ἀκολουθεῖν μετ’ αὐτοῦ
(Antiatt. α 122, Phryn. Ecl. 330, Σ^b α 747 [= Phot. α 789])

D. General commentary

Several entries of Atticist lexica touch upon the syntax of verba sequendi. Verbs meaning ‘to follow’, such as ἀκολουθέω, ἕπομαι, ὀπηδέω, ὁμαρτέω (see K–G, 430–1) and those prefixed with συν‑ (e.g. σύνειμι), are normally constructed with the dative, which expresses association (see Bortone 2010, 113–4); so too are several adjectives indicating relationship and fellowship (for instance, κοινός and ἀκόλουθος). However, prepositional phrases replace the dative on occasion. Given that μετά + genitive is the most common of these prepositional structures, lexica tend to focus on this particular construction. Nevertheless, prepositional phrases with κατόπινκατόπιν or ὀπίσωὀπίσω are also possible: the earliest occurrence of κατόπιν + genitive with ἀκολουθέω is found in Aristophanes (Pl. 13), while it is often used with verba sequendi on its own as an adverb (on κατόπινκατόπιν as the Attic equivalent of ὄπισθεν, see Moer. κ 62Moer. κ 62). Moreover, although μετά – when used in a comitative sense – is typically construed with the genitive, the dative may also occur (see e.g. Hes. Op. 230: μετ’ ἀνδράσι λιμὸς ὀπηδεῖ ‘[neither] hunger follows men’). The prepositions employed differ in various respects: for example, while μετά is widely attested, the usage of ὀπίσωὀπίσω is almost entirely restricted to the SeptuagintSeptuagint and Christian commentaries thereon (see below; for the use of improper prepositions as a typical post-classical phenomenon, see Bortone 2010, 180–1).

The replacement of the dative by prepositional phrases + the genitive in verba sequendi constructions is likely to have originated in Pre-classical Greek (see Bortone 2010, 155–6; Wackernagel 1928, 218, for the instrumental function) as a means of emphasising the idea of accompaniment (for which, in Attic Greek, μετά + genitive gradually replaced σύνσύν + dative, see DELG, s.v. μετά, and Bortone 2010, 114, 154 and especially 156, where he duly talks about a ‘reinforcement of the case with a redundant preposition’; on the semantic difference that originally existed between σύν and μετά, see Bortone 2010, 166–7, with bibliography). Attic shows a decline in the use of the dative in prepositional phrases, whereby it tends to be preserved only after prepositions that cannot take any other case and is otherwise replaced with the genitive or, increasingly, the accusative (see Bortone 2010, 154–5). This phenomenon foreshadows the predilection for prepositional phrases and the decline of the dative, which are conspicuous features of the koine. According to Horrocks (2010, 97, commenting on Polybius’ language), in middle- and high-register koine, the dative is often replaced with prepositional phrases, particularly in comitative expressions.

The verba sequendi’s trajectory thus aligns with a wider trend in koine Greek, whereby the dative is increasingly eschewed in favour of prepositional phrases. Papanastassiou (2007, 611) dates the replacement of ‘the dative denoting instrument or means’ with μετά + genitive to the 4th century BCE. The texts quoted below appear to suggest that, around the same time, the same replacement also occurs after verba sequendi, with some early examples dating to the 5th century. The chronological sequence may be reconstructed as follows: (1) plain dative expressing a comitative meaning; (2) prepositional phrase with the dative; (3) prepositional phrase with the genitive. All three options already coexisted in classical Attic (on such coexistence in the specific case of verba sequendi, see Bortone 2010, 155–6). On more general issues concerning the demise of the dative, see Humbert (1930); on the overlap between dative and genitive in personal pronouns, resolved at the expense of the dative, and on the interchangeability of the dative and genitive in papyriPapyri, see Horrocks (2007, 628–9) and Stolk (2017), respectively.

Lexica typically acknowledge verba sequendi to be a coherent group characterised by common syntactic features, as evidenced by the fact that Σ^b α 747 (A.3) treats ἀκολουθέω and ἕπομαι together and discusses the constructions associated with both. Similarly, the scholium to Ar. Pl. 823 (B.1) draws a parallel between Aristophanes’ use of the expression ἕπου μετ’ ἐμοῦ and Pl. Mx. 249d.6, where the verb employed is ἀκολoυθέω, demonstrating that the scholiast was interested not merely in the syntax of ἕπομαι exclusively but in the syntax of the verba sequendi as a group.

Atticist lexica show no consensus regarding the legitimacy of prepositional phrases with the genitive in place of the dative. The construction of ἀκολουθέω with μετά + genitive is defended by a gloss in the Antiatticist (A.1; see Valente 2015, 50), which argues that the construction is Attic on the basis of an example from Lysias (fr. 61 Carey = C.1). The Synagoge’s gloss (A.3) is on the same page and broadens the group of Attic authors who set a precedent for the use of this construction. These include Menander (C.4, C.5), who is generally considered by Atticist lexicographers to be representative of a later, post-classical Attic. Alpers posits that the source of Σʹʹʹ for this entry is a fragment of Orus (fr. B 7)Orus fr. B 7, while de Borries argues – considerably less convincingly – that it derives from Phrynichus’ PS (fr. *115)Phryn. PS fr. *115 (see below). Valente (2015, 50) agrees with Alpers’ identification and claims that a direct derivation of the Σʹʹʹ entry from the Antiatticist is unlikely in any case. In his Eclogue (A.2), Phrynichus adopts a stricter Atticist line, proscribing the prepositional phrase and recommending instead the traditional use of the dative. That entry (A.2) appears to respond directly to the Antiatticist (A.1: on the use of the Antiatticist in the second book of the Eclogue, see Sicking 1883, 5; Latte 1915, 374; Valente 2015, 52–4).

The entries in the Antiatticist (A.1) and in the Eclogue (A.2) attest to two antithetical standpoints in an Atticist debate on the admissibility of ἀκολουθέω (et similia) with μετά + genitive in place of the more usual construction with the plain dative. Both have their starting point in Lysias’ use of the construction with μετά + genitive while ignoring passages from authors who are certainly well-regarded by the Atticists (notably, Aristophanes). We cannot discount the possibility that these glosses echo a specific Hellenistic viewpoint on the authenticity of Lysias’ orations. Works by Alexandrian scholars are recognised as primary source material for the Antiatticist (see Valente 2015, 32–3), and it is not unlikely, therefore, that the Atticist debate on syntax originates from an Alexandrian source with philological interests. The possibility that the construction attested in Lysias’ passage is spurious (νοθεύειν) is openly expressed in Phrynichus’ entry (A.1), which speculates as to whether the construction might constitute a syntactical error on Lysias’ part. This entry also highlights several of the methodological principles on which Phrynichus bases his prescriptions, including the importance of correctly identifying authentic texts and the fact that, despite the value of the canon, even Attic authors are not infallible but rather should be evaluated critically (see entry βρέχει). Another criterion that Phrynichus uses to assess a form’s acceptability is its rarity: his condemnation of lexical hapax legomena is particularly noticeable in the Eclogue (see entry βρέχει). The same criterion seems to be applied to syntax in Ecl. 330 (A.2), where the use of μετά + genitive is described as a καινὸν σχῆμα and a ξένη σύνταξις. Phrynichus concludes that, regardless of the authenticity of Lysias’ usage of ἀκολουθέω μετά τινος, such a construction should be avoided because it is uncommon. On the contrary, it is noteworthy that, while the expression ξένη σύνταξις does not occur elsewhere, Phrynichus (PS fr. *5 [= Σ^b α 243, Phot. α 101, ex Σʹʹʹ])Phryn. PS fr. *5 (= Σ^b α 243, Phot. α 101) is said to recommend the construction ἀγανακτῶἀγανακτέω + genitive in place of ἀγανακτῶ + dative because of its uncommonness. The expression καινὸνκαινός σχῆμα thus appears applicable to uncommon uses when they are both perceived as inappropriate and regarded as innovative (for Phrynichus’ doctrine on ἀγανακτῶ + genitive, see entry ἀγανακτῶ σου).

Phrynichus is not mistaken in classifying the use of the prepositional phrase with ἀκολουθέω as unusual. The usage of μετά + genitive in such contexts, although well attested in Classical Greek, is in fact rare compared to the use of the dative. In classical texts, μετά + genitive is primarily attested in prose, occurring in (among others) Thucydides (with ἀκολουθέω in 7.57.9, with ἐφέπομαι in 7.52.2), Plato (with σύνειμι in Smp. 195b.4, Lg. 639c.4–5; with ἀκολουθέω in La. 187e.1–2, Mx. 249d.6, with ἕπομαι in Phdr. 250b.7–8), and Lysias (with ἀκολουθέω in 2.27, 12.12, fr. 61 Carey [C.1]). While σύνειμι used with μετά + genitive in Eur. El. 943 (see F.1) appears to be a unique exception in tragedy, prepositional phrases with σύνειμι (Aristom. fr. 2.3, Eub. fr. 8), ἀκολουθέω (Antiphan. fr. 120, Men. Dis exapaton 59–60, Men. fr. 293 [C.5]) and ἕπομαι (Men. Dysc. 969 = Mis. 466 = Sic. 423 = fr. 903.21, C.4) occur more frequently in comedy, though they remain less common than constructions with the dative.

Aristophanes’Aristophanes work warrants closer examination. As noted by Willi (2003a, 253), the dative continues to exert a robust presence in Aristophanes, although, in some instances, a switch to other cases may indicate a weakening. In the vast majority of those instances, Aristophanes regularly uses the dative with ἕπομαι (occurring circa 25x), ἀκολουθέω (circa 16x) and σύνειμι (25x), but prepositional phrases are noticeably more frequent in his Plutus, where we find μετά + genitive used with σύνειμι (l. 504) and ἕπομαι (l. 823), and κατόπινκατόπιν + genitive with ἀκολoυθέω (l. 13) and perhaps ἕπομαι (l. 1209; note that Wilson here prints the dative τούτοις in place of the genitive; see also εἰσέρχομαι κατόπιν τινός at line 1094). Poultney’s (1936, 174–7) study of Aristophanes’ use of the genitive notes that the comitative μετά + genitive is ubiquitous with intransitive verbs of motion. Specifically regarding the peculiarity of the Plutus, Willi (2003b, 49) interprets the use of prepositional phrases with συνίεμι and verba sequendi as one of the syntactic changes attested in the play, differentiating its language from that of Aristophanes’ other works: a ‘sign of lateness’ and a first indication of the decline of the dative. Regarding ἕπομαι κατόπιν τινός, it occurs in Ar. fr. 509: ἐρείδετον, κἀγὼ κατόπιν σφῷν ἕψομαι (‘Move forward, you two, and I will follow you’, on which see Bagordo 2020, 62–3).

Phrynichus’ aversion to the use of prepositional phrases in such contexts, which was rarer in classical Attic but common in koine Greek, may be a reaction to the fact that it had also begun to appear, albeit only occasionally, in Atticising authors (see below). Concerning the koine, it is worth noting that in the SeptuagintSeptuagint, which clearly shows a dative that is in decline and increased use of prepositional phrases (see Horrocks 2010, 107–8, Bortone 2010, 177–94), the use of μετά + genitive with verba sequendi is still sporadic and dependent on the verb with which it occurs. In both the Septuagint and the New TestamentNew Testament, the most common verbum sequendi is ἀκολουθέω, which spread at the expenses of ἕπομαι (the latter occurring only twice; see DELG, s.v. ἀκόλουθος) and of both ὀπηδέω and ὁμαρτέω, which had already fallen out of use. ἀκολουθέω takes a preposition + the genitive only in a handful of occurrences out of a hundred cases, while the simple dative remains common. ἀκολουθέω ὀπίσω + genitive prevails in the Septuagint (R. 3. 19.20.3 + 3x) and also occurs in NT Ev.Matt. 10.58; while ἀκολουθέω μετά + genitive, which does not occur in the Septuagint, can be found in the New Testament in Ev.Luc. 9.49 and Ap.Io. 6.8, 14.13. σύνειμι, by contrast, is used only used with the dative. In Biblical HebrewHebrew, ‘to follow’ is often literally conveyed by the verb meaning ‘to go after’: the Hebrew Bible's influence on the Septuagint – and, to a degree, the New Testament – may thus have affected the spread of prepositional syntax.

In the imperial period, ἀκολουθέω, ἕπομαι, and σύνειμι are occasionally constructed with μετά + genitive by authors who generally write in ‘good’ Attic and regularly use the dative in the vast majority of cases. For example, Lucian uses that prepositional phrase with ἕπομαι (Nav. 1.12), ἀκολoυθέω (Nav. 30.2) and σύνειμι (DDeor. 10.2.11); ἕπομαι occurs once with μετά + genitive – again in place of the dative – in Aelius Aristides (10.34: δεῖ γὰρ οἶμαι μετὰ τῶν ἄλλων ἕπεσθαι, ‘it is necessary, I think, [that you] follow the others’).

E. Byzantine and Modern Greek commentary

High-register Byzantine texts imitate classical syntax and thus routinely construe verba sequendi with the simple dative. In some instances, however, prepositional constructions with the genitive are also used. In his translation of Ovid’s Metamorphoses, for instance, Planudes systematically uses ἀκολουθέω + dative, but he constructs the verb with κατόπινκατόπιν + genitive at least once in his letters (see Maximus Planudes Ep. 99.45: ἀλλ’ οὐκ ἐκεῖνος κατόπιν τῆς σῆς ἠκολούθησε δόξης, ‘but he did not follow your doctrine’). Nicetas Choniates, for the sake of variatio, uses both the dative and a prepositional construction in the same sentence, juxtaposing ἀκολουθέω ὀπίσω + genitive with συνακολουθέω + simple dative (see Hist. 250.213–6: στείλας τριήρεις, ὅσαι ὀπίσω τοῦ μεγάλου δουκὸς ἠκολούθησαν, καὶ τῶν συνακολουθησάντων αὐτῷ ἐκ τῶν κατὰ πάροδον χωρῶν στρατιωτικῶν καταλόγων ἅπαν ἐπίλεκτον πόλεμον, ‘after dispatching triremes, which followed behind the megas doux, and after enrolling a whole elite troop of soldiers who followed him from the neighbouring regions’). The same applies to σύνειμι, which Photius, for instance, regularly uses with the dative while also twice constructing it with μετά + genitive (e.g, in his lexicon, σ 804: συνουσία· τὸ συνεῖναι μετ’ ἀλλήλων, ‘συνουσία: the [fact of being] one together with one another’). Anna Comnene uses this construction once (10.6.4: αὕτη ἡ φήμη τοὺς μετὰ τοῦ Πέτρου συνόντας καταλαβοῦσα δεινῶς συνετάραξε, ‘this report, reaching those who were together with Peter, upset them terribly’), but the dative is otherwise the rule in her Alexiad.

In the lower registers of Medieval Greek, by contrast, the dative is almost entirely absent (see Horrocks 2010, 337, 350, regarding Digenes Akritis and the 12th-century Chronicle of Morea). The transitive use of ἀκολουθέω, which will later become the standard construction in Modern Greek, is widespread: see, for instance, the Chronicle of Morea 3283 (μὲ ὅσους τὸν ἠκολούθησαν ‘with all those who followed him’, although different redactions in this passage fluctuate between accusative and genitive) and 8986 (νὰ ἠκολουθοῦν τὰ φλάμουρα, ‘to follow the banners’). The replacement of the dative in Medieval Greek is not a monolithic process but rather varies geographically with respect to extent and speed. The genitive and the accusative vie with one another in replacing the dative: in northern regions, verbs such as ἀκολουθῶ and βοηθῶ and those with συν- prefixes are assimilated to the transitive paradigm, and the dative they originally took is thus replaced by the accusative. Southern regions, by contrast, make these verbs transitive at a later stage and tend to replace datives with genitives, with or without prepositions (see CGMEMG vol. 3, 1951–2, 1957–8). In Modern Greek, the northern dialects, unlike southern ones and Standard Modern Greek, use the accusative to mark indirect objects, i.e. in replacement of the ancient dative and instead of the modern genitive or prepositional phrases (for instance, σε δίνω, instead of the standard σου δίνω ‘I give you’).

In Modern Greek, μετά + genitive is no longer in use for comitative and instrument but survives only in rare idioms, such as μετὰ βίας (‘hardly’): on μετά in Modern Greek see Bortone (2010, 207) and Luraghi (2005, 152–3). ακολουθώ is still in use with objects only in the plain accusative, without any prepositions; the same applies to its compound παρακολουθώ (‘to follow’ in the sense of ‘to attend’), which is also common in Modern Greek. έπομαι also survives but as a learned katharevousa form artificially reintroduced. Its use is rare and limited almost exclusively to the 3rd persons and to the present tense; it is usually intransitive and, when it does take an object, the object will be in the genitive (see, for instance, the saying μια συμφορά έπεται της άλλης, ‘one misfortune follows another’); it never takes prepositional phrases.

F. Commentary on individual texts and occurrences

(1)    Eur. El. 941–4 (C.6)

Euripides’ use of μετά + genitive after σύνειμι is remarkable as an exception in tragedy, where prepositional phrases do not occur in the construction either of this verb or of verba sequendi. This passage, although notable, has received little commentary, and it is not listed among Euripides’ colloquialismsColloquial language by either Amati (1901)    or Collard (2005; 2018). Stevens alone (1937, 187), in discussing περί + accusative, notes that Euripides uses prepositional phrases and the accusative with prepositions more than the other tragedians do and describes this feature of Euripides’ style as ‘relevant rather to the consideration of the prosaic than the colloquial element in his language’.