I'm sure you've all heard the expression "Take with a grain of salt". While some sources go into serious detail on the origins of this expression, the widely accepted definition is
take something with a grain of salt
to consider something to be not completely true or right - ie I've read the article, which I take with a grain of salt.
My experience with genealogy research is that one should take in information with a
POUND of salt.
While parts of this diary may duplicate some of the information contained in Andrew C. White's excellent diary, my focus is mainly on Black research in the South (though a lot of the issues are applicable to everyone).
Note: This diary began to take on a life of its own once I started writing and is ending up longer than I intended. Because of that, the end is rather abrupt. If there is interest, I will continue on in a future diary.
First - A brief Genealogy Researcher CV :)
I've been researching my family history for about 15 years which includes a couple of significant breaks. I've used the Internet (Ancestry, FamilySearch, USGenweb, Rootsweb, various state/county vital stats databases, etc.), visited the National Archives in DC and Regional Archive in California, State archives in MS, GA, & Al, a number of County Circuit and Chancery courts, cemeteries, LDS FHL, called up strangers and basically have used anything and everything I can think of to fill in the blanks. I'm skeptical of anything where I have not viewed the source and I prefer to have one or two corroborations before I'm comfortable accepting something as fact.
As I read over what I've written, I see it's not so "Brief".
Before I descend into negativity, let me preface by saying I LOVE GENEALOGY!! The absolute thrill when a theory is proven correct or you uncover a crucial link to an earlier generation has got to rival the highest of highs. It inspires you to keep churning that microfilm reader handle until your arm is about to drop off or endure having all your belongings examined like you are a criminal just to access the archives building. On the other hand, because I've been researching so long, I've learned not to assume ANYTHING. One possibility is that my ancestors were some of the lying-ist, inconsistent, illiterate, moral degenerates EVER and my experiences won't apply to anyone else OR it's best not to take everything at face value.
Census
First up, census enumerators can't spell and their penmanship is often horrific. Get used to it, don't question it, it's a fact!!
I've seen new researchers say something to the effect of "oh, but my family spelled Smyth with a Y". Those thoughts should be expunged from your mind. Anyway there is to mess up a name, you will see it on the census. For some common name variations like Peterson/Petersen, Phillips/Philips, or even Gipson/Gibson, Soundex will still capture the name in a search. For other cases like Michael(M240)/Michaels(M242) or Wraight(W623)/Raitt(R300) you need to expand your thinking. Also, in this error era of online research, mistakes or poor legibility is codified. You may be looking for your "Tasker" ancestor and overlook the "Lesker" mistake. And Nicknames....Need I say more?
Surnames of former slaves in the 1870 census may have no relation to who last held them in bondage.
Many people interested in tracing former slaves believe the last name of their ancestor was the same as the family that held them in bondage. To me, this seems to be more the exception than the rule. Once emancipated, former slaves may have chosen a name from a prior owner or a parent's prior owner. They may have chosen the name of someone they respected or a name like Freeman that describes their new condition. You could also find that in the 1870 census they had one surname but in 1880, they were using a different one. This could have been an enumerator's error or they might have decided to start using or go back to a different surname.
Last names of children may be different than what is recorded in the census or that either the listed mother or father may not be the biological parent of the child/children
I don't know if this hold true in less rural areas or with people of means, but the names of children may differ from what is written in the census. This could occur in a couple of ways: A woman comes to the marriage with children from a prior relationship. It's possible they were adopted by her husband but more likely, the child's true last name is not recorded correctly in the census. Also, children shown under a husband and wife may only be the children of the husband since the relationship shown is to the head of household. Keep an eye out if you notice a gap between children, children not listed by descending age, oldest child/children born before the couple's marriage, etc.
Age is but a number
Usually, the closer to the event, the more accurate the number but that rule also has exceptions. If for example, the enumerator got info from a neighbor or made an (un)educated guess, the earlier number could be less accurate. Use all available data like school records, permission for marriage if the girl was under age, wills that indicate when children were minors to zero in on the correct age.
Racial classification
In the 1870 through 1930 census, race was very important so in my experience, it's almost never in error as in writing down a B instead of a W. On the other hand, you'll see differences from census to census when it comes to Black/Negro/Colored vs Mulatto. Mulatto does not indicate a White parent. The person could just have "looked" mixed to the enumerator or they may be part Native American or maybe they had a White great grandparent. Mulatto seems to have disappeared in the 1930 census (or at least the areas where I'm researching).
To conclude (Part I?), I would slightly modify the old Ronnie aphorism:
Don't Trust, But Verify