εἰσιέτω, ἰέτω
(Phryn. Ecl. 141)

D. General commentary

Phrynichus (A.1) reproaches the sophist Publius Hordeonius Lollianus for using the analogical imperative εἰσιέτω in place of εἰσίτω. The analogical forms ἰέτω/‑ιέτω and ἰέτωσαν/‑ιέτωσαν are the product of hypercorrectionHypercorrection, presumably triggered, as Phrynichus plausibly suggests, by the expansion of the analogical infinitiveAnalogy ἴναι at the expense of the correct form ἰέναι (see ἴναι).

The analogical forms of the imperative of εἶμι are sub-standard throughout the history of Ancient Greek, and they are almost unattested in classical and post-classical texts (literary and documentary). Thus, Lobeck (1820, 16) concluded that the few instances found in the manuscript tradition should be removed. Such an approach is surely advisable for classical texts. Editors of Xenophon’s Anabasis rightly emend the variant reading ἰέτωσαν in An. 1.9 into ἴτωσαν. In the case of the Hippocratic treatise De morbis 2 (C.1), Littré alone does not emend the transmitted προσιέτω, while Jouanna (1983, 150) and Potter (1988, 194) correctly restore προσίτω and regard προσιέτω as a Byzantine slip. Not only are the analogical forms of the imperative entirely absent in classical literary sources, but they are also unattested in extant documentary texts, such as papyri and inscriptions. The third-person imperative, singular or plural, is mostly used in official texts: an element of vernacular language such as ἰέτω/‑ιέτω and ἰέτωσαν/‑ιέτωσαν would have been inadmissible in such contexts. Phrynichus’ prescription (A.1) that ἰέτω is unacceptable in higher Greek is therefore confirmed by the available documentation, although the question of what use Lollianus may have made of εἰσιέτω is destined to remain unsettled (see F.1). Things become more complicated when we move to later Christian sources, where the normalisation of the analogical ἰέτω/‑ιέτω and ἰέτωσαν/‑ιέτωσαν into regular forms is not always recommendable. For instance, the occurrence of εἰσιέτω in a text such as the apocryphal Acts of John (C.2) is acceptable, in that it is likely to be an element of genuinely informal and vernacular language fully in keeping with the linguistic register of this writing. On the other hand, Origenes is a more complicated case. In his corpus, the analogical form of the imperative (C.3) occurs only once, whereas the regular forms occur 11x. Since Origenes is a highly erudite author, one may be tempted to normalise the one isolated analogical form of the manuscript tradition into ἀπίτω, thus considering ἀπιέτω a Byzantine slip.

To conclude, while ἰέτω/‑ιέτω and ἰέτωσαν/-ιέτωσαν are out of place in classical and post-classical mid- to high-brow texts, be they literary or documentary, the analogical imperatives should be retained in the sources for which there is evidence of a more vernacular register, such as early Christian literature. It remains, however, unclear where one should draw the line, and whether the alternation of regular and analogical imperatives can be allowed for in the work of educated Christian writers or whether one should normalise the analogical forms, restoring the regular conjugation. Such a problem also remains unresolved in some of the Byzantine sources (see E.).

E. Byzantine and Modern Greek commentary

Byzantine texts tend to bring to an extreme the tendencies already discernible in Ancient Greek. While the analogical imperatives ἰέτω/-ιέτω and ἰέτωσαν/-ιέτωσαν are entirely foreign to texts written in the higher register, they are extremely common in monastic regulations, as well as in juridical and didactic writings, which make no pretence to elevated language. Let us consider two exemplary cases. In the Geoponica, the analogical imperatives are attested three times (C.6, in addition to 12.20.5 εἰσιέτω and 17.13.1 προσιέτωσαν), while the regular forms ἴτω and ἴτωσαν are unattested. In Alexius Aristenus’ commentary on the Synopsis canonum, the analogical forms are attested three times in the canons on which Alexius comments (Conc. 5, canon 22, line 2 ἀπιέτω; Conc. 11, canon 38, line 2 προσιέτωσαν; Conc. 13, canon 46, line 4 ἐξιέτω). The regular forms are also found, and they occasionally alternate with the analogical forms in the same text (Conc. 2, canon 19, line 2 προσίτωσαν; Conc. 13, canon 41, line 2 εἰσίτω; Conc. 13, canon 41, line 3 ἐξίτω; Conc. 13, canon 48, line 2 εἰσίτω; schol. Basil. Magn. canon 75, line 1 εἰσίτω). In writings like the Geoponica and the Synopsis canonum, the analogical imperatives are an aspect of informal language that we should not regard as surprising or unwarranted. Even though the analogical forms occasionally alternate with the regular ones in the same writings, we may confidently retain both and allow for a lack of homogeneity. In support of this conclusion, one may note that in erudite sources the analogical form ἰέτω is also used as the interpretamentum of ἴτω (B.1), which confirms that, for the Byzantines, analogical forms like ἰέτω had become an established, and surely more intelligible, alternative to the regular imperatives ἴτω/-ίτω and ἴτωσαν/-ίτωσαν.

The situation becomes more complex when analogical imperatives occur in texts written in a higher variety of Greek (an early example being ἰέτω in pseudo-Dionysius the Areopagite, C.6), and especially when they alternate with the regular forms. In such cases, it is doubtful whether the analogical forms are copyist slips or whether they are different forms that the writers saw as legitimate. Some instances are particularly noteworthy. Regarding John Chrysostom, Migne does not emend εἰσιέτω and προσιέτω into the regular forms (C.4, C.5), even though εἰσίτω and προσίτω are attested 10x and 18x respectively in the rest of John Chrysostom’s writings. Along the same lines, only one example of the analogical imperative is attested in the writings of Theodorus Studites (C.8), while the regular forms are attested 8x. Finally, Matthew Blastares alternates on more equal terms the standard and analogical forms of the imperative, and the distribution of these options apparently depends on the individual compounds (προσίτω and προσίτωσαν in Syntagma alphabeticum α.5.11, κ.28.6, and κ.28.97 Plotes–Ralles, ἐξιέτωσαν in Syntagma alphabeticum α.5.6 Plotes–Ralles and Syntagma alphabeticum ε.28.15 Plotes–Ralles). In these cases, modern editors generally avoid normalising the analogical forms, no matter how isolated, into the regular ones. However, the analogical imperatives represent a clear minority in John Chrysostom and Theodorus Studites. John Chrysostom and Matthew Blastares are learned, Atticising writers whom we would not expect to admit the unclassical ἰέτω/-ιέτω and ἰέτωσαν/-ιέτωσαν. Further investigation into the language of these writers may prove fruitful.

F. Commentary on individual texts and occurrences

(1)    Phryn. Ecl. 141 (A.1)

Phrynichus’ main aim is to poke fun at Lollianus’ (alleged) linguistic incompetence. Lollianus    was a prominent and successful figure in Athenian political life and in the sophistic circles during Hadrian’s time (see Philostr. VS 1.23 and Schissel 1927). Not only was Lollianus in charge of administering Athens’ grain supply, but the emperor Hadrian also assigned to him one of the two recently founded chairs of rhetoric in Athens. Additionally, during his life Lollianus was a successful and well-paid teacher, and after his death he was honoured with two statues, one of which was inscribed with an honorary epigram (IG 2².4211 [Athens, 2nd century CE]). Phrynichus’ teasing of Lollianus is paralleled by a scoptic epigram attributed to Lucian (AP 11.274), in which the sophist is depicted as a know-it-all who never stays silent and might even hyperbolically claim to be teaching Hermes how to do his duty as ψυχοπομπός.

The analogical imperatives ἰέτω/-ιέτω and ἰέτωσαν/-ιέτωσαν are sub-standard in classical and post-classical texts (see D.). We should then ask how reliable Phrynichus’ gloss may be regarding Lollianus’ use of εἰσιέτω. Indeed, εἰσιέτω is a markedly different case compared to the other glosses in which Phrynichus reproaches Lollianus’ mistakes, which are standard post-classical forms with a very large diffusion in high-register Greek (see Ecl. 140Phryn. Ecl. 140 on Lollianus’ use of accusative τὰς νῆας in place of τὰς ναῦς and Ecl. 152 on the pluperfect ending -εσαν in ἐδεδίεσαν).

We can envisage three possible scenarios: (1) Lollianus’ εἰσιέτω is a learned but incorrect hypercorrection. This may sound odd, but one might compare, e.g., the gloss (Ecl. 182Phryn. Ecl. 181) in which Phrynichus criticises those who believed that, based on analogy, the metaplasm ὤτοις was a legitimate alternative to ὠσίν as the dative plural of οὖς, ὠτός (see slip ὤτοις); (2) εἰσιέτω is an occasional slip of the tongue Lollianus makes, e.g., while giving a public speech, and his blunder becomes notorious. One may compare how the 5th-century Athenian actor Hegelochus famously mispronounced γαλῆν instead of γαλήν’(α), an episode mentioned several times in Old Comedy (see Ar. Ra. 303–4, Sannyr. fr. 8, Stratt. frr. 1 and 60); (3) Lastly, εἰσιέτω is a scribal mistake in one of Lollianus’ writings, which Phrynichus took as though it were Lollianus’ own. With (2) and (3) εἰσιέτω would be a sub-standard form and therefore a slip into vernacular language. The distribution of analogical imperatives in ancient literary and documentary sources (see D.) strongly suggests that (2) and (3) are the most likely alternatives.