Performance as a witness
A recording can witness one event of performance; its transcription is an editorial representation of that event. Harvard’s Milman Parry Collection preserves recordings, notebooks and dictated songs from South Slavic oral traditions. A notebook line and a disc recording may document related but different events. Neither automatically stands for every performance by a singer or for an original version.
Homer Multitext treats papyri, quotations and medieval manuscripts as evidence of a tradition with multiform readings. The extant written sources can show variation; they do not let us replay an ancient performance. A comparison with documented living performance can illuminate how oral composition works, but it does not prove that a particular Homeric line was sung in exactly the reconstructed way.
Editorial choices
When publishing an oral text, state the performer, place and date if known; whether the source is audio, dictation, field notes or later copy; and how pauses, repetitions and repairs are represented. When publishing a written witness of an oral tradition, avoid silently merging versions into one fluent narrative. Variant performance can be meaningful evidence rather than a defective copy.
Practice: compare the collection description of a recording and a notebook transcription. List what each source can show directly and what requires editorial interpretation.
Modern-language examples on this page clarify a linguistic pattern. They do not establish unattested sounds or forms for an ancient language. Follow the cited sources below and keep editorial readings separate from the surviving trace.