Historians routinely draw quite far-reaching conclusions about all aspects of society from archaeological artifacts and textual documents that – naturally – represent only a minuscule fraction of all the artifacts and documents that existed in the time and place that is being investigated.

I have always held doubts about the validity of the boldest of those conclusions and have devised a simple experiment that I believe would shed some light on the strength of the standard methods of the historical-archaeological sciences:

  1. Take a modern society that the historians participating in the experiment (say, United States-based historians specializing in classical European antiquity) are not familiar with (e.g. rural Japan)
  2. Randomly sample artifacts and documents from that society, emphasizing such artifacts and documents that are capable of surviving several centuries
  3. Provide those samples to the historians without any context whatsoever (except for the location of the "find")
  4. Compare the conclusions drawn by the historians with reality

Has an experiment of this type ever been conducted?

  • 1
    The academic discipline you're looking for is "experimental archaeology" - it's a pretty active field. In addition to applying their techniques to modern societies, they recreate artifacts using period technology and test their usage in real-world conditions. May 6, 2014 at 13:19
  • 1
    what he describes is not what experimental archaeology does. EA is trying to recreate artifacts that we don't find in the dirt. The poster wants to test the acuteness of the conclusions found by archeology with a 'test' set of finds that we know the results about. The problem is that you won't get true archaeologists to start a new career digging up fake artifacts to start a new field to compare with the actual results. Humans aren't lab rats for your amusement.
    – Oldcat
    May 6, 2014 at 23:33
  • 1
    This would test the historians' ability to draw valid conclusions from archaeological data. I strongly suspect that the results of such experiment would be discouraging. Therefore I doubt that historians would knowingly subject their reputation to such a test. It would be a magnificent experiment though.
    – Michael
    May 7, 2014 at 22:54
  • 2
    I think it is paranoid to suggest that not only this generation but all generations' archaeologists are fakes and liars trying to save their reputations rather than people trying to discover the truth the best they can.
    – Oldcat
    May 7, 2014 at 23:48
  • 2
    @pew I'm not sure I follow. As far as I can tell (I just skimmed an article) authors claim that it is possible to build the Great Pyramid of Giza in 10 years with 13200 men using modern project management and ancient technologies. Sorry, I couldn't find any far-reaching conclusions on how pyramid was actually built by ancient Egyptians. May 27, 2014 at 17:16

1 Answer 1


Not that I know of, but I'd think that such a process would be so far removed from quantifiable science that it'd be rendered pointless.

The job of a historian is to make best guesses given the evidence that's presented itself. Given that it's easy for two different historians to look at the same evidence and draw different conclusions. Then without really knowing the reality of the situation you can't quantify accuracy of the historian's predictions.

Instead we could go about the process you described but that wouldn't represent an accurate model of different civilizations and circumstances. In the life sciences we do testing on animals that represent a model that's close enough for practical purposes, but there are way too may variables to do the same with history.

With all of that in mind it's not so much important to gauge how correct historians are, but rather that when historians make predictions to make the reader aware of the small amount of evidence involved and how likely what they are saying is to be true.

  • Agreed; for science as well as history (and hell, most fields), the reporting of it (whether by the researchers or journalists of the field) often presumes the reader knows just how precarious the findings may or may not be... and the confidence of many researchers is astounding to a degree worthy of psychological study.
    – Smithers
    Dec 5, 2014 at 22:13

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.