Adverbs of manner derived from adjectives with the suffix -ως constitute a productive category in Ancient Greek whose development ran parallel to similar morphological formations in other Indo-European languages. This study deals with a very specific sort of -ως ending adverb of manner, one whose special feature is its derivation from participles. In Ancient Greek, this specific type is rather uncommon, as it is in other Indo-European languages, although examples can be found beyond those mentioned in the introduction (Lithuanian, Modern English, and Old High German).[1] As -ως ending adverbs of manner are built on adjectives, those derived from participles necessarily have an adjectival nature. It is, however, not that simple, and as this study makes clear, the verbal side of participles also plays a role in the development of such adverbs. In fact, one of the hypotheses considered in this book is the possible classification of the participle-derived, -ως ending adverbs as converbs by assuming Haspelmath’s definition of converbs as “verbal adverbs”[2] – a hypothesis dismissed by the author.
After an introduction in which the corpus, goals and structure of the book are set out, the morphology of the adverbs retrieved from said corpus is discussed in relation to their chronological development (chapter 1). Participle-derived, -ως ending adverbs are not common: they barely appear in Homeric and Archaic poetry – ἐπισταμένως and ἐσσυμένως being the only examples – but do appear more frequently in Classical and post-Classical prose, where adverbs such as εἰκότως and ὄντως are commonly attested. In all, Mathys has only found 106 different adverbs – most of them with fewer than five attestations – in a corpus that comprises Homer, Archaic poetry, Herodotus, the Hippocratic Corpus, the Attic orators, Thucydides, Xenophon, Plato, the three major tragedians plus Aristophanes, as well as the Argonautica of Apollonius of Rhodes, the Septuagint, and Polybius’ Histories. Of them 58 are derived from perfect middle-passive participles, 36 are derived from present active and middle-passive participles, and 11 from perfect active participles. Other tenses remain undocumented in this kind of formation, except for τυχόντως (aorist active stem), probably created as an alternative form to τυχόν, an adverbialised neuter with its own distinct development.[3] According to Mathys, the higher number of perfect middle-passive participles is due to their suitability for adjectival meaning, as interferences with verbal adjectives in -τος would suggest.
This distribution has led Mathys to separate the syntactic analysis of the adverbs formed from perfect middle-passive participles (chapter 2) from those formed from present and perfect active participles (chapters 3-6). The last two chapters before her conclusions are more general in focus, though limited almost exclusively to adverbs belonging to the second group: chapter 7 deals with some illustrative cases of adverbs that develop propositional uses (εἰκότως, ὁμολογουμένως) and others that are found functioning as intensifiers (διαφερόντως, ὑπερβαλλόντως). In chapter 8 the adjectivisation of the participle is established as a necessary requirement for the formation of adverbs of manner with the suffix -ως.
The presence of a higher number of items in the first group should explain why comparatives and superlatives are found among them only, cf. κεχαρισμενώτατα (X. Eq.Mag. 1.1), πεφυλαγμένως μᾶλλον (Is. 8.97), μάλιστα πεφυλαγμένως (X. An. 2.4.24). However, their syntactic behaviour is almost identical and they do not differ from other adverbs of manner: they specify the way in which the main action is carried out, and can further develop propositional uses like ὄντως ‘really, actually’; occasionally, they can be found modifying adjectives, cf. νὺξ ὑγρὰ διαφερόντως ‘an extraordinarily wet night’ (Pl. Criti. 112a). It is true, as Mathys points out, that the verbal nature of the participial base can license government. But these constructions are rather exceptional and limited to νοῦν ἐχόντως (and its variant λόγον ἐχόντως) which should be analysed as a hypostatic compound, and to ἐχόντως ἑαυτόν in one Platonic passage (Phlb. 64a).[4] There are even some rare cases in which middle and passive meanings can be distinguished for the same adverb, compare Κῦρος ἐπορεύετο ἠμελημένως “Cyrus proceeded carelessly” (X. An. 1.7.20) to οὐδὲ ταύτας ἠμελημένως ἐχούσας “(maids) who also were well cared for” (X. Mem. 3.11.4).
Most of the book deals with what Mathys calls the syntactic integration of these adverbs, particularly regarding the potential overlapping of similar elements, such as circumstantial participles, predicative adjectives and adjective-derived adverbs of manner which are semantically close to them. As for present and perfect active participles, some specific cases of this potential overlap are thoroughly examined, in which adverbs related to impersonal verbs such as δεόντως are contrasted with corresponding accusative absolute constructions (chapter 4); ἑκόντως and ἀκόντως with the adjectives ἑκούσιος, ἀκούσιος and the corresponding adverbs of manner ἑκουσίως, ἀκουσίως (chapter 5); νοῦν ἐχόντως with the participial clause νοῦν ἔχων (chapter 6).
Despite the level of detail in her analyses, the conclusions are somewhat prosaic: participle-derived, -ως adverbs of manner are pretty much like adjective-derived, -ως adverbs of manner. As a matter of fact, their increased frequency in Classical texts coincides with that of all other -ως ending adverbs of manner, which is inversely proportional to the decrease of the adverbial accusative of corresponding adjectives. Regarding their syntactic integration, there is no real overlap with other elements, although some contexts can trigger their use – and even creation – instead of them. One such context involves coordination with other -ως adverbs, another the accumulation of participles with the same head noun.
All these facts lead Mathys to propose that the formation of an -ως adverb of manner from a participle implies its lexicalisation as an adjective. The onus probandi of this hypothesis is put on the use of these very same participles as predicative complements in copulative constructions, although in the case of perfect middle-passive participles it is not easy to distinguish this use from a periphrasis of the type ἀφιγμένοι εἰσί “they have arrived”. It is true that the number of participles taking part in their formation is limited but it still remains unclear how far that lexicalisation goes. In fact, once an adverb was derived from a participle, after it underwent this kind of lexicalisation, the morphological process may have been extended to compounds, as in ἀρκούντως > ἐξαρκούντως, ἐπαρκούντως. But it may also have spread from adverbs built on adjectives, those ending in -μενως or -ντως, to adverbs formed from participles by Reimwortbildung, i.e., contamination between rhyming words, as might be the case of ἡδομένως ‘with joy, gladly’ after ἀσμενῶς ‘gladly, readily’. Finally, the formation of adverbs of manner seems to have spread among semantically close participles, as in ἀγαπόντως ‘gladly, contentedly’ and ἡδομένως.
To conclude, as Mathys herself claims, this is the first study devoted to Ancient Greek -ως adverbs of manner formed from participles. One might question the necessity of a 227-page book on such a narrow topic, but the scientific quality of this study makes its reading highly recommendable as it is an example of thorough and rigorous philological investigation.
Notes
[1] For instance, see Lat. ornātē ‘ornately, elegantly’, meritō ‘deservedly’, audenter ‘boldly, fearlessly’, caesim ‘by cutting, with cuts’, etc.; cf. R. Kühner and F. Holzweissig, Aufürhliche Grammatik der lateinischen Sprache. Erster Band: Elementar-, Formen- und Wortlehre. Hannover: Hahnsche Buchhandlung, 1912, 1004-1013.
[2] Cf. M. Haspelmath, “The Converb as a Cross-Linguistically Valid Category”. In: M. Haspelmath and E. König (eds.), Converbs in Cross-Linguistic Perspective: Structure and Meaning of Adverbial Verb Forms – Adverbial Participles, Gerunds, 1-55. Berlin: Mouton de Gruyter, 1995.
[3] Cf. C. Denizot and S. Vassilaki, “La fabrique de l’éventuel en grec : les fortunes de τυχόν”. In: F. Lambert (ed.), Les futurs grecs et leur histoire, 253-283. Louvain-la-Neuve: Peeters, 2017.
[4] According to Mathys, construction with the genitive or the dative cases is difficult to analyse as it can be assumed to be analogical, e.g. κεχαρισμένως + dat. after ἡδέως + dat., or explained through free uses of these cases. The reader is referred to one of her papers for a detailed discussion of the data, cf. “Adverbes en -ως dérivés de participes et rection verbale en grec ancien”. In: C. Le Feuvre, D. Petit and G.-J. Pinault (eds.), Verbal Adjectives and Participles in Indo-European Languages, 197-215. Bremen: Hempen, 2017. Note that, as Mathys herself recognises, de la Villa considers the origin of this construction to lie in verbal government, cf. “Adverbs as Part of Speech in Ancient Greek”. In: E. Crespo, J. de la Villa and A. Revuelta (eds.), Word Classes and Related Topics in Ancient Greek, 422-423. Louvain-la-Neuve: Peeters, 2006.