χθές, ἐχθές, χθιζός, χθιζινός, χθεσινός
(Moer. χ 6, Phryn. Ecl. 294, Phryn. PS 127.9, Philemo [Vindob.] 393.35–394.4)

D. General commentary

Atticist scholarship discusses the admissibility of χθές and ἐχθές, i.e. the monosyllabic and disyllabic forms, respectively, of the adverb ‘yesterday’ (A.1, A.4, B.7), as well as of its adjectival derivativesDerivatives χθιζός, χθιζινός, χθεσινός (‘of yesterday’) (A.1, A.2, A.3, B.4, B.7).

The adverb (ἐ)χθές ‘yesterday’ is an inherited Indo-European word (cf. Latin heri, Vedic hyás, Old High German gesterēn, etc.). The cognate languages point to a form with an initial consonant cluster, variously reconstructed as *ǵʰ-di̯es (Schindler 1977, 34), *ǵʰh₁-di̯es (Vine 2008), *dʰǵʰi̯es (Puhvel 1987), or *dʰǵʰés (Kloekhorst 2014, 44–6). The variation ἐ-/zero in Greek is therefore not a phonological phenomenon, but probably the result of the addition of the same deictic particle ἐ- seen in the demonstrative adjective/pronoun (ἐ)κεῖνος ‘that’ (see DELG, EDG s.v.; NIL 69–70 s.v. *dei̯-). From the root of (ἐ)χθές were derived several adjectives meaning ‘of yesterday’ were derived. The morphology of the oldest form, χθιζός, attested since Homer (see below), is not entirely clear. If it derives from the adverb χθιζά ‘yesterday’, also Homeric, the latter may in turn derive from the zero grade *χθισ- (provided that the root of (ἐ)χθές contained an *-i̯-) plus the adverbial suffix -δα (cf. e.g. κρύβδα ‘secretly’; see Puhvel 1987, 318 n. 11). If, instead, the adjective is primary and χθιζά represents its adverbial neuter plural (as suggested by the fact that the singular χθιζόν is also used adverbially in the direct tradition of Il. 19.195), its formation nevertheless remains unexplained. At a later stage, another adjective, χθιζινός ‘of yesterday’, was formed in Attic by adding to χθιζός the suffix -ῐνός to χθιζός (reanalysed from adjectives in -νό- based on locatival forms in -ι, e.g. ἔαρι ‘in the spring’ > ἐαρινός ‘of spring’). The suffix became specialised in forming temporal adjectives and remained productive in this function into Modern Greek (see Chantraine 1933, 200–1; Bologna, Dedè 2021, 11–2). Given the variation between χθές and ἐχθές, a variant ἐχθιζινός could easily have arisen alongside χθιζινός. Later still we find the further variants χθεσινός and ἐχθεσινός, which are clearly analogically based on (ἐ)χθές.

The adverb (ἐ)χθές is attested from the archaic age, its first occurrence being in the Homeric Hymn to Hermes (C.2) in the form χθές. Attic literature provides a rich harvest of occurrences of (ἐ)χθές in both comedy (Eupolis: χθές 3x; Aristophanes: χθές 8x, e.g. C.4, ἐχθές 8x, e.g. C.3) and prose (among others, Thucydides: χθές 3x; Xenophon: χθές 3x, ἐχθές 2x; Plato: χθές 33x, ἐχθές 1x; Demosthenes: χθές 6x, ἐχθές 2x). Although most authors appear to use both forms indiscriminately, the monosyllabic χθές clearly prevails; a case in point is Plato’s language, in which ἐχθές is almost always avoided, as noted by B.6. ἐχθές, by contrast, seems to be preferred by authors of Middle and New Comedy, where it outnumbers the occurrences of χθές (cf. Anaxandr. fr. 41; Antiph. fr. 276; Crobyl. fr. 5; Alex. fr. 287; Euphro fr. 9, in which ἐχθές is metrically required). In Biblical Greek one always find ἐχθές, although manuscripts often oscillate between the two forms.

The adjective χθιζός (‘of yesterday’) is attested from Homer onwards (cf. e.g. Od. 2.262: ὃ χθιζὸς θεός, ‘yesterday’s deity’). In the Homeric poems, in which χθές does not occur, χθιζός is also used adverbially as an equivalent of the adverb (cf. C.1), including in the adverbial neuter plural χθιζά (cf. Il. 2.303: χθιζά τε καὶ πρωΐζ[α], ‘yesterday or the day before’). As Phrynichus, not without reason, warns his reader (A.2), χθιζός later becomes typical of Hellenistic poetry: see its use in a Theocritean epyllion (25.56), in Callimachus (e.g. Epigr. 14.2 Pfeiffer), and in Apollonius Rhodius (e.g. 4.1397). In the imperial period it occurs, among others, in Julia Balbilla’s Epigr. 30.1 in the Aeolic form χθίσδος. Conversely, its absence from the diction of classical prose is striking, the only exception being Herodotus 1.126.21: εἰσὶ ὑμῖν πόνοι τῷ χθιζῷ παραπλήσιοι ἀναρίθμητοι, ‘countless toils, similar to those of yesterday, await you’. Occurrences of χθιζός in prose increase in the imperial age. Aside from Plutarch, who uses it several times (cf. e.g. Quomodo quis suos in virtute sentiat profectus 75e.5), Atticising writers also do not disdain it, despite Phrynichus’ warning: both Lucian (cf. e.g. Herm. 1.12: τὴν γὰρ χθιζὴν συνουσίαν, ‘yesterday’s lesson’) and Aristides (28.2 Keil = 49.492.4 Dindorf: ὁ χθιζὸς ἡμῖν ἑστιάτωρ, ‘our host of yesterday’) employ it, as B.7 notes. Literary occurrences of χθιζινός are scanty: in Classical Greek they amount to two Aristophanic passages (C.5, Ra. 987), where the form has been restored by editors in place of χθεσινός for metrical reasons (despite MacDowell’s attempt (1971, 172) to defend the manuscripts’ reading; see Austin 1973, 133‒4). χθιζινός is later employed by Galen and Alciphron; otherwise it occurs only in erudite sources. In the form ἐχθιζινός, its unique literary occurrence is in Menander (C.6), which most likely is the locus classicus commented on by Photius in B.3. The analogical χθεσινός is first attested in Lucian (C.7), but its occurrences become more frequent in late antique literature: it is employed, among others, by Basil of Caesarea (Ep. 42.1.30), Apollinaris of Laodicea (Epistula ad Dionysium 1.7), and John Chrysostom (MPG 49.68.7).

Atticist scholarship takes an interest in (ἐ)χθές and its derivatives for several reasons. Moeris’ concern (A.1) is with the longer form ἐχθές, which he rejects as non-Attic. In making this dialectal attribution, Moeris follows a grammatical tradition that regarded ἐχθές as the earlier and more complete form, and χθές as an Attic peculiarity – see B.5, where ἐχθές is ascribed to users of common Greek: οἱ κοινοί – derived by a process of aphaeresis supposedly characteristic of this dialect. This theory, attributed by Apollonius Dyscolus to Tryphon (B.1), was supported with an etymologyEtymology tracing ἐχθές back to the adverb ἐκτός ‘outside’, with an exchange between plain voiceless and aspirated stops. A similar doctrine recurs in many later scholarly sources (besides B.2 and B.5, cf. schol. Dion.Thr. 97.16; for the etymology, cf. Orio 162.10–1, Epim.Hom. χ 15 = Hdn. Περὶ παθῶν GG 3,2.182.2–22). A competing etymology deriving ἐχθές from ἐκθεῖν ‘to run out’ (B.6, Orio 162.10–1, Schol. (Tz.) Ar. Pl. 881) likewise assumes that the form with ἐ- was the original one. Whereas in the pre-Atticist grammatical tradition – and also among non-Atticist 2nd-century-CE grammarians such as Apollonius and Herodian – an Attic form such as χθές could be felt to be irregular with respect to analogy, and therefore negatively evaluated (see Probert 2011, 275), in the eyes of Atticists such as Moeris the same dialectal peculiarity made it preferable. This strict position is questioned by Philemon (A.4), later followed by Thomas Magister (B.7). Philemon notes that such intransigence would paradoxically lead to the exclusion of Aristophanes himself from the number of Attic authors (see further F.1). A tolerant position is also expressed in an entry of Photius’ lexicon, possibly dependent on Orus (B.4), where χθές, χθιζινός and ἐχθές, ἐχθιζινός are all considered ‘Greek’, i.e. linguistically correct. Although Phrynichus does not focus on the problem of the prothetic vowel, his discussion of only forms with initial χθ- may implicitly suggest that he, like Moeris, rejected those with ἐ-.

A point of disagreement between Phrynichus and Moeris concerns instead the acceptability of the poeticPoetic language χθιζός, which Moeris accepts, perhaps owing to his notion of Homer as a ‘proto-Attic’ author (see entries ἥρῳ and οἶσε). By contrast, Phrynichus – who proves to be particularly attentive to the forms’ stylistic appropriateness – advises his audience to avoid χθιζός as belonging too markedly to poetic (A.2) and Homeric (A.3) diction (as indeed confirmed by its distribution, see above). To achieve a more πολιτικός effect ‒ that is, among other things, greater suitability for rhetorical speech and urban eloquence; on Phrynichus’ use of πολιτικός in the context of stylistic evaluation, see Tribulato (2025, 200–11) ‒ Phrynichus instead recommends the later χθεσινός (A.2), while also admitting χθιζινός (A.3). Interestingly, achieving a πολιτικός style was also the purpose which, according to Phrynichus, motivated Aristophanes’ choice of words, in which the epic χθιζός has no place. In the form in which it has come down to us, Phrynichus’ entry transmits a corrupted version of Aristophanes’ text, since the line does not scan unless one restores χθιζινός in place of χθεσινός (in both C.5 and Ra. 987, see above). A couple of entries in the Synagoge (B.8, B.9) attest to the existence of an abusive use of both χθιζός and χθεσινός meaning ‘slow’, ‘useless’.

E. Byzantine and Modern Greek commentary

As seen above, Byzantine erudition recognises both ἐχθές and χθές as correct (B.4, B.7) and both forms are in use in Byzantine literature. Nonetheless, authors writing in the more classicising literary register show a clear preference for the Attic χθές, which is the only form in which the adverb is used, for instance, by Anna Comnene and Michael Choniates. χθεσινός likewise remains in continuous use throughout the Byzantine period in both literary and documentary sources. Whereas it frequently occurs in works written in mixed and low register, including the Digenis Akritis (cod. E 849, cod. Z 5.1840), χθεσινός is only sporadically used by Atticising writers, who appear instead to prefer χθιζός in both prose and poetry (e.g. among others, Michael and Nicetas Choniates). Notably, Nicephorus Gregoras mostly employs it in the expression χθεσινὴ κραιπάλη, ‘yesterday’s drunkenness’ (e.g. Antirrhetica priora 249.19), which quotes Lucian’s passage (C.7).

In Medieval Greek, (ἐ)χθές underwent manner dissimilation of the cluster χθ (fricative + fricative) into χτ (fricative + stop), yielding (ἐ)χτές (see CGMEMG vol. 1, 185–6, 188–9). Note that the dissimilation (ἐ)χθές > (ἐ)χτές is attested as early as the 2nd century CE (see P.Oxy. Hels. 33.22 = TM 24974), although this may represent an isolated case due to the influence of a local variety. Indeed, the dissimilated spelling does not appear to surface in literature until a rather late period, close to the Humanist age, whereas literary texts that more closely reflect the spoken variety still show the non-dissimilated form ἐχθές (cf. e.g. Digenis Akritis cod. Z 3.738, Bellum Troianum 8489).

χθες/χτες is still the standard adverb for ‘yesterday’ in Modern Greek. As regards the adverb’s spelling, both forms with -χθ- and -χτ- are retained; on a more general level, the preservation of consonant clusters with two fricatives at the beginning (or in the middle) of a word is typical of Katharevousa, while Demotic phonology adopts the dissimilated spelling. In the case of the doublet χθες/χτες, however, the conservative spelling is far more widespread according to the data available in the Hellenic National Corpus (for their distribution in written texts, see Mikros, Gavrielidou, Lambropoulou, Doukas 1996, and Tseronis 2002). The noun το χτες ‘yesterday’ is also in use, as is the case with other adverbs of time (cf. αύριο, το αύριο ‘tomorrow’). The adjective χτεσινός/χθεσινός is likewise used in Modern Greek (e.g. το χθεσινό ψωμί, ‘yesterday’s bread’); cf. also the compound χθεσινοβραδινός/χτεσινοβραδινός (‘of yesterday evening’).

F. Commentary on individual texts and occurrences

(1)    Philemo (Vindob.) 393.35‒394.4 (A.4)

This entry in Philemon’s lexicon, transmitted only by cod. V, is the longest preserved fragment of the iambic trimetersIambic poetry in which the work was originally composed (see entry Philemon, Περὶ Ἀττικῆς ἀντιλογίας τῆς ἐν ταῖς λέξεσιν); as such, it offers a glimpse into the style and structure of this lexicon before its abridgement (for a comparable example, see the entry ἐρυγγάνω, ἤρυγον, ἐρεύγομαι, ἠρευξάμην). The first line expresses the same prescription of χθές against ἐχθές that is found in Moeris (A.1); however, the following lines challenge this view, pointing out in a sarcastic tone that such a prescription would render even Aristophanes a non-Attic author, since he used ἐχθές in a line of the Clouds (C.3), which is then quoted in full (we follow Reitzenstein in restoring ἡμῖν, unanimously transmitted in the MSS of Aristophanes, for ἐκεῖ of cod. V, but it cannot be ruled out that Philemon read this line in a corrupted form). As Brown (2008, 190; 228) observes, Philemon’s polemical stance towards ‘overzealous Atticists’ is reminiscent of the Antiatticist; indeed, the practice of invoking the authority of canonical authors such as Aristophanes in defence of a form of disputed acceptability is also paralleled in the Antiatticist (on the latter’s use of Aristophanic quotations, see Fiori 2022). The lexicon’s polemical tone is unsurprising given the title Περὶ Ἀττικῆς ἀντιλογίας τῆς ἐν ταῖς λέξεσιν (On Attic Controversy about Words), recorded by Choerob. in Heph. 183.3 Consbruch. The language of Philemon’s iambics is itself Atticising, echoing canonical authors such as Aristophanes and Plato. Tellingly, the formulaic expression (οὕτως) ὄναιο ‘may you (so) profit’, i.e. ‘bless thee’, a colloquialism expressing ‘exaggeration or emphasis’ (Collard 2018, 137; apparently not recognised by Reitzenstein, who placed a question mark after οὕτως), is attested in Attic drama (Soph. OC 1042; Eur. Hel. 1418, IA 1008, 1359; Or. 1676‒7; Ar. Th. 469; Men. Pc. 400) and oratory (D. 28.20), before being revived in the 2nd century CE by Lucian (Philops. 27.7, Icar. 20.3, Merc.Cond. 34.3, D.Meretr. 4.1.3, Symp. 23.10) and Aulus Gellius NA 14.6.5: ὄναιό σου […], doctissime uirorum, ταύτης τῆς πολυμαθίας et librum hunc opulentissimum recipe nil prosus ad nostras paupertinas litteras congruentem (‘Most learned of men, may you profit from this erudition of yours, and take back this very opulent book, which is entirely unsuited to our poor learning’). As in Philemon, Gellius employs the expression with a sarcastic force, disparaging another scholar’s learning. The rare diminutive τεχνύδριον ‘petty art’, likewise with disparaging force, comes from Plato (R. 475e.1‒3: τούτους οὖν πάντας καὶ ἄλλους τοιούτων τινῶν μαθητικοὺς καὶ τοὺς τῶν τεχνυδρίων φιλοσόφους φήσομεν;, ‘Are we going to refer to all of these as philosophers, as well as others keen on learning of any sort, even those who practice minor crafts?’) and is not attested again until the imperial period (Clem.Al. Strom. 1.3.22.2, 1.8.41.2). The expression Μουσῶν στόμα ‘mouth(piece)/voice of the Muses’, on the other hand, is not attested before the Hellenistic age, appearing first in Theocritus (7.37). It is frequent in epigrams as a celebratory epithet applied to several poets (Homer, Pindar, Stesichorus; see Skiadas 1965, 84‒5; Ercoles 2013, 597; Valerio 2016, 261). Lastly, the future περιελεῖς from περιαιρέω is late (LXX+), but cf. Ar. Eq. 290: περιελῶ σ’ ἀλαζονείαις, ‘I’ll harass you with quackeries’, where περιελῶ is now generally taken to derive from περιελαύνω, although it may have been understood as the future of περιαιρέω by some ancient authorities, as suggested by the schol. Ar. Eq. 290a (VEΓ³ΘΜ), where the two explanations are apparently confused. Note also that the imperative περίελε occurs twice in cod. V (393.34, 394.26), making it likely that this verb formed part of Philemon’s critical vocabulary, although it is not used by other Atticists.