Wednesday, July 17, 2013

What Exactly is "Indirect Evidence?"

There was a spirited discussion on the Transitional Genealogists Forum message board earlier this month.  The first message I see in July 2013 was by Michelle Lewis, titled "direct vs. indirect."  There are many responses to this post that involved many of the best minds in the genealogy world.

Thomas W. Jones new book, Mastering Genealogical Proof, defines "Direct Evidence" as (page 14):

" information item that answers a research question all by itself.  When we consider the possibility that an information item describes what actually occurred we are using that information item as direct evidence."

And "Indirect Evidence" as:

" ...a set of two or more information items that suggest an answer to a research question only when they are combined."

The discussion on the TGF board quickly went to "it depends on the question being asked."  Example questions and answers included (my interpretation, I hope I got them right!):

*  With a question of "How old was John Doe on 1 June 1850?" an answer of 49 years old is "Direct Evidence" (because it answers the question - it doesn't have to be exact, or correct).

*  With a question of "What year was John Doe born if he was age 49 on 1 June 1850?" (and therefore he was born in either 1800 or 1801), the answer would be "Direct Evidence" (because it answers the question, but not exactly).

*  With a question of "What was John Doe's birth date if he was age 49 on 1 June 1850?"  The answer would be "Indirect Evidence" because it does not answer the question exactly.

*  With a question of "What is the birth date of John Doe if he died on 1 June 1850 and his age at death was 49 years, 4 months and 3 days?" the answer would be "Direct Evidence" (because this single piece of information provides enough data to permit calculation of an exact birth date).

*  With a question of "What is John Doe's birth date?" if a death certificate says the birth date was 29 January 1801, the answer would be "Direct."

Elizabeth Shown Mills added this in the discussion:

"If we assume that direct evidence gives a full answer and anything less than that is INdirect, then we have simplified both our expectations and our research. If we redefine INdirect as "something that addresses the question but doesn't give a full answer," then our focus--and our research  methodology--stays at base level."

In my assignment of Direct or Indirect Evidence to source citations that provide information about an exact date (day-month-year), I've been using "Direct" for information that states a specific day-month-year, and "Indirect" for information that requires additional information to determine a specific day-month-year.  

My practice has been to assign "Indirect" to the first four examples above if the question was "What is John Doe's birth date?" My view was that I needed more information to determine an exact day-month-year in the first three cases, and that I need to make a calculation in the fourth case (since there were two different pieces of information - a death date and an age at death).  I find age at death quite a bit in the Massachusetts Vital Records, 1841-1915 collections and use them to determine birth dates.  Some gravestones list a date of death and an age at death also. 

Is a death record that includes both a death date and an age at death one piece of information or two?  I've been treating it as two, but it seems like it's one now to me.  Am I quibbling over a small point?

The TGF message board also highlighted that assigning Direct or Indirect to evidence items was not the most important factor here - the purpose of the assignment of the term is to help the researcher evaluate all of the available evidence and apply the Genealogical Proof Standard to draw a soundly reasoned conclusion that answers the question at hand.  

Elizabeth Shown Mills summed it up nicely with:

"Success with a BCG portfolio--indeed, success with genealogical problem solving--doesn't hinge upon learning textbook definitions or knowing what label to apply to something. What makes us successful is (a) the ability to recognize evidence that doesn't jump off the page and slap us in the face; and (b) the ability to use that unobvious information to solve a problem. "

I encourage interested readers to read the entire thread of responses.  The TGF message board is one of the best freely available forums to ask questions and receive answers from knowledgeable and respected genealogists.  

Am I correct with my assignments of Direct vs. Indirect in the five examples above?  What examples do you have that you are puzzled about?

The URL for this post is:

Copyright (c) 2013, Randall J. Seaver


Cousin Russ said...


Another great post and on a subject I was confused about in that thread of messages.

My take is knowing the right question to be asking. For birth (date) information I only have ONE question. What is my subjects Birth Date. Most sources fail to completely answer that question. I agree with you that the first four are indirect. I may get close but not jump off the page complete answer, which would be direct as I understand it.

Having found the direct answer, I don't stop looking, but continue to look for conflicting information even I might find more confirming information.

I may not be focussing on that specific question, but be aware that there is a question that might have another answer


Elizabeth Shown Mills said...

Russ, given your stated question (What is my subject's birth date?), let me pose a pair of situations that, IMO, better shows the difference between direct and indirect evidence.

Let's say your subject is John Puzzler. Let’s say we have his memoir saying that he was born during a blizzard the first year his parents lived in North Dakota. Let’s say we find an affidavit by Jenny (Puzzler) Jones who, we know, was his sister. Jenny says: “On Christmas Day, right after we moved to North Dakota, my mother died in childbirth. I helped deliver my brother because we were in the middle of a blizzard and no one could go get a doctor.” Jenny does not even mention John by name. She was not addressing the issue of John’s birth date. But her testimony IS evidence we can use to determine John’s birthdate. It’s INdirect evidence.

By comparison, let’s say that we find testimony by Sam Puzzler, whom we can otherwise prove to be John’s first cousin. Sam testified: “Both John and I were born on Christmas Day, in the middle of a blizzard, right after our families moved to North Dakota.” Sam has supplied us with a day and month. Sam does not specify the year, so we don’t have ALL the information we want. But he did directly, explicitly, address the issue of John’s birth date; and he explicitly, directly stated part of that date. He gave us direct evidence for our question: What was John’s birth date?

If the evidence directly addresses our question, it is direct evidence, whether it tells us EVERYTHING we want to know or not. We wouldn’t say: Oh, yeah, it tells me 2/3 of the answer I want but it doesn’t give me the last 1/3 so I have to relabel it and call it INdirect.

Paul K. Graham said...


I think you bring up one of the most critical responsibilities of a genealogist. Even if you've found the answer to a question, you must be willing to accept the possibility that new evidence will challenge what you think to be true.

After finding direct evidence for a question in one source, it's important to find other sources that either corroborate or conflict. However, it's equally important to recognize that the reasonably exhaustive search does not require us to search endlessly for conflicting evidence, especially as consistent direct evidence stacks up.


Cousin Russ said...


Great. Thank you so much. That was very helpful. I think I am getting much closer in understanding the difference.


Cousin Russ said...


I totally agree with you. That is why I also am on the look out for conflicts, or differing data.

One point that I keep in the back of my mind though, is that I don't have to create a "brick wall" because I only found the Direct Evidence from once source. I may have also incomplete or Indirect Evidence as well. My experience and from I have heard from others, is that we will find that information, conflicting, or direct, when we least expect it.


Anonymous said...

"Indirect evidence" - another genealogical misnomer. It isn't the evidence that's indirect, it's the reasoning that proceeds from the evidence that's indirect.

Linda Schreiber said...

Thanks, Elizabeth. I had thought I understood direct-indirect difference early on as a pretty simple thing, but there seems to be so much discussion and confusion.... Thought maybe I was wrong.
If the source states something, it is "direct". Could end up being wrong, could be partial information, certainly may be not proven. But it is indeed Direct.
Source states "born in a blizzard" is direct. "Probably not born in July" is indirect... ;-) Though likely!