# ΦΑΙΝΕΤΑΙ ΜΟΙ after *Classical Philology*: the archive's answers, the two readings compared, and the corrections owed

Working dossier, 16 November 2026. Not deposited. It is to be read beside the reading of the report (CP-report-reading.md).

The search covered 124 deposits that bear on Sappho 31. The records used here, by role:
- **The earlier version and its companions:** #1270 (ΦΑΙΝΕΤΑΙ ΜΟΙ v1.0, 9 January), #283, #313, #134, #828.
- **The errata on the edge (15 August):** #1052, #1474, #1475, #1476, #1478, #1484.
- **The Catullus room:** #576.
- **Research notes and later work:** #1548, #1645, #1749, #1752, #1756.

The submitted file is dated 24 July 2026. Most of what follows postdates it.

---

## 1. Where the archive has already answered the report, or gone past it

### 1.1 κῆνος (the report's objection 1)

- **v1.0 had more than the submission kept.** #1270 §II.1 cites Bonifazi on κεῖνος in Pindar: "archaic poets exploit this distancing function for complex effects: temporal displacement, the evocation of figures removed from…". It also sets the poem among archaic objects that address their reader: "The Nestor cup speaks as 'I' to 'you' (the drinker who is also reader). The Dipylon oinochoe addresses 'whoever of all the dancers performs most gracefully.' Mantiklos dedicates himself…" (CEG 454; Svenbro 26–43). The CP revision dropped Bonifazi, the cup, the Dipylon jug and Mantiklos. It answered the report's objection with less than the archive already held.
- **The cup itself says κῆνον.** Pithekoussai, c. 740–720: ὃς δ' ἂν τοῦδε π[ίησι] ποτηρί[ου] αὐτίκα κῆνον / ἵμερ[ος αἱρ]ήσει καλλιστ[εφάν]ου Ἀφροδίτης, "whoever drinks from this cup, *that one* desire of fair-crowned Aphrodite will seize at once" ([text](https://bafybeiemxf5abjwjbikoz4mc3a3dla6ual3jsgpdr4cjr3oz3evfyavhwq.ipfs.4everland.io/wiki/Cup_of_Nestor.html)). The oldest Greek verse inscription built around a future user does three things: it uses the distal κῆνον for that user, picks him out by an indefinite relative (ὃς δ' ἄν), and gives him to Aphrodite's ἵμερος. Sappho 31 has the same three: κῆνος, ὄττις, and the ἰμέροεν laugh. No record in the archive yet makes this point.
- **The archive's later grammar.** #1476: "the poem's first noun-phrase is a distal deictic plus an indefinite relative — **a position, not a person**." #1484 calls it "a seat, not a person". With ὄττις, the triangle's "that man is close to the girl and distant from me" stops being enough. The triangle needs a man; the grammar gives a seat for whoever sits.
- **The third reversal.** #1756 Q1, posed in #1475: Sappho as the one who takes the seat herself, the resurrected Sappho. It is dated and still open.

### 1.2 The addressee (objection 2)

- **The answer the reader's own parallels give.** Anaktoria's ἔρατόν τε βᾶμα κἀμάρυχμα λάμπρον προσώπω (fr. 16) is a step and a face, things seen. Dika's ἀπάλαισι χέρσιν (fr. 81) are hands, touched. The addressee of 31 is given voice and laughter only.
- **The archive, by way of Lardinois.** #1548 §2 reads Lardinois's own method: the memory fragments about young women "recall **performances**", and "Bierl 2003 read[s] Anaktoria's ἔρατον βᾶμα as dance move". So the reader's counterexample is a performing body remembered. Fr. 31's addressee is the one woman given nothing a body does except sound.

### 1.3 Literacy, papyrus and γράμματα (objection 3)

**The ancient record of the maker.** #1752:
- Longinus quotes fr. 31 "to show how a maker works".
- §2.6a: the fifth-century Attic hydria, "reading from a roll to three women".
- §2.8: Ovid, *Heroides* 15, "studiosae littera dextrae".
- Its finding: "The illiteracy is the reception's: a figure it produces, a voice without a hand." #1756 records the literal-illiteracy proposition "as unargued in the seated record".

**The modern debate, mapped.** #1548:
- Lardinois 2008 names the written-futurity position: West 1970: 315, Parker 1981: 161, Stein 1990. Lardinois argues against it.
- The archive answers him (#1548, "for the CP piece"): "the deictic-slot mechanism is compatible with reperformance — a future *singer* occupies the slot as fully as a future reader — which converts the strongest objection into support for the mechanism's medium-independence". And on fr. 147's chorus: "the mechanism is **speaker-agnostic as well as medium-agnostic**."
- This is the developed answer to "anachronism". The seat of §§I–II does not depend on papyrus. Only two things do: the χλωρός identification of §IV, and the supplement γράμμασι.

**What the archive does not yet hold, and the record supplies:**
- *Sappho's family in Egypt.* Herodotus 2.135 names Charaxos, "Sappho's brother" (ἀδελφεὸς Σαπφοῦς τῆς μουσοποιοῦ), trading at Naukratis, the Greek port in the papyrus country.
- *The Ionians' own name for the roll.* Herodotus 5.58 says the Ionians had long used papyrus rolls, calling them διφθέραι from the skins used when papyrus was scarce.
- *Writing as a vehicle in epic.* Iliad 6.168–69: γράψας ἐν πίνακι πτυκτῷ θυμοφθόρα πολλά.
- *Inscribed verse addressed to its future reader,* 8th–6th c.: the Nestor cup, the Dipylon jug, Mantiklos, and Phrasikleia (Svenbro).
- *The noun.* γράμματα in the sense of letters is first attested in the fifth century. The supplement's word should be marked as the restorer's, not Sappho's.

### 1.4 πένητα (objection 4)

The archive corrected this before the report, on 15 August.
- **#1474**, the sixth erratum: "editors accept πὰν τόλματον as Sappho, dagger καὶ πένητα, repair the accents to do so, and record that the metre does not work". Bergk "brackets ἐπεὶ καὶ πένητα as **ad Longini orationem pertinere**; conjectures ἐπεῖπεν". The manuscript is "an unbroken run of letters… ἐπεὶ καὶ πένητα into οὐ θαυμάζεις", with Perseus's section [3] "falling mid-phrase".
- **#1475**: "Six words follow — ἀλλὰ πᾶν τολματόν, ἐπεὶ καὶ πένητα — and **not one of them is Aeolic**. At the rate the poem has been running, one and a half markers were due. Zero arrive."
- **#1476 and #1478**: the hanging line read as the complete poem.
- **#1476**: the Catullan inverse "reaches πένητα independently — derived from reges et urbes run in reverse, not from the Longinus crux".
- **#1749**: ἐπεὶ καὶ πένητα are "words whose attribution to Sappho is disputed".
- **#1756 INV-5**: "The restoration μολπὰ γράμμασι ζώοι is one more filling, which is to say: open."

So the reader's objection 4 lands on a position the author left three months ago. It also misses the developed one, in which the crux is evidence: the seam where the poem ends and Longinus begins.

### 1.5 The Catullan operator (what the report did not reach)

The derivation is developed at length:
- #1052, the fifth erratum, "the hanging line as transmission instrument";
- #576, the Catullus room;
- #1474, decompression: "Catullus decompresses into Latin and inverts. Longinus, if ἐπεὶ καὶ πένητα is his, decompresses into koine and does not invert";
- #1476 on the two routes to the low case.

ChatGPT's challenge, whether the two derivations are independent, is answered in #1476. The low case is reached from reges et urbes reversed, without the crux.

---

## 2. The two readings, point by point

**The conventional reading, at its strength.** The current consensus reading (#1756 INV-8): a love triangle, the man near the girl, the speaker watching and undone. Its forms are jealousy (formed in the eighteenth century, written into the text by Hermann in 1816, consolidated by Page in 1955), erotic seizure without jealousy (Smyth 1900: "not an expression of jealousy"), and the wedding song (Wilamowitz 1913).

**The future-reader reading.** The archive's current form, not the CP text: κῆνος ὄττις is a seat for whoever receives; the poem builds that receiver; the edge is the design.

**Degrees.** **E** explains: the datum follows from the reading. **A** accommodates: compatible, does no work. **S** strains: needs an extra assumption. **—** silent.

| # | Datum | Conventional | Degree | Future-reader | Degree |
|---|---|---|---|---|---|
| 1 | φαίνεταί μοι … φαίνομ' ἔμ' αὔτᾳ: the poem opens and closes on appearing, the speaker in both positions at the end | a frame of perception, the speaker watching, then seeing herself undone | A | the speaker becomes both the one appearing and the one appeared to, the structure of inscription | E |
| 2 | κῆνος, distal | far from the speaker, near the girl | E | points out of the scene to the one who receives | E |
| 3 | ὄττις, indefinite: "whoever" | a rhetorical generalising of the man | S | a position, not a person (#1476): the seat for any receiver | E |
| 4 | ἴσος θέοισιν | the lucky man; or praise of a bridegroom (Wilamowitz) | E | presence with the voice across the mortality line | E |
| 5 | ἐνάντιος … πλάσιον … ὐπακούει: sits facing, near, listening | the man beside the girl | E | the posture of a reader or hearer before a song | E |
| 6 | the addressee given only voice and laughter | Sappho's economy of one or two features (frs. 16, 81) | A | the one addressee given only what is heard (§1.2) | E |
| 7 | τό … ἐπτόαισεν: what "this" is | the sight of the pair; disputed | S | the projected scene of reception | E |
| 8 | the order of failures: voice, tongue, skin, eyes, ears, sweat, trembling, colour | the symptoms of erotic seizure (Longinus 10.1: τὰ συμβαίνοντα ταῖς ἐρωτικαῖς μανίαις παθήματα) | E | the staged withdrawal of the performing body | E |
| 9 | the order as an order | not required by seizure | — | voice fails first because voice is what transmission replaces | E |
| 10 | γλῶσσα ἔαγε, the hiatus at line 9 | admired as mimetic and intended (Nagy) | E | the first of two ruptures (#1475) | E |
| 11 | χλωροτέρα ποίας | pallor; "greener than grass" an awkwardness | S | colour of the substrate (needs papyrus known: §1.3) | S |
| 12 | τεθνάκην … ὀλίγω 'πιδεύης | the lover near death | E | the threshold: the body prepared for transfer | E |
| 13 | ἀλλὰ πᾶν τόλματον: adversative, agentless verbal adjective | a turn to resolve or endure, its agent the speaker (supplied in translation: "I must") | S | the second open position: no one is said to dare (#1749) | E |
| 14 | ἐπεὶ †καὶ πένητα†: zero Aeolic markers in six words (#1475); Bergk brackets them as Longinus's | a crux | — | the seam where the poem ends and the commentator's sentence begins (#1474) | E |
| 15 | the four modern endings (Pearce 1724, Hermann 1816, Welcker 1816, Wilamowitz 1913) each name the man's relation by supplying the close (#1752) | each restorer's reading | A | every receiver decides the poem's completeness; the edge is designed to be filled | E |
| 16 | Catullus 51: stanza 1 translated, 2–4 compressed, stanza 4 a reversal | Roman moralising coda | A | an inverse of the operator at the fifth position (#1052, #1476) | E |
| 17 | Longinus's reason for quoting: selection and binding of extremes (10.1) | the catalogue as craft | E | the maker's craft (#1752) | E |
| 18 | ancient reception: no jealousy consensus; Plutarch reads it as love | holds for the erotic form; strains for jealousy | A / S | holds | A |
| 19 | the medium c. 600: performance; writing known, papyrus traded | performance only | E | the seat is medium-agnostic (#1548); only row 11 and the supplement need writing | A |
| 20 | fr. 147, fr. 55: remembrance in the future | reperformance futurity (Lardinois) | E | the seat that 147's "someone" fills (#1645) | E |
| 21 | the Nestor cup: κῆνον, ὃς δ' ἄν, ἵμερος of Aphrodite, a future user picked out at a distance | — | — | the same three elements, a century before Sappho | E |

**What the table shows.** On the transmitted lines the conventional reading explains the scene: rows 2, 4, 5, 8, 10, 12. It does so by supplying a person where the grammar gives a position (row 3) and an agent where the verb gives none (row 13). It is silent on the order and on the edge (rows 9, 14). Where it is the reading of the ending, it is a supplement (row 15).

The future-reader reading explains the same scene rows and the ones the conventional reading leaves open: rows 3, 6, 7, 9, 13, 14, 16 and 21. It strains at one point, χλωρός as papyrus (row 11), which depends on the material facts of §1.3. Its medium question is otherwise answered by #1548 (row 19).

The decisive rows are 3, 13 and 14: the two open positions, and the seam. The report addressed row 14 only as a defect in the submission, and correctly so. It did not address rows 3 and 13, and it did not carry its own correction at row 14 back into the derivation.

---

## 3. Corrections owed, all verified

| Locus | As submitted | Correction | Source of the error |
|---|---|---|---|
| §I | "The jealousy reading, dominant since Wilamowitz" | Wilamowitz 1913 read a wedding song; the jealousy reading was formed in the eighteenth century, written into the text by Hermann in 1816 ("refinxi") and consolidated by Page 1955 (#1752) | the submission |
| §IV, n. 7 | Renehan, "The Meaning of ΧΛΩΡΟΣ in Homer and in Later Greek Poetry", *CP* 71.4 (1976) 321–25 | the pages fall inside Adkins's article on πολυπραγμοσύνη (*CP* 71.4, 301–27); no such Renehan article is located. v1.0 cited Irwin, *Colour Terms in Greek Poetry* (1974) 31–63. Restore Irwin, and check the "dried and cut" end of the spectrum against Irwin, who reads χλωρός with moisture and freshness. Irwin is not open; declared, not read | first in #134 (May 2026), n. 15, then carried into the submission; #134 needs an erratum |
| §V.2, V.4, VI | "The transmitted πένητα, whatever the following crux proves to be" | the editions obelize †καὶ πένητα†; the crux begins at καί. Take #1474–1476: the six words carry no Aeolic marker, Bergk assigns them to Longinus, and the low case is reached by the Catullan inverse without them | the submission; corrected in the archive 15 August |
| §V.4, stanza line 1 | ἐπεί σε, Κύπρι … σώζεις | σε is accusative; the subject of σώζεις wants σύ (same quantity) | the submission |
| §V.4, stanza line 3 | καὶ πόλιν ὀλίγαν ὑπερέχεις ἄτας | twelve syllables, and – ∪ ∪ at the opening: not a Sapphic hendecasyllable (– ∪ – x – ∪ ∪ – ∪ – x). Line 2, καὶ πένητα καὶ βασιλῆα σώζεις, scans | the submission |
| §V.4, Adonic | μολπὰ γράμμασι ζώοι, called an Adonic | seven syllables; an Adonic is – ∪ ∪ – x. γράμμασι ζώοι alone is an Adonic, and v1.0's γράμμασι μολπὰν was one | the submission |
| §V.4, n. 17 | ζώοι "aorist optative" | present optative of ζώω: let it go on living. The aspect changes the gloss, and the present suits "live on" better than the aorist | the submission (and #576, which builds its σ_S on an aorist optative) |
| §I–II | Bonifazi, the Nestor cup, the Dipylon jug, Mantiklos dropped | restore from v1.0 and add κῆνον on the cup (§1.1) | the revision |
| §I, the stanza line 1 | ἀλλὰ πᾶν τολματόν, position 5 short | inherited from the transmitted line, which the editions mark as unmetrical; say so | the transmission |
| §IV | West 1970, 314, "absurd" | to check against West's text before it is kept | unverified |

## 4. What a revised version would carry

These are the expansions, each drawn from the records above:
- **§I**: the position, not the person (#1476). Bonifazi restored. The Nestor cup with κῆνον as the governing comparandum. A concession: the triangle explains the distance, and what it cannot explain is ὄττις (row 3).
- **§II**: the addressee given only sound (§1.2).
- **§IV**: the material facts of §1.3 (Herodotus 2.135 and 5.58, Iliad 6.168–69, the hydria, Ovid); the medium-agnostic seat (#1548); the illiteracy proposition recorded as unargued (#1752, #1756).
- **§V**: the seam (#1474–1475); the low case by the inverse alone (#1476). The convergence claim then has to be restated. With καί inside the obelus, the grammatical route keeps ἀλλά and πᾶν τόλματον and loses ἐπεὶ καί. The two derivations meet on the turn and the universal, and only the Catullan inverse supplies the comparandum; the stanza repaired in its three metrical points, with γράμμασι marked as the restorer's word; and the reconstruction placed as one more filling (#1756 INV-5) beside the reading of the edge as the poem's design (#1749).
- **A new section**: the comparison table (§2), the conventional reading met point by point.

How much of v1.0's stronger apparatus to restore, and whether the stanza stays in the paper or moves to an appendix as one filling among the four modern endings, are the author's calls.
