Negative Evidence? To Use or Not to Use?

16 August 2014

Evidence is messy.  Because it is a mental construct, it rarely gives us the clear and simple answers that we seek. Sources, by contrast, are physical: we can touch them, see them, smell them, hear them. Information is also physical, visible, audible. Evidence, however, is intangible. It’s only what we think certain information means. That’s all it can be―until we make something concrete from it by processing it and molding it into a meaningful and convincing form.

Especially is this true for negative evidence. Even scholars differ in their opinion as to whether negative evidence is useful or even valid. For example:

  • David Hackett Fischer—who always writes elegantly and persuasively—contended a generation ago that "evidence must always be affirmative" and that so-called negative evidence "is no evidence at all" because it is "a contradiction in terms."1
  • David Henige more recently advises, to the contrary, that "historians must respect contextually suggestive silence. It must be plumbed, found to be true—or not—and brought into arguments where it is relevant."2

EE, which defines negative evidence as an inference one can draw from the absence of information that should exist under given circumstances (glossary, p. 826), takes the stance that negative evidence can be immensely valuable, when reinforced by thorough research in reliable sources and skilled analyses.

How do you do that in your own research? Or do you?

1. David Hackett Fischer, Historians' Fallacies (New York: Harper Collins, 1970), 68.

2. David Henige, Historical Evidence and Argument (Madison: University of Wisconsin Press, 2005), 175.

Photo credit: "Grungy film strip or photo negative," CanStockPhoto ( : downloaded 4 August 2014), used under license.


Submitted byyhoitinkon Sun, 08/17/2014 - 15:25

I agree that negative evidence can be very valuable. I've often been able to narrow down the period for a death by looking at the names of the grandchildren. In the area where my paternal ancestors came from (the Achterhoek in the province of Gelderland, the Netherlands), people named their children after the children's grandparents, but only if the grandparents had died already. The first son was typically named after the paternal grandfather. So if you have an ancestor called Hendrik and you see two of his grandchildren named Jan and Hendrik, the fact that the first was wasn't called Hendrik is evidence that Hendrik sr. had not died then. He must have died between the births of Jan and Hendrik.

The burial records for this area are very sketchy, often referring to people as 'the old man at the Hoitink farm' or similar non-descriptive wordings. But often the grandchildrens' names can narrow the period to a couple of years, sometimes even a couple of weeks. In combination with a thorough analysis of who was living on the farm at that time, that can help to identify who that old man must have been. 

Submitted byRachel Mackin Evanson Mon, 08/18/2014 - 19:14

I agree that negative evidence can help support a case. My great-grandfather and his three brothers were born Szodry but after their mother remarried "adopted" their stepfather's surname of Dorn. Most records produced after their births gave their surname as Dorn and Robert, the stepfather, as their father. The only two that contradict this are a brother's death record, where my great-grandfather was the informant, saying the father was Charles Dorn. This could be explainable if the person taking information assumed that since both men were Dorns that their father's name was Charles Dorn and not Charles Szodry. The second document is the 1910 census where the sons are listed as stepsons of Robert. I use negative evidence to prove that there were no children born to Robert Dorn and Bertha Offhaus during the time surrounding their birth dates and that they did not even marry until after all the sons were born. However, on those days a woman named Bertha Offhaus, same age, same birthplace, same address as Robert Dorn's wife had children with the man my great-grandfather stated was his father.

I feel more confident being able to say that I did a search of the records of that time and could find no matching Dorns. Although birth records were filed sporadically at that time period it helps that I tried to disprove the fact that they were born Szodry. This is also not the only evidence I have proving my case. It was just the evidence I felt relevant to the comment.

Submitted byrworthingtonon Tue, 08/19/2014 - 19:28

A very wise Editor once told me

You haven't looked in the right places

Since I saw that comment I "record" the negative evidence in my Research Log and put it in my exhausing research to do list.

My case was to find if an individual served in the Civil War. I found a record that a person signed a local CW registration log book, but that is the only record I have found. The only place I can record that is in my Research Log. There isn't enough information on the entry to "prove" that is was my person. But that doesn't answer my research question.


To this date, I have not found the answer to my question, but I have a number of "not found in this record index" or this website, so I'll keep looking. Before that comment, I probably would have thought follow up research as negative evidence, but not now.

I have put that question on my "probably won't be able to answer that question" unless I find some record that said that he DID serve.

For what ever that is worth.

Love the questions.



Submitted byJadeon Fri, 08/22/2014 - 15:35

In one family group, a persistent omission in several accounts pointed to extensive research on the "why" of it -- and the omission itself supported a surprising likely solution.

There is almost always a logical reason for negative evidence.

Thanks for this reminder :D