Shelf numbers for books in archive

Real-size facsimile of the Codex Gigas. Image from Wikipedia, by Michal Maňas, 2007.

This article was originally posted on Encased in Steel on 24th February 2017. It has been edited and improved for posting here.

You may have observed that when discussing original source material, people will sometimes refer to sources by their shelf numbers: a series of letters and numbers, rather than using a more readable name. This much more common with medieval sources (particularly handwritten manuscripts) than with printed books, as printed books usually have their own title, whereas manuscripts often exist without a title.

What do these combinations of letters and numbers mean, and how can we understand them?

Generally speaking, it is because each archive (be it a museum, a library, a university, a private collection, whatever) needs to have some way to identify and catalogue everything in their collection. This identifying code needs to be unique.

Why is uniqueness important? Well, every item needs to be stored in a specific place, so that it can be found again when it is required for study or maintenance. People need to be able to refer to one particular item in the collection when they use it in their own research. If several items share the same identifier, then each item cannot be identified uniquely, and this can cause all manner of problems and confusions. Therefore, each item must have its very own unique identifier.

Different archives, museums and libraries all have different methods for constructing their identifier codes; some are simple listings, others perhaps record some useful data about the item’s place in the collection or about the item itself.

A simple example

A simple example is that of the so-called “Codex Döbringer”, the HS 3227a. In this case, the HS is an abbreviation for “Handschrift”, the German word for “handwritten”. This book is held in the Germanisches National Museum in Nuremberg, in the HS collection of handwritten manuscripts, and it is item number 3227a within that collection.

Another example

The so called “Codex Ringeck” has the identifier Mscr.Dresd.C.487 and is a manuscript (as opposed to a printed book), held in the Sächsische Landesbibliothek in Dresden, in collection C, and is item number 487 within this collection.

In this collection are several sources that are of interest to us in historical fencing, including:

– Mscr.Dresd.C.13 (Johann Georg Pasch)

– Mscr.Dresd.C.15 (by Heinrich von Günterode)

– Mscr.Dresd.C.93 (first volume of Paulus Mair)

– Mscr.Dresd.C.94 (second volume of Paulus Mair)

– Mscr.Dresd.C.487 (which includes Ringeck’s longsword gloss)

Not all manuscripts have a title, nor do that all have a single author. Many manuscripts are collections of different treatises, bound together for whatever reason. For example, the Mscr.Dresd.C.487 contains the following treatises and elements:[1]

– 1r-2v              front matter

– 3r-9v              a copy of Liechtenauer’s Zettel

– 10v-48v         Ringeck’s gloss of Liechtenauer’s Zettel for Bloßfechten

– 49r-54r          Ringeck’s additional longsword material

– 54r-55v          Lignitzer’s treatise on sword and buckler

– 55v-57r          a fragment of Liechtenauer’s Zettel

– 57r-59v          anonymous gloss of Liechtenauer’s Zettel for Bloßfechten

– 66r-77v          anonymous treatise on wrestling

– 78r-84r         Ott’s treatise on wrestling

– 84r-86v         an anonymous treatise on wrestling

– 88r-108r       Ringeck’s gloss of Liechtenauer’s Zettel for Kampffechten

– 109r-121v      Ringeck’s gloss of Liechtenauer’s Zettel for Rossfechten

It can be seen that this manuscript includes work by Liechtenauer, Ringeck, Lignitzer, and Ott, as well as some anonymous work. There is no information about who compiled the manuscript, what the overall title of the work should be, or even a date when the manuscript was bound together. With such bibliographic information lacking, it can be very difficult to refer to this work so that other people would know immediately which manuscript you mean; but by referring to the Mscr.Dresd.C.487, it is instantly clear which source you are citing.

Collections within collections

Some archives have very extensive collections, and therefore they need an identifier system that can track items within collections within collections within collections. The “Kölner Fechtbuch” and the “Glasgow Fechtbuch” are two good examples to show this nesting effect:

In the Historisches Archiv Köln, the “Kölner Fechtbuch” has the identifier MS Best.7020 (W*)150, which stands for a manuscript in the Bestand collection, within collection 7020, then within the Wallraf collection, item number 150.

In the Glasgow Museums, the “Glasgow Fechtbuch” has the shelf number MS E.1939.65.341, so it is item 341 within collection 65 within collection 1939 within collection E.

Collection E.1939.65 is the RL Scott Library, which contains many fechtbücher and treatises relevant to historical fencing, including Donald McBane’s book The Expert Sword-Man’s Companion (MS E.1939.65.498). Even though McBane’s book has an author and a title, it still has a unique identifier to help the Glasgow Museums to track it and to differentiate it from any other copies of this work that may be held elsewhere or that come into possession of the Glasgow Museums.


This may seem quite overwhelming, as there is a lot to learn, and every archive employs a different system of identifier. It may also seem strange to a modern person that a book doesn’t actually have a title, an author, or a date of publication. The medieval world of books and transmission of knowledge was quite a different world from that in which we live today, and trying to impose modern conceptions of what a book “should be” and how to refer to it can be problematic in many instances.

How could someone go about learning all of this? To be honest, I don’t really have a good answer for that, since I have picked it up over the last six years or so of doing HEMA and becoming more involved in the research side of things. I have not received any formal training in any of this, so there are undoubtedly gaps in my knowledge, and I may have missed some important points in the explanation above.

If anyone knows of a good basic resource to help beginners understand this sort of material, I would be delighted to hear about it, so that I can go and learn more about the matter. If anyone has corrections for any of the statements in this article, I would welcome those corrections.

In the meantime, I hope this article gives a basic grounding to help people understand a little more about the issue!


[1] Information drawn from the Wiktenauer: