Decoding an Anglo-Saxon Runic Inscription

Professor John Hines
Madeline McLeod
Cardiff University

I am going to talk about an object with a runic inscription on it. I have worked on this find together with colleagues within Cardiff, plus others in the museum sector in Wales and England, to try to understand what was written on it and why somebody should wish to do that. The item and inscription prove to be very interesting in their own rights, but this is also a valuable example of the importance of different areas of in Archaeology and Conservation working supportively together. 

The object was found in January 2019 near Winchester in Hampshire and promptly reported to a local Finds Liaison Officer under the Portable Antiquities Scheme. It is silver, and so falls under the provisions of the Treasure Act. That it has runes on it was clear to the Finds Liaison Officer from the outset. 

It is immediately identifiable as an Anglo-Saxon strap-end. These are quite common stray finds reported to the PAS. They were the terminals of straps or girdles worn by both men and women; there is some evidence that the wider and larger examples are from male costume. 

The runic writing system is essentially like our own. In fact it is almost certainly derived from the same Latin script, and had been developed by the second century AD. It is made up of individual graphs (the runes) which represent the sounds of the spoken language. 

It appears to start with that we have eleven runes on one side of the strap-end and twelve on the other. These include three examples of a distinctively Anglo-Saxon rune for o, confirming that the inscription is broadly ‘local’. The process of decoding a runic inscription is firstly to identify the individual runes and to ‘transliterate’ them into familiar modern letters, and then to see if we can recognise specific words or word-endings. 

What immediately stood out is a sequence that crosses the end of one line and the start of the other w o r o/o h t æ þ i s, which it is relatively easy to identify with the Old English words worhte þis: literally ‘wrought this’, and meaning ‘made this’. It is common to have inscriptions on artefacts that identify the maker, and so we can expect this pair of words to be preceded by a word or name representing the maker and to be followed by a noun that gives us the word for a strap-end. 

But there are problems with both of those words. Initially the name appears to read either æiemele or æierele, neither of which are names that exist (we have many recorded personal names from this period). There is, however, a simple and plausible solution: i.e. that the name is Hemele. (Slide 4, second click) The fact that it has become obscured seems to be explained by the fact that this character Hemele was at best semi-literate, and mis-copied an exemplar, dividing what should have been the single rune h into what looks like æi

At the other end, what is the name of the object? We can read the last four runes as æsil, but the first letter is obscured by corrosion. The nature of that corrosion led us to hope that x-ray could see through it to the underlying surface. At this point, though, we faced a practical problem in that the British Museum’s system for processing Treasure Act finds does not normally include x-ray, and is not designed take account of questions such as that posed by a runic (or indeed any other) inscription on a find. 

Fortunately, careful and patient discussions with staff at the British Museum (in two different departments) and Finds Liaison Officers both in England and in Wales enabled us to bring the object to Cardiff in late summer last year, where MSc student Madeline McLeod has taken charge of the analytical study. 

Following multiple carefully targeted x-rays, it proved that the object was still largely sound, although still difficult to read because the x-rays showed the text on two sides simultaneously so clearly: nonetheless eventually we could identify the hidden rune as g, giving us (for the very first time) the Old English word for a strap end as gæsil.  

Although we have not seen this word before, we can relatively easily explain it as an example of a word borrowed into Old English (and other contemporary Germanic languages) from Latin sigilla, ‘jewellery’: a word we are now most familiar with as ‘seal’. This Latin word has been given a purely Germanic prefix gæ-, which has a ‘perfective’ function: referring, in other words, to an item of jewellery which finishes off or terminates something — a good description of a strap-end. 

The date of the object and its inscription is around the later eighth century. It is exceptional in having been made so clearly to display the pride of its maker, Hemele, and so it would seem likely that it was intended to be a badge on his own personal costume. Presumably he remained unaware of the extent to which it exposed his own limited literacy, and yet it testifies to a community in which a skilled metalworker could claim status, and writing was familiar. 

2 thoughts on “Decoding an Anglo-Saxon Runic Inscription”

  1. The spacing of the runes suggests that the first rune on ‘Side’ is indeed a runic ‘h’, and I can’t really see why the authors (or their source) might have thought it was the two runes for æ+i [if there is a narrow space, it is enough only to suggest that the carver inscribed the two diagonals before inscribing the second upright of a runic ‘H’]. As for the final word (on Side ‘B’), the ‘g’ has been inserted to support its identification with an OE *gæsil, which is otherwise unrecorded (or, as the authors put it, “we have not seen the word before”, shortly after their bold claim that gæsil is “the OE word for a strap-end”). In fact the OE borrowing of Latin ‘sigilla’ is ‘sig(e)l’ (other spellings – including ‘sægl’ – are available), which is frequently glossed in OE sources (in Latin) as ‘fibula’ (meaning clasp/clamp/brooch). The *gæsil which the authors propose is not only dependent on the insertion of a ‘g’ as the intitial letter, but overlooks the absence of a rune representing the medial ‘g’ in OE sig(e)l. If the reading of the obscure initial rune as a ‘g is correct, the runes can be rearranged to form (the equivalent of) *sigæl. That way we have a recorded OE word with a meaning which suits a strap end. The authors themselves provide a potential explanation of the situation when they suggest in support of their own re-interpretation of the first word in the inscription that the carver was “at best semi-literate”.


  2. æiereleworo ogtæþis-æsil
    It looks very much as if the spelling aims at numeric rune magic
    Sides A and B count 11 runes each (2×11)
    Both sides together produce a runic value of 299 + “hyphen” value 1 adds up to 300.
    So the rune master knew his “orthography” but altered spellings for the sake of the total.
    The Frank Casket is full such numeric speculations which in the end form a calender expanding over an aon!
    Find out on my runic calculator


Leave a Reply

Fill in your details below or click an icon to log in:

Gravatar Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s