Making History More "Replicable"
Some things that you can do
“Does History Have a Replication Crisis” inspired me to do three things. First, I redoubled my efforts to locate, and, through the magic of Substack, correct, mistakes that I have made in my own writings. Second, I searched for practices that might mitigate the problems that Anton Howes, the author of that piece, identified. Third, I wrote a short piece, for publication on the Tactical Notebook, that makes use of the aforementioned techniques.
The first thing that an historian can do to promote the reproducibility of his work is to provide his readers with copies of his sources. If the source can be found on a reasonably stable website (such as Gallica, Hathi, or the MDZ), a link will suffice. If, however, the source has no home on the interwebs, an historian can organize a website for the storage and retrieval of PDFs. (He can do this with the help of a file storage service, such as Dropbox or Mega, or a repurposed discussion forum.)1
The second thing that an historian can do is show his work. Thus, if (as is often required in my own corner of Clio’s realm) he did some calculations, he can type those up neatly and file them in the same place as his sources. Likewise, if his research took him down some blind alleys, he may wish to describe those in a brief note. (“I looked in this place, but found nothing.”)
Finally, when engaging in speculation, an historian can provide his readers with fair warning that he has embarked on a flight of fancy.
I look forward to the day when Substack provides its authors with an online storage-and-retrieval site for their sources and working documents.
I have in recent years come to the conclusion that every time an historian writes or speaks on history the first words should be something like “as best we can tell now from the available sources this is what we think happened...”
Thanks for this.
Was Leopold von Ranke Robert Graves’s grandfather or something?
(Just an aside.)