DCLP/67/67136.xml (TM 67136, LDAB 8405, https://papyri.info/editions/cla/7/991) appears to combine the catalogue record of one document with the edition text of another.
1. The record contradicts itself, independently of any text analysis.
<term type="overview">sacramentarium celticum : Visitatio infirmorum (fragm.)</term>
...
<langUsage>
<language ident="en">English</language>
<language ident="grc">Greek</language>
</langUsage>
The record is described as a Celtic Latin sacramentary (Lowe, CLA 7 no. 991: parchment codex, Irish majuscule, Ireland, s. VIII, St. Gallen, Stiftsbibliothek 1395) while declaring the language of its text as Greek. This does not look like a stray default: across the 14,842 files in DCLP/, ident="la" occurs in 1,266 records and ident="grc" in 5,543, so the distinction is maintained consistently elsewhere.
2. The edition text is Homer, Iliad 3.
<div type="edition"> holds Greek hexameter matching Iliad 3.267–347:
- col. i (heavily damaged):
Ἀγαμέμν[ων], ἔχευ[αν], ἔπει[τα] → Il. 3.267, 3.270, 3.271 (preparations for the oath);
- col. ii, near-verbatim Il. 3.320–347:
Ζεῦ πάτερ Ἴδηθεν μεδέων κύδιστε μέγιστε (3.320), κνημῖδας μὲν πρῶτα περὶ κνήμῃσιν ἔθηκε (3.330 ff.), πρόσθε δ᾿ Ἀλέξανδρος προΐει δολιχόσκιον ἔγχος (3.346–347).
Quantitatively: 136 of 167 character 12-grams of the visible text match tlg0012.tlg001 in canonical-greekLit — 81.4%. This counts only text outside <supplied reason="lost">, so it does not rest on editorial restoration; 58.1% of the edition's characters are visible on that measure.
3. Note on provenance of the error. The only entry in <revisionDesc> is 2014-12-10 "Crosswalked to EpiDoc XML". That may or may not be where the two documents were joined; I have not investigated, and the diagnosis is not the point of this report — the mismatch is.
I am not proposing an attribution or a fix, only reporting that the metadata and the edition text of this record describe two different documents. Happy to supply the matching script if useful.
DCLP/67/67136.xml(TM 67136, LDAB 8405, https://papyri.info/editions/cla/7/991) appears to combine the catalogue record of one document with the edition text of another.1. The record contradicts itself, independently of any text analysis.
The record is described as a Celtic Latin sacramentary (Lowe, CLA 7 no. 991: parchment codex, Irish majuscule, Ireland, s. VIII, St. Gallen, Stiftsbibliothek 1395) while declaring the language of its text as Greek. This does not look like a stray default: across the 14,842 files in
DCLP/,ident="la"occurs in 1,266 records andident="grc"in 5,543, so the distinction is maintained consistently elsewhere.2. The edition text is Homer, Iliad 3.
<div type="edition">holds Greek hexameter matching Iliad 3.267–347:Ἀγαμέμν[ων],ἔχευ[αν],ἔπει[τα]→ Il. 3.267, 3.270, 3.271 (preparations for the oath);Ζεῦ πάτερ Ἴδηθεν μεδέων κύδιστε μέγιστε(3.320),κνημῖδας μὲν πρῶτα περὶ κνήμῃσιν ἔθηκε(3.330 ff.),πρόσθε δ᾿ Ἀλέξανδρος προΐει δολιχόσκιον ἔγχος(3.346–347).Quantitatively: 136 of 167 character 12-grams of the visible text match
tlg0012.tlg001incanonical-greekLit— 81.4%. This counts only text outside<supplied reason="lost">, so it does not rest on editorial restoration; 58.1% of the edition's characters are visible on that measure.3. Note on provenance of the error. The only entry in
<revisionDesc>is2014-12-10 "Crosswalked to EpiDoc XML". That may or may not be where the two documents were joined; I have not investigated, and the diagnosis is not the point of this report — the mismatch is.I am not proposing an attribution or a fix, only reporting that the metadata and the edition text of this record describe two different documents. Happy to supply the matching script if useful.