Orthography

Here we attempt to establish a standard for spelling Old English words. This is mainly used to determine the canonical form of a given word and is not intended as a standard for general use.

This standard is still in development and the dictionary may not necessarily follow this standard yet.

Our standard is based largely on Early West Saxon. This is because West Saxon is the best-attested dialect, and Early West Saxon is, unsurprisingly, more conservative than Late West Saxon, thus conveying more information about the history of the word. However, Late West Saxon is much better attested.

Leveling

Prefer forms such as burg to burh, as these forms are more etymological and more consistent with other forms.

Word-final double consonants

Some words have a stem that ends with a doubled consonant, but sometimes it is not doubled when there is no suffix: one writes cattas, but cat and catt are equally valid. In all such cases, we choose the doubled form, and so the lemma should be catt.

Short a/o before nasals

Prefer a except when it is clear that o is closer to the Proto-Germanic form.

cg/gg

Write as cg when hard and ċġ when palatalized: frocga, seċġan.

Unstable i

Written as ie, even if the word is unattested with it. For example, ġiefan, not ġifan or ġyfan.

In late manuscripts, the vowels i and y sometimes alternate seemingly at random. This is not the same thing as unstable i; we know for instance that cyning was never read as ciening in West Saxon, even though we find spellings such as kining; rather, the y was produced by i-mutation of u (PGmc *kuningaz > OE cyning).

eo/io

Check the etymology if possible:

• PGmc eu becomes ēo (īo with i-mutation)
• PGmc e becomes eo (io with i-mutation)
• PGmc iu becomes īo
• PGmc i becomes io
• PGmc ju becomes ġeo

When the etymology is uncertain, e.g. ācweorna, or if the word is a loanword, e.g. prēost, use eo or ēo.

This is a somewhat artificial convention based on sound change laws. It may not be that anyone ever wrote according to these rules, and we certainly find manuscripts with words such as <gesion> where one might expect <geseon> (from PGmc *gahsehwaną).

An earlier version of this orthography instead proposed ġio for PGmc ju (e.g., ġiong), since /j/ and /i/ are related sounds. We have decided, however, that the e in ġeo is strictly orthographic, serving only to mark palatalization of the preceding consonant; in such cases, e should be used.

-u/-o

Spell with -u: beadu, not beado.