In many Greek morphology projects, I’ve wanted a way of conveying the surface form of an inflected word while also conveying the underlying components prior to the application of the sandhi rule. A couple of years ago, I came up with a simple representation for inline annotation.

Say you want to convey the fact that φιλοῦμεν comes from φιλε + ομεν by application of the rule that ε + ο → ου. In the representation I’ve been using you’d write `φιλ|ε>ου<ο|μεν`

.

This enables you to see the stem and affix easily but also the result of sandhi.

So what `A|B>C<D|E`

means is there is a sandhi rule that B + D → C and that rule has been applied in AB + DE to form ACE.

Using Stump’s terminology introduced in a previous post:

- A / φιλ is the
**theme** - CE / ουμεν is the
**distinguisher** - AB / φιλε is the
**stem** - DE / ομεν is the
**affix**

It also means that you can search for `|B>C<D|`

to find where that particular sandhi rule has been applied.