In a typical question-and-answer format, first an unknown speaker questions a shell about who took it from the sea and how it became a dedication; then the shell itself responds regarding who dedicated it (Dionysius, son of Protarchus) and to whom it was dedicated (the nymphs of the grotto), and indeed how it came to be dedicated (the straight threw it upon the shore).

The first half of the epigram’s opening line is a matter of concern on a textual critical level. The Palatinus and the Suda have the unmetrical εἰν ἁλὶ λαβύρινθε. εἰνάλιος λαβύρινθε is found in some apographs; a nominative adjective in place of the vocative is uncommon but can be paralleled.Footnote ² Seelbach as well as Gow and Page prefer the correction εἰνάλι’ ὦ λαβύρινθε found in several other of the apographs.Footnote ³ Since it is the lectio difficilior, I accept this reading. The fact that in miniscule ω and ος are easily confused palaeographically makes plausible a scribal correction from εἰνάλι’ ὦ to εἰνάλιος. Moreover, ὦ commonly attends vocative addresses and makes clear earlier in the hexameter the dialogue form of the epigram. Although the adjective creates infelicity with ἐξ ἁλὸς in line 2, repetition seems rather to be a hallmark of the epigram (see below).

The subsequent characterization of the object, namely that it is quarry from the sea, makes it patent that a seashell is meant, or something very much like it. Still, the representation of the shell as labyrinthine requires comment: what kind of shell ought to be imagined? Gow and Page simply note that ‘probably any convoluted shell (and possibly its gastropod owner) might be so called’.Footnote ⁴ Seelbach similarly notes that it must be understood as a spiral-shaped shell.Footnote ⁵ This might seem obvious to the modern mind and perhaps it did to the ancient mind too; after all, gazing into a spiral shell can present a ‘room’ with no end in sight. Notably, however, there is little ancient support for an explicit connection between spiraliform shells and labyrinths.Footnote ⁶ In the iconographic tradition, the concentric spirals of the ancient Cretan maze can be readily evidenced.Footnote ⁷ Yet it is difficult to identify any sure cases of shells represented on a plane that highlights their concentric spiral cross-section: typically, spiraliform shells are represented in profile, and so resemble horns or cornucopia, or their cavity faces the viewer.Footnote ⁸

It is also difficult to evidence in texts. Gow and Page go on to observe that ‘the word [that is, λαβύρινθος] does not occur elsewhere in this connexion’.Footnote ⁹ Later lexicographical works do record a direct connection between the labyrinth and the shell. Hesychius glosses the labyrinth thus: λαβύρινθος· κοχλιοειδὴς τόπος (‘labyrinth: spiral-shell-shaped place’, Hsch. λ 33 Latte–Cunningham). It is possible that this is the first surfacing in the extant record of a long-standing connection between the labyrinth and the shell: some of Hesychius’ material at least is drawn from Pamphilus’ first-century c.e. lexicon.Footnote ¹⁰ The Suda, however, not only takes over the Hesychian gloss, but it also then quotes the first couplet of Theodoridas’ epigram, introducing it as follows: σημαίνει δὲ καὶ τὸ τῶν Ἀντριάδων Νυμφῶν ἀνάθημα (‘it [sc. the labyrinth] also signifies the dedication of the grotto-nymphs’, Suda λ 11, line 4 Adler). On three other occasions the Suda references a word and quotes part of the epigram (Suda α 346 [ἀγρέμιος], α 2766 [Ἀντριάσι] and π 952 [πελώριος] Adler). Hesychius’ lexicon, as it survives, does not exhibit any knowledge of the epigram in his glosses on those same words. The epitomization of his work has affected many of the entries, and authors and works are named only sporadically.Footnote ¹¹ It is possible that Hesychius’ entry originally made reference to Theodoridas. However, there is little evidence that Hesychius consulted epigram collections.Footnote ¹² Thus, whereas Hesychius’ gloss suggests that in later antiquity the connection between a spiral shell and a labyrinth could at least be made in the abstract, the Suda’s references to Theodoridas’ epigram equally suggests that it stood out as the most readily available text associating the shape of the shell with the labyrinth. If there was a common association beyond Theodoridas’ epigram, then it has left little trace in the earlier visual or written record.

Moreover, Gow and Page as well as Seelbach view the shape of the shell as conflicting with its geographical source.Footnote ¹³ It is identified as a gift from ‘sacred Pelorias’ (ἐξ ἱερᾶς … Πελωριάδος, 4). Cape Pelorias is at the north-eastern tip of Sicily on the straits of Messene (Polyb. 1.11.6). There was a temple to Poseidon on the promontory (Diod. Sic. 4.85), which may explain the adjective ἱερᾶς (4). It was known for its mussels, according to the Sicilian Archestratus (fourth century b.c.e.) as cited in Athenaeus: Μεσσήνῃ δὲ πελωριάδας στενοπορθμίδι κόγχας (‘[take] Pelorias clams at Messene with its narrow straits’, fr. 7.4 Olson–Sens = Ath. Deipn. 3.92d). Theodoridas specifically puts the reader in mind of this location by describing how ‘the strait spat [it] out’ (ἐξέπτυσε πορθμός, 5). Archestratus and Athenaeus (Ath. Deipn. 1.4d) use the term κόγχαι, which covers both bivalve shells and shells notable for their capacity (LSJ s.v. κόγχη I.1–2): some of these may have been spiral-shaped.

Pelorias may also refer to the nymph of that name, which would likewise be supported by the adjective ἱερᾶς at line 4.Footnote ¹⁴ So, while the shell is the actual gift of Dionysius dedicated to the nymphs of the grotto, it is also the gift of Pelorias, the strait from which it comes. At the same time, it can be understood metaphorically as the gift of the nymph Pelorias. This nymph is depicted on coins together with fan-mussels though not, as far as I can discover, with spiraliform shells.Footnote ¹⁵ Therefore, on current evidence, even when Pelorias (the location or the nymph) was connected with shells, their spiral shape was not in view. The connection to Pelorias by itself is not sufficient to explain the shell’s characterization as a labyrinth. An explanation is required that appreciates the epigram’s uncommon characterization of the shell as labyrinthine and the fact that it is a marked contrast given the source of the shell.

One piece of evidence which has not been sufficiently discussed in relation to this epigram is the story told about Daedalus and his flight from Minos and the Cretan labyrinth and Daedalus’ arrival on Sicily.Footnote ¹⁶ There, Daedalus goes to Camicus, the Sicanian capital on Sicily, and to the court of Cocalus, its king, who conceals him. However, Minos in his hunt for Daedalus arrives at Cocalus’ court and promises a gift to whoever is able to pass a thread through a shell. Cocalus accepts the challenge but delegates it to Daedalus. He attaches a thread to an ant and then bores a hole through the shell to allow the ant to pass through it. Since he solves Minos’ challenge, Minos detects that Daedalus is in fact at the court of Cocalus and demands that he be handed over. Versions of the narrative can be found in the mythographical works of Apollodorus (Epit. 1.14–15) and Zenobius (Zen. vulg. 4.92 = Paroemiogr. 1.112.9).Footnote ¹⁷ This sequel reverses the more popular myth of Daedalus building the labyrinth, with Minos gifting a new challenge now to a king, Cocalus, whose very name evokes ‘seashells’ (κοκάλια, cf. Arist. Hist. an. 528a9). Whereas Daedalus’ ingenuity enabled him to construct the labyrinth to keep the Minotaur in and he was only foiled with Theseus’ use of Ariadne’s thread to guide him out of the labyrinth, here the shell is a pre-existing object and Daedalus’ ingenuity is in employing the self-same strategy of threading which Theseus had used against him. This myth makes the shell a miniature version of the labyrinth.Footnote ¹⁸

Was this myth available to Theodoridas writing in the third century b.c.e.? Sophocles’ Camicians appears to have dealt with the same myth as recorded by the mythographers. A tantalizing fragment reads: ‘… of this seashell, child, if we could discover someone …’ (ἁλίας στραβήλου τῆσδε, τέκνον, εἴ τινα | δυναίμεθ’ εὑρεῖν, fr. 324 Radt). The address to a child is typically taken to be Cocalus addressing one of his daughters, perhaps before presenting the shell to Daedalus. The challenge of discovering someone by means of a seashell specifically accords with the later sources for the myth: both Apollodorus and Zenobius refer to it as a κόχλος or κοχλίας (Epit. 1.14–15; Zen. vulg. 4.92.58–65). That the solution offered in the play made a connection between the shell and the labyrinth would find support if Nauck’s supplement to the fragment were accepted; he prints in the second line <ὅς διείρειεν λίνον> (‘who might pass the thread through’, Soph. fr. 301.2 Nauck) on the basis of the similar language found in Apollodorus and Zenobius. The supplement is attractive. The similar language would support the argument put forward, although not proven, that Apollodorus drew from the hypothesis to Sophocles’ play.Footnote ¹⁹ But here there is a risk of spiralling into circular reasoning. Nevertheless, it is difficult to think what other challenge could have been offered by means of a seashell. It is highly probable that Sophocles’ play recounted the myth of Daedalus threading a shell and that this recalled on some level his construction of the labyrinth.

Given the lack of clear parallels for a shell on the model of a labyrinth elsewhere, it is plausible that Theodoridas’ collocation evokes this myth. The opening couplet’s question ‘who found you’ can thus also be understood as directed at the reader, prompting them to consider the origins of the shell-as-labyrinth and perhaps who ‘discovered’ or even ‘invented’ it (see LSJ s.v. εὑρίσκω II–III). The lemma identifies the epigram as follows: Θεοδωρίδα ἐπὶ κοχλίῳ θαλασσίῳ ‘Theodoridas, on a seashell’. Despite the Messenian location for the shell, it was interpreted by the lemmatist as a spiral shell in the same manner as Apollodorus and Zenobius.

One reason why this myth would be known and would appeal to Theodoridas is because Sicily is probably his homeland. Athenaeus cites Theodoridas as hailing from Syracuse (Ath. Deipn. 15.699e), he identifies his use of the Syracusan term τήγανος to refer to a casserole dish (Ath. Deipn. 6.229a–b), and Pollux identifies his use of a Tarentine word (Onom. 10.187) which at least supports a background in Magna Graecia. Moreover, his epigram in Anth. Pal. 6.222 (= 3520–3 Gow–Page, HE) on the ribs of some unknown sea-creature called a σκολόπενδρα, washed up on the shore and dedicated to the gods by sailors, locates the transformation of marine detritus into divine gift in South Italy, ‘on the headland of Iapygia’ (Ἰαπύγων … ἐπὶ σκοπέλους, 2).Footnote ²⁰ Theodoridas is likely to have been Syracusan.

The epigram evidently brings Sicilian matters into focus. Whether Pelorias is read as a place or as a nymph, it figures the shell as Sicilian. There is no explicit mention of where the shell washes up, although somewhere in Magna Graecia is likely; it is a fair assumption that the shell would not have travelled too far along the coast before being thrown up on the shore. Indeed, it has not yet been noted that the mention of ‘sleek grotto-nymphs’ (λιπαρῶν … Ἀντριάδων, 6) in the final verse brings to mind Lipara, the largest of the Aeolian islands—just a shell’s throw away from Sicily and Cape Pelorias in particular. One might possibly view the shell as a metaphorical gift from a local nymph dedicated to the collective cave nymphs situated in a nearby area, a divine form of gift exchange that would explain the lack of focus on the dedicator, Dionysius.Footnote ²¹

Leveraging the myth of Daedalus and characterizing the shell as labyrinthine would further situate it within a Sicilian context and such a connection is developed when the epigram calls the strait σκολιός (5) which spat out the shell. It has been proposed that the etymology of λαβύρινθος is derived from a root *labyr meaning a ‘coil’ or a ‘curve’.Footnote ²² Moreover, something which has not been noted either in Gow and Page or in Seelbach is that the adjective σκολιός had been used by Callimachus in reference to the Cretan labyrinth (σκολιοῦ λαβυρίνθου, Callim. Hymn 4.311), a poet with whom Theodoridas is undoubtedly in dialogue (see below).Footnote ²³ Callimachus is the first to use the term λαβύρινθος of the Cretan labyrinth in extant sources.Footnote ²⁴ If this is not simply due to the fortunes of manuscript transmission, then it is possible that Theodoridas is alluding to Callimachus’ novel coinage. At any rate, later, in the first century b.c.e., Dionysius of Halicarnassus observes how Thucydides has the Athenians respond in the Melian dialogue ‘more crookedly than a labyrinth’ (λαβυρίνθων σκολιώτερα, Dion. Hal. Thuc. 40). His use of the phrase demonstrates a certain popularity to the idea of a labyrinth as crooked. Not only does the opening characterization of the shell evoke a myth connecting Daedalus’ labyrinth to Sicily, then, but the epigram then transfers its crooked shape and winding form on to the Sicilian landscape.

A focus on Theodoridas’ striking association between the shell and the labyrinth calls for a reconsideration of the poem in relation to other shell epigrams. It has long been observed that Theodoridas’ epigram can be set in productive dialogue with Callimachus’ epigram on a nautilus shell dedicated by one Selenaia to Aphrodite Zephyritis (1109–20 Gow–Page, HE = Ath. Deipn. 7.318b):Footnote ²⁵

Theodoridas recalls Callimachus’ epigram in several ways. Most immediately, both epigrams depict a shell’s reuse. Here Theodoridas and Callimachus are in the company of two third-century b.c.e. epigrams from the Milan Posidippus (AB 11 and 12) and two epigrams from the Palatine Anthology, one anonymous and of unclear date but likely Hellenistic (3896–9 Gow–Page, HE = Anth. Pal. 9.325), the other of Imperial date attributed to a certain Quintus (3358–63 Gow–Page, GP = Anth. Pal. 6.230). Whereas the anonymous epigram follows Posidippus in the ecphrastic epigrammatic tradition, commenting on how the shells once dwelled in the sea but have now been refashioned as works of art, Theodoridas and Quintus follow Callimachus in exploiting the dedicatory associations of epigram and in having the shells turned into dedications for diverse deities (ἀνέθηκεν, 3524 Gow–Page, HE; ἄνθετο, 3526 Gow–Page, HE; and θῆκε, 3363 Gow–Page, GP; cf. ἄνθεμα, 1110 Gow–Page, HE). Callimachus’ nautilus shell is dedicated to Aphrodite at Alexandrian Zephyrium (in the recently constructed temple there for Arsinoe) and Quintus’ ‘trumpet shell’ (κήρυκ’, 3360 Gow–Page, HE) to Apollo upon the ‘Bithynian coastland’ (Βιθυνίδος … χώρης, 3358 Gow–Page, HE) by Damis the fisherman. By contrast, Theodoridas has Dionysius dedicate his shell to the cave-nymphs as a group (although it may also be a gift from the nymph Pelorias).

Theodoridas’ characterization of the nymphs as ‘sleek’ (λιπαρῶν, 6), however, forges a link to Callimachus’ address to a ‘sleek goddess’ (λιπαρὴ θεός), Calm,Footnote ²⁷ who produces the ‘sleekness’ of an unruffled sea surface.Footnote ²⁸ Further connection is made by Theodoridas’ παίγνιον Ἀντριάδων in a final clause in the second half of the final pentameter which echoes Callimachus’ παίγνιον Ἀρσινόη, also in a final clause, in the same position in his fourth pentameter. While Damis’ dedication is in exchange for protection—‘that he see Hades without disease’ (νούσων ἐκτὸς ἰδεῖν Ἀίδην, 3363 Gow–Page, HE)—and is highlighted by the shell’s characterization as ‘well protected … with natural spikes’ (φρουρητὸν … αὐτοφυεῖ σκόλοπι, 3361 Gow–Page, HE), the epigrams on the shells dedicated by Selenaia and Dionysius do not reflect on the kind of benefit received in return (indeed, there is none requested in Theodoridas), but instead emphasize the close connection between divine dedicatee and the playful function of the dedicated object. That being said, in Callimachus’ epigram there is considerable focus on the dedicator, whereas in Theodoridas’ epigram the focus is on the nymphs. Thus, while epigrams on shells are an emerging genre at the time during which Theodoridas is writing (and continue to be written afterwards, too, given the evidence of Quintus), there are intimations that he is responding to Callimachus’ epigram on the nautilus shell specifically.

Accepting the myth of Daedalus and the shell as operative in the epigram fleshes out the intertextual relationship Theodoridas fashions with Callimachus’ epigram. As several commentators have noted, Callimachus’ epigram takes a particularly natural historical approach to the dedicated shell. The shell describes its own self-propulsion either by raising its sails or by rowing with its feet, an account that accords with several other ‘scientific’ descriptions of its movements (Arist. Hist. an. 6222b5; Ael. NA 9.34; Plin. HN 9.103).Footnote ²⁹ Totally without parallel, though, is the claim that the halcyon would typically lay its eggs within the recesses of the living nautilus. Whether this is Callimachus’ invention or he is drawing from a lost source, Kathryn Gutzwiller is surely right to connect this to the themes of marriage and female reproductive capacity that would be so resonant for a young woman dedicating the shell to Aphrodite ‘of good delivery’.Footnote ³⁰ Theodoridas’ ‘labyrinth of the sea’ evokes a rather different natural historical perspective. This shell is not a sea creature like the nautilus but rather its hard exoskeleton. There are no bird eggs in the shell’s ‘chambers’ (ἐν θαλάμῃσιν, 9) that are hatching from out of it, but an ant has instead made its way through the divisions of the shell which, on the labyrinth analogue, may likewise be understood as rooms or chambers. The myth further serves to explain the tubular perforations between the walls separating the chambers of spiral shells. Daedalus’ success in treating a shell like a labyrinth can be viewed as an aetiology for the supposedly perforated, chambered spiral shells that one can discover washed up on any beach. If the Daedalus myth is recalled in Theodoridas’ epigram, it underscores human ingenuity and intervention in the natural world, whereas Callimachus’ living nautilus shell had evoked only collaboration within the natural order.

Theodoridas’ ‘labyrinth of the sea’ also employs travelling shells to a different geopoetic end. Callimachus’ image of the nautilus able to travel under its own propulsion and carry the halcyon’s young over great distances is mirrored in the dedicatee Selenaia’s travels from Smyrna to Egypt and to the temple at Zephyrium, where Arsinoe is worshipped as Aphrodite. The natural history of the object and its spatial movements together work to aestheticize and valorize the waterborne mobility enabled by the Ptolemaic empire.Footnote ³¹ Conversely, the image of the labyrinth evokes a paradox of movement: a confused wandering within a highly restricted and bounded space. In a sense, this is the movement that Theodoridas’ epigram depicts: the shell is from Pelorias, possibly dedicated to nymphs near Lipara, and has a shell form associated with a myth about Sicily. It is even from a strait that itself is evocative of the labyrinthine. One might say that, whereas Callimachus’ epigram makes the shell a symbol of travel, transformation and a networked Mediterranean, for Theodoridas the labyrinthine shell, straying only within the local Sicilian area—an area where it gained its current natural form, if my aetiological reading of the myth is correct—emblematizes travel and wandering of a much more restricted, local and insular form.Footnote ³²

Furthermore, there is the double designation of Theodoridas’ shell as a plaything (3 and 6). This amplifies Callimachus’ singular identification of the nautilus as a plaything (8). Callimachus’ use of the term παίγνιον has often been taken as a marker of metaliterary reflection, namely that the epigram is a ‘game’ or ‘poetic trifle’ as much as the dedication is a ‘toy’ (LSJ s.v. παίγνιον I and III.1 and 3).Footnote ³³ In Theodoridas’ epigram, the oral connotation of a strait that ‘spat out’ (ἐξέπτυσε, 5) the shell further encourages the reader to parallel the object and the verbal artefact about it. Adrien Gramps has recently stressed the importance of the παίγνιον as περίσκεπτον in Callimachus’ epigram; on the one hand, the toy is ‘admired’, while, on the other hand, the literary ‘game’ is one of examination and speculation (LSJ s.v. περισκοπέω II.1–2). This accords well with an epigram which through its display of natural historical knowledge allows the reader to probe the nautilus shell with ‘an inquisitive gaze’.Footnote ³⁴ No such inquisitiveness is elicited about Theodoridas’ shell. Nevertheless, Jan Kwapisz has proposed that it still constitutes a literary παίγνιον of sorts in that the whorls of the shell are reflected in the repetition of παίγνιον Ἀντριάδων (6) at the line end after the line initial παίγνιον Ἀντριάσιν (3) and of εἰνάλι’ and ἐξ ἁλὸς in the opening distich.Footnote ³⁵ The verbal responsion could also be viewed as appropriate to a shell dedicated to cave nymphs: caves are notorious for generating echoes, and shells in their own small way echo the sound of the sea when held to the ear.Footnote ³⁶

This distinction between the two epigrams’ use of παίγνιον is sharpened by the characterization of the shell as a labyrinth. Having identified a λαβύρινθος as a κοχλιοειδὴς τόπος, Hesychius’ entry continues: λέγεται δὲ ἡ λέξις ἐπὶ τῶν φλυάρων, παρὰ τὸ πολλοῖς κύκλοις λόγων κεχρῆσθαι (‘the phrase is used for silly talk, derived from the use of many circles of speech [or: of many circular arguments]’, Hsch. λ 33 Cunningham). In Plato’s Euthydemus circular reasoning is compared to ‘falling into a labyrinth where, although we think we have now come to the end, we turn the corner again and appear to be at the beginning of the enquiry’ (ὥσπερ εἰς λαβύρινθον ἐμπεσόντες, οἰόμενοι ἤδη ἐπὶ τέλει εἶναι, περικάμψαντες πάλιν ὥσπερ ἐν ἀρχῇ τῆς ζητήσεως ἀνεφάνημεν ὄντες, 291b7–9). This is not an exact comparandum, since Hesychius’ gloss seems to be about silliness rather than about faulty logic, although Socrates’ use in passing does highlight the ease with which a link to illogical or formally ill-fitting or unexpected speech could be made. However, the Suda adds: ‘said about silly talk. [derived] from not reaching the door’ (λέγεται δὲ ἐπὶ τῶν φλυάρων. παρὰ τὸ μὴ λαβεῖν θύραν, Suda λ 11, lines 1–2 Adler). If this folk etymologizing also underwrites Plato’s use of the architectural analogy, then the connection between silly talk and labyrinths would be old indeed. If the gloss has at least a Hellenistic pedigree, then Theodoridas’ use of λαβύρινθος to describe the shell which is also a παίγνιον can be understood as a response to Callimachus’ epigram. While Callimachus infuses his epigram with natural scientific language and thought, Theodoridas describes a shell the natural shape of which symbolizes nonsensical verbal articulation. Theodoridas suggests that, if a dedicatory epigram on a shell is truly to be a poetic παίγνιον, then the physical form of the shell ought also to reflect that playful, riddling and non-serious literary form. And Theodoridas the Sicilian offers up a shell with a markedly Sicilian provenance that provides just the opportunity for the intertwining of dedicatory epigrammatic subject matter and literary form in a playful key.