duplication as stated.

One is equally descended from all these people, and if one could do it, each of these lines would be most interesting to explore. What a pile of research results that would make in one’s library!


This volume: Next Article | Table of Contents | Other Issues

Other Volumes | Keyword Search | White River Valley Quarterly Home | Local History Home

Copyright © White River Valley Historical Quarterly

 Springfield-Greene County Library