This paper draws on Juvenal's intertextual relationship with comedy to solve a textual crux involving fish-names. The monograph by Ferriss-Hill will no doubt warn scholarship away from the treatment of Roman satire's intertextuality with Old Comedy for a time.Footnote ² Yet, Greek comedy's influence on Roman satire is far from exhausted, and this paper will show that this influence goes more widely, and more deeply, than is usually seen. In time, one might hope for a renewed monographic treatment of the subject.

Towards the end of Juvenal's fifth satire, the pathetic client is humiliated—once again. He has endured different crockery (37–48), different table water served by rude staff (49–65), different bread (66–79) and different seafood (80–91). The fish course, likewise, features much more appetizing dainties for the master than for the client: while the master dines on mullet from Corsica, and lamprey from Sicily, for the client (Juv. 5.103–6)

There are a number of problems with line 104 as it is transmitted. No fish are known by the name ‘Tiberinus’; nor are fish known to change colour because of the temperature of the water. The notion that ice was a problem for a fish swimming up a sewer (behaviour attributed to the fish in lines 105–6) was scoffed at by Housman, whose characteristically sardonic remark bears quoting in full: ‘glacie nemini, quantum scio, praeterquam mihi et Schradero et Hadriano Valesio admirationem mouit: ceteris exploratum est frigore pisces maculosos fieri, eos praesertim qui torrentem cloacam, locum frigidissimum, penetrare soleant.’

The opinion of the ceteri are therefore (a) that fish can turn spotty from cold; (b) that sewers are particularly cold, and therefore that fish which swim in sewers are likely to suffer this affliction. A polemical reply to Housman by Bradshaw, in the most able defence of the line undertaken, dismissed unreasonably the link Housman draws between the sewer and the cold: ‘The reason for its being spotty has nothing directly to do with the sewage which made it fat.’Footnote ³

Indeed—but that is not Housman's claim. If it is true that the fish is solitus penetrare cloacam—no mere accident therefore but a regular visitor—then the sewer must be reckoned as one of the cold environments contributing to the fish's condition. Are sewers loci frigidissimi? I read Housman's statement as ironically reflecting the contrary view.Footnote ⁴

Bradshaw's second defence is equally weak: ‘What the rest of us have to be convinced of is, not that cold makes fish spotty, but that Juvenal might have had reason to think that cold made the Tiberinus spotty. The distinction is neither subtle nor trivial.’Footnote ⁵ The distinction is indeed an important one. Yet, Bradshaw fails to produce evidence that anyone in antiquity believed this, let alone that it was what Juvenal believed. He argues, therefore, that not every belief that was current in antiquity is transmitted to us. Also true. But his argument is pinned into a circle as a consequence: the belief about fish growing spotty in the cold is assumed only to explain the text under question. As a result, we are entitled to consider the problem unsolved: we are being asked to take on trust that this was a belief plausible for Juvenal to hold. Furthermore, Bradshaw's claim that his argument is ex silentio is wide of the mark. For it is particularly strange that no fish is said to grow spots in the cold when differences between fish at different times of the year are so often the subject of ichthyological literature of various types (cf. Archestratus, frr. 27, 29, 31–6, 42, 45, 50 Olson–Sens; Xenocrates, fr. 3 Ideler; Damoxenus, fr. 2.14–20 K.–A.). If this literature avoids mentioning the belief reconstructed by Bradshaw, despite having the motive, the weapon and the opportunity to do so, we are entitled to conclude that no such belief existed.

Bradshaw draws on the fact that Galen knew a fish called the Τιβέρινος (de alim. fac. 3.30) to argue that no fish name is lacking from Juv. 5.104;Footnote ⁶ this passage has been alleged by others to indicate that Tiberinus is not an adjective but the proper name of a fish. Yet, the true interpretation of this passage was seen in 1938 by Thompson, who points out that Galen (and Varro; cf. Macrob. Sat. 3.16.12) refers simply to ‘Tiber fish’ in general, and is not naming a species.Footnote ⁷ Bradshaw's defence, which among modern commentators on Juvenal was accepted by Ferguson and anticipated by Duff, can therefore be dismissed.Footnote ⁸ Furthermore, if Tiberinus refers to a fish's contingent location rather than essential identity, then passages referring to the lupus Tiberinus (Hor. Sat. 2.2.31; Columella, Rust. 8.16.4; Xenocrates 6 Ideler; cf. Oribas. Coll. Med. 2.58.9) need not lead us to the conclusion that it is the lupus, Greek λάβραξ, the sea-bass, or pike,Footnote ⁹ which is meant in this passage.Footnote ¹⁰

If Tiberinus is an epithet, not the name of a fish, then we must turn our attention again to the line. Suspicion is ranged against glacie in particular. Housman's sarcasm is quite correct: sewers are not particularly cold, and that fish turn spotty in the cold has been shown above to be an implausible opinion for Juvenal to hold. Solutions to the problem have been extremely varied, as has the willingness of editors to accept conjectures into the text. The modification of glacie has brought in its train the modification of aspersus; editors read sparsus or aspersus depending on the metrical requirements of their decision about glacie. Since the words are synonymous, this need not detain us. I present first a synopsis of suggestions on the text:

The name glanis was proposed on three separate occasions independently, as the only pyrrhic Latin fish name beginning with gl-, but refuted by Thompson on ichthyological grounds: the glanis, it seems, does not swim into estuaries (and a fortiori not into cloacae).Footnote ¹¹ Owen's suggestion is, as he points out, extremely stylistically apt (the repetition is rather typical of Juvenal); yet, it is hard to explain what glacie is doing in the text at all. Owen's notion that glacie was an interlinear gloss hardly satisfies: what was it glossing? If it was supposed to offer an explanation of the maculis appearing on the fish, then we are back in Bradshaw's territory, and can raise the same objection.Footnote ¹² glacus was derived by Thompson from Greek γλάκος, but the lack of attestation in Latin might make one sceptical.Footnote ¹³ Thompson had earlier argued for wholesale deletion; he was followed in this by Knoche, but effectively countered by Adamietz.Footnote ¹⁴ Campbell's idea, glutto, is approved of by Courtney, though it is unclear to me why: is a ‘Tiber glutton’ an obvious way of describing a fish?Footnote ¹⁵ If not, then Courtney seems wide of the mark to say that this ‘meets all criteria’.Footnote ¹⁶ Campbell's argument from intertext (in his case, Lucilius) is however intriguing; the methodology, at least, can be endorsed.Footnote ¹⁷

Two final suggestions lead to my own. Clausen thought of glaucis; although he did not put it in his own text, the suggestion was taken up by Willis in the Teubner edition, and by Braund in her Loeb (though not in her commented edition of Book 1, where she marks a crux). We still lack a fish name. My own solution is therefore to write the fish name glaucus into the text (and thus to write sparsus after it). While working on this article, I found that I had been anticipated by an anonymous contributor to Rose's paper on the passage.Footnote ¹⁸ I take this anticipation to indicate that my thought is correct, or at least plausible. I supplement it now with the intertextual evidence from comedy.

What clinches this solution is that the juxtaposition of glaucus and anguilla—the eel—is a widespread feature in the catalogues of fish in Greek comedy. Since these catalogues indeed name many kinds of fish, it might be thought far-fetched to assume that they point to a particular association between these two species. However, what is significant is that even within these catalogues the eel and the glaucus are found close together. Juvenal, as a reader of comedy (however we might imagine that happening), can have formed a mental association between the two fish.Footnote ¹⁹

The earliest example of this juxtaposition which I have found is from Cratinus (fr. 171.49–50 K.–A.):

This fragment, preserved by Athenaeus, overlaps with a papyrus (PSI 11 1212), which has placed it into larger (if not on that account very much clearer) context. Kassel and Austin ad loc. compare Ar. Av. 716 for the thought: the fish here presented are delicacies to which the speaker is implicitly comparing himself (assuming that this is the parodos and that the chorus of πλοῦτοι is still speaking).

It may well be objected that, if these fish are delicacies, they hardly fit well into our text of Juvenal. Yet, it is clear that, whatever has happened to the fish in our satire, they are not as appetizing as one might wish; this very juxtaposition, of a mouth-watering menu and a disappointing execution, sharpens the point of the lines. Furthermore, other fragments, and ancient commentary upon them, show that the status of these fish is not always the same. Here is Philemon (fr. 82.18–24 K.–A.):

From Eustathius, who quotes 20–1, we learn that the text was understood in antiquity to indicate that the κάπρος, at least, enjoyed high favour (particularly in Argos); by implication, the same must go for the glauciscus (for the morphology, see below) and the eel, said to be the food of the gods themselves. But on turning to Athenaeus, who preserves the fullest version of this fragment (by the standards of comic fragments, a whopper at 26 lines), we discover that there were other opinions (Deipn. 7.288c):

Hikesios says that these [γόγγροι, eels] are the toughest of the eels, spongey-fleshed and rather unnourishing, much inferior in good chyle but good for the digestion.

To illustrate these eels, Athenaeus then quotes Philemon, fr. 82 in full. Hikesios, at least, seems to have differed from Philemon's assessment of the fish's value. His Περὶ Ὕλης (in at least two books) is a fairly frequent source for Athenaeus’ lore on dietary matters (particularly in Book 7, on fish, in which he is cited 26 times). On the whole, Hikesios has a positive view of the value of eels (cf. Ath. Deipn. 7.298b), which makes his critical remarks about γόγγροι all the more surprising and significant.

Philemon refers to the γλαῦκος as the γλαυκίσκος in this fragment; the diminutive morphology is relatively common in biological terminology (cf. ἀστήρ >> ἀστερίσκος)Footnote ²¹ and we probably need not think that this represents a semantically charged diminutive (either appreciative or deteriorative). It is however germane to our Juvenal passage to note that a fragment of Amphis uses exactly the same strategy (fr. 35 K.–A.):

This fragment, once again, shows the grouping together of the γλαῦκος with the eel. It will be objected that the λάβραξ, the very fish that is mostly assumed to lie behind the identity of the Tiberinus of the satire, is also mentioned in this fragment. Yet, the λάβραξ forms a much less regular pairing with the eel than does the γλαῦκος, and is therefore a less suitable candidate for our Juvenal passage than the γλαῦκος. Antiphanes, fr. 130 does not include the λάβραξ at all, but in line 4 we read (fr. 130.4 K.–A.):

The cooking instructions at Antiphanes, fr. 221, however, include so comprehensive a list of fish (eleven types in 8 lines) that we need not worry about the co-occurrence of a λαβράκιον with a γλαυκίδιον (1–2), especially given the presence of a γόγγρος and an ἐγχέλειον (4–5). The collocation looks highly significant to us, because it begins the fragment; we cannot assume, however, that it had an equally striking position in the original context of the text the fragment was taken from.

Aside from the association of the γλαῦκος with the eel (ἔγχελυς, γόγγρος, anguilla), the texts presented here stimulate rather wider reflections on Juvenal. First, although I began my paper with a reference to Ferris-Hill's work on satire and Old Comedy, the texts that illustrate my point come from a rather wider range of comic traditions, representing Old, Middle and New Comedy. Of course, that comic tropes are passed between these texts is nothing to be much surprised at. However, it is revealing that Juvenal's intertextual relationship with comedy is by no means restricted to the Old Comedy ‘triad’ of Aristophanes, Eupolis and Cratinus in the tradition of Hor. Sat. 1.4.1. Indeed, one might think that the closest fragment in content to our satire is Philemon, fr. 82, where the stress on river fish and their unappealing diet (echoed even for sea-fish by Men. fr. 27) is refocussed on fish names found to be appealing. The connection is less paradoxical than it might seem: even fish known on other grounds to be desirable, Juvenal implies, are made disgusting by their association with (a) the river (a trope paralleled in this fragment of Philemon) and (b) the Tiber in particular. This suggests, then, that satire's debt to New Comedy is underexplored. Second, it is revealing that the comic authors on whom I have drawn are not necessarily the best known; Cratinus is less of a surprise than Amphis in this context. Again, the presence of the same trope is no sure guide to Juvenal's reading material. But if we take the parallel with Amphis seriously, what this might point to is that the anthology literature to which we presumably owe what meagre scraps of comedy we have (mediated by Athenaeus et al.) is in the process of developing. In other words, Juvenal's knowledge of Amphis need only be the same as our own; the tradition of the comic fragment has, then, already begun, but remains part of the ‘Old Comedy’ that Juvenal draws on for his own poetic creation.

The fish's fortunes in Latin writers are much less generous: I find only Plin. HN 9.16.25. Yet, that is itself an encouraging sign that this fish was known in the times of Juvenal, even aside from literary reminiscence. Pliny's note may hold the clue to a further difficulty in the passage, though a less crucial one for the constitution of the text: quidam rursus aestus impatientia mediis feruoribus sexagenis diebus latent, ut glaucus, aselli, auratae (‘some, again, lie hidden for sixty days because of their intolerance of the heat at the hottest point of the year, as the glaucus, the haddocks, the gilt-breams’). Presumably the fish is ‘hiding’ in the depths of the river or perhaps even in the mud in the banks or at the bottom. This might explain the difficult expression uernula riparum, if this refers to the fish emerging from the bank after this period of dormancy. Such a fish could reasonably be described as a ‘native of the banks’. Pliny's theory about the fish's behaviour has itself a comic precedent (though to conclude that Pliny was dependent on this comic fragment for his information is presumptuous) in a passage of Damoxenus referred to already in this paper (fr. 2.16–20 K.–A.):

Damoxenus fits into the scientific tradition which accorded a difference between the quality of fish in summer and winter. However, the specific reference to the γλαῦκος in this context is suggestive. The fish which buries itself in river-mud in the hot periods of the year, according to Pliny, seems an excellent candidate for having different qualities of flesh at different times as well, as indicated by Damoxenus. The picture of this fish is consistent with the portrayal of the dinner of the Juvenalian client.

In spite of this consistency in the ancient sources for this paper, it must be admitted that modern scholarship has drawn something of a blank about what kind of fish the γλαῦκος actually was. Thompson, though ultimately despairing of a solution, suggests that there may be two species meant: a kind of shark and the Scomber glaucus.Footnote ²³ The behaviour of the fish in gathering its young into its mouth (Oppian, Hal. 1.749) suggests some kind of mouth-brooder. The glaucus seems to have been a large fish, and both of Thompson's suggestions fit that impression. But the frequent use of diminutives in Greek comedy is highly suggestive (γλαυκίδιον in Antiphanes, fr. 221; γλαυκίσκος in Philemon, fr. 82.21 and in Damoxenus, fr. 2.18; γλαυκινίδιον in Amphis, fr. 35.2);Footnote ²⁴ if these diminutives do denote a fish that is physically small—which is however not at all certain—then a small, Mediterranean mouth-brooder might be a candidate for (one of) the species referred to. These criteria are met by the cardinal fish (Apogon imberbis), which might be a further candidate for the elusive γλαῦκος. As an ocean-going fish, to find it in the Tiber might not augur well for its quality.

The final paragraph represents some speculative thoughts; those better versed in fish-lore, ancient and modern, should take the story up. The aim of this paper was to show how Juvenal's text can be improved by consideration of Greek comic intertexts. Greek comedy is taken to refer to the genre as a whole, not only to the Old Comedy so often claimed as a Greek step-parent to Roman satire: Philemon was as useful to us as Cratinus. Furthermore, Amphis was as useful as Cratinus: the presence of Greek comedy in Juvenal ranges beyond the ‘Big Three’, likely because extract-books of comedy had already been produced, making—in our terms—fragments of unusual texts available more widely. Importantly, this gives us insight into Juvenal's actual textual practice when it comes to his exploitation of Greek comedy, rather than following the tracks beaten by the Roman satirists themselves. This also allows us to flesh out the sketch offered by Ferriss-Hill: Juvenal's reading of Old Comedy is as extensive, perhaps, as Horace's, and must have taken advantage of florilegia as much as complete plays.Footnote ²⁵ The emendation of a passage of Juvenal thus gives us a broad insight into the relationship between Greek and Latin literature, the history of the textual transmission of Greek comedy, as well as deepening our understanding of Juvenal's text.