The Italics Test – How Strong is Your Tree’s Foundation?

Geoff Rasmussen

04 Oct 2012

We have heard the story about two men and their homes – the first who builds his house on the sand and the second who builds his house on the rock. When the winds and rains come there is only one house that remains. The one with the solid foundation stays strong.

Our family trees fit well into this analogy. The first family tree is built on a foundation of compiled and published resources – books and other records with few primary sources, which rely heavily upon the research of others. The second family tree is built on a foundation of original records and careful analysis, using the Genealogical Proof Standard (GPS) as their guide.

Not living close to the Family History Library or archives that house my ancestors' original records, I turn first to compiled records – books and other publications online. They are just more accessible than most original records. I am fine with using these types of records for their clues, but the problem comes when the foundation of a family's pedigree is based on compiled records alone. I have personally disproved many pedigrees as I have turned to the original records.

How does one weigh the stength of their pedigree? Judy Russell and Michael Hait, in their recent webinars, Building a Family from Circumstantial Evidence and What is a 'Reasonably Exhaustive Search'?, explained this very well as they taught the about Genealogical Proof Standard.

Another quick method of visualizing the strength of a family's evidence is to perform the Italics Test. I learned about this from Elizabeth Shown Mills, author of Evidence Explained: Citing History Sources from Artifacts to Cyberspace at a national genealogy conference a few years back. While I do not recall her exact words, following is what I took from her class.

In a bibliography, titles of published records are italicized. Titles of original records are shown in normal case. If you look at a book's bibliography and find that the majority of works cited are italicized, then you know that the author/compiler relied mostly on the works and interpretations of others which may or may not be accurate. If the bibliography showed few italicized titles, you know that the author/compiler relied on original records.

Which foundation is your pedigree built on?

You can perform the italics test one family at a time. Here's how.

In Legacy Family Tree, create a Family Group Record (go to Reports > Report Menu > Family tab)
Select the option to include the Bibliography (Report Options button > Sources tab)
Preview the report and view the bibliography (the last section of the family group record)

Below is the Bibliography for Asa Brown's family group record. It contains a mixture of italicized (published) and non-italicized (original records) sources, but the majority of sources cited are original records.

Compare this with the bibliography of John Smith's family group record:

Here the researcher relied solely upon the compilations of others, which has its place in the research process, but I would hesitate to base my entire tree on these kinds of records.

You need to be using Legacy's SourceWriter for this to be effective. As you are entering your sources, the SourceWriter knows to automatically italicize (or not) the title based on the type of source you selected. The SourceWriter is available in both the free standard edition and the deluxe edition of Legacy.

While there will never be a magic number of how many sources you need, or which sources you need to consult to begin to make a sound conclusion (again, see Michael Hait's webinar where he answers the question of how many sources you really need) this italics test is a good start to quickly visualizing how often we rely on the works of others to build our own genealogical foundations.

Comments (4)

I have stopped using compiled trees from Ancestry.com. I rely on source records and I look at every one of them before adding to my trees. I have so many errors because of just clicking and adding. Very bad idea. I have had to delete over 1000 names. Not to mention some of those census records are not indexed right! I have just found two census records for 1880 of the same family done six months apart. I couldn’t figure it out then I realized the original transcriber must of done something wrong. I have a sister for my great-great-great grandfather which I didn’t have in one census record but the street address are the same the peoples names and birthdays are the same. Weird.

While I do essentially agree with this article but “proof” is a strong word. I can’t often really offer proof about what I did two days ago. Original records are very often mistaken or mistranscribed (by the original scribe), unavailable or just plain misinterpreted. . While one must aspire to the greatest accuracy possible, proof is often a dream. And if a secondary source is used, it should show that it had access to a primary source. More than proof, common sense and a Sherlock Holmes type of deduction is what makes one’s research valuable.

I also have found pedigrees online that were based entirely on one original record, the 1900 census, or on three census records… no vital records (compiled or otherwise). Yikes!

This should headline every site on the Internet, free or subscribed to. Ancestry.com has become a tangle of errors with their “click and attach” feature. I think more trees have errors than do not. New genealogists (I use the term loosely) seem to think if it is the same name, it must be the same person, never mind that a person born in 1842 could not possibly not be a child of person born in 1878! Or is in two or three vastly different places at the same time with large families in all places with different names and ages. Good research takes time and can’t be done exclusively with a click of a button. Family trees complies by others should be closely examined and only used for reference to see the proof for yourself.

SH
susan Hiland

14 years ago

I have stopped using compiled trees from Ancestry.com. I rely on source records and I look at every one of them before adding to my trees. I have so many errors because of just clicking and adding. Very bad idea. I have had to delete over 1000 names. Not to mention some of those census records are not indexed right! I have just found two census records for 1880 of the same family done six months apart. I couldn’t figure it out then I realized the original transcriber must of done something wrong. I have a sister for my great-great-great grandfather which I didn’t have in one census record but the street address are the same the peoples names and birthdays are the same. Weird.
MQ
Martin Quartermaine

14 years ago

While I do essentially agree with this article but “proof” is a strong word. I can’t often really offer proof about what I did two days ago. Original records are very often mistaken or mistranscribed (by the original scribe), unavailable or just plain misinterpreted. . While one must aspire to the greatest accuracy possible, proof is often a dream. And if a secondary source is used, it should show that it had access to a primary source. More than proof, common sense and a Sherlock Holmes type of deduction is what makes one’s research valuable.
PR
Paula Ryburn

14 years ago

I also have found pedigrees online that were based entirely on one original record, the 1900 census, or on three census records… no vital records (compiled or otherwise). Yikes!
NR
Norma Rogers

14 years ago

This should headline every site on the Internet, free or subscribed to. Ancestry.com has become a tangle of errors with their “click and attach” feature. I think more trees have errors than do not. New genealogists (I use the term loosely) seem to think if it is the same name, it must be the same person, never mind that a person born in 1842 could not possibly not be a child of person born in 1878! Or is in two or three vastly different places at the same time with large families in all places with different names and ages. Good research takes time and can’t be done exclusively with a click of a button. Family trees complies by others should be closely examined and only used for reference to see the proof for yourself.