The use of dual forms in Hesiod: Observations about its gradual dwindling in the Theogony and in Works and Days                                                                                         

Ruprecht-Karls-Universität Heidelberg

The considerable quantitative disparity between the occurrences1 of dual forms in the Theogony and in Works and Days becomes outright evident by means of a closer examination of the data: in Works and Days almost three times as many dual forms as in the Theogony are employed: such a peculiarity seems to be not completely accidental and likely depends on the different stadiums of the language´s evolution which reflects Hesiod´s specific stylistic decision, correspondingly to the respective themes and addressees of the poems.

The author of the Theogony appears not to employ the dual in an active way for he uses this numerus only where a plural would not be metrically2 suitable. On the contrary, if the option between dual and plural is open, then will the author mostly opt for a plural.3

In Works and Days the quantity of dual forms in proportion to the verses is by a third higher than in the Theogony4. Furthermore, in Works and Days the choice between dual and plural seems to be made for metric reasons in four cases only5: where a dual occurs, almost always6 also a plural form would be possible and vice versa.

Additionally it is possible to assert that the dwindling of the dual appears to follow the same path as in Attic Greek7: its decomposition starts with the verb and is then extended to nominal forms.  In fact, the duals in these poems are mostly nouns8: only three occur in a verbal form9, two of them in the Theogony.

The mixture of dual and plural forms10 in the same passage, which occurs mainly in Works and Days, seems, through the concept of "Zweiheit"11 or through the so called "Dualité-Unité"12 not to be clarified enough13. This combination is perhaps better explainable by researching about the process, how traces of dialects persist in the correspondent high language and in which manner the lower version of the dialect tends to dwindle14. Actually the substitution of dual through plural progressively increased with the time, until the dual, approximately at the time of Alexander the Great, died out15.

Regarding the diverging quantities and distributions of dual forms in the two poems, it is possible to assume that the difference of topic and register plays a central role in this sense, for these are dissimilarly influenced and vary according to the respective contexts and addressees.

In fact, the dual is unanimously estimated16 as a product of a primitive society, which perceived and conceived the world and humans mostly as groups or couples17: eventually, this perception would, with the civilisation of the society, progressively disappear18. Although the use of the dual can be interpreted as an interesting signal of the stadium of the language´s evolution19, it is not necessary for this reason to postulate a different authorship for the two poems20, as, according to the sacral-mythic theme of the Theogony, Hesiod could have adopted a, so to speak, more modern and intellectual register which almost ignores dual forms, while for Works and Days, according both to the theme and to the receiver of the poem21, the rural Boeotian oral dialect was perceived as more adequate.


Critical Editions

  • Solmsen, F., Hesiodi Theogonia, Opera et Dies, Scutum, Oxford, 1970.
  • West, M. L., Hesiod Theogony, Oxford, 1966.


  • Rzach, A., Hesiodi Carmina accedit Homeri et Hesiodi certamen, Leipzig, 1902.
  • Wilamowitz-Moellendorf, U. von, Hesiods Erga, Berlin, 1962.


  • Chantraine, P., Grammaire homérique, Paris, 1963.
  • Kühner, R., Blass, F., Gerth, B., Ausführliche Grammatik der Griechischen Sprache, Hannover, 1869.
  • Rix, H., Historische Grammatik des Griechischen, Laut und Formenlehre, Darmstadt, 1992.
  • Zinsmeister, H., Lindemann, H., Färber, H., Griechische Grammatik, München, 1954.

Lexica and Encyclopediae

  • Der Neue Pauly: Enzyklopädie der Antike, hrsg. von Cancik, H., und Schneider,  Stuttgart, 1996-2003.
  • Lexikon des frühgriechischen Epos, begründet von Snell, B., Göttingen 1979ff.
  • Liddell, H. G., Scott R., Jones, H. S., A Greek-English Lexikon, Oxford 1968.

Secundary Literature

  • Bechtel, F., Die griechische Dialekte, Erster Band, Berlin, 1921.
  • Buck, C. D., The greek dialects, Grammar, Selected Inscriptions, Glossary, Chicago, 1928.
  • Corbett, G., Number, Cambridge, 2000.
  • Cuny, A., Le nombre duel en grec, Paris, 1906.
  • Edwards, G. P., The language of Hesiod in its traditional context, Oxford, 1971.
  • Giacalone Ramat, A., Ramat, P., Le lingue indoeuropee, Bologna 1994.
  • Gonda, J., Reflections on the numerals "one" and "two" in ancient Indo-European languages, Utrecht, 1953.
  • Hoffmann, O., Debrunner, A., Geschichte der griechischen Sprache, Berlin, 1953.
  • Humboldt, W. von, Über den Dualis, Berlin, 1828.
  • Illek, F., Der Dual bei Hesiod, Zeitschrift für die österreichischen Gymnasien, n. 39, 1888, SS. 97-102.
  • Lühr, R., Zum Gebrauch des Duals in der Indogermania,in 125 Jahre Indogermanistik in Graz, edd. Ofitsch, M., Zinko, C., Graz2000, SS. 263-274.
  • Malzahn, M., Die Genese des indogermanischen Numerus Dual, in 125 Jahre Indogermanistik in Graz, edd. Ofitsch, M., Zinko, C., Graz2000, SS. 291-315.
  • Meier Brügger, M., Griechische Sprachwissenschaft, Berlin, 1992.
  • Meillet, A., L'emploi de duel chez Homère et l'èlimination du duel, in Mèmoires de la Sociètè de Linguistique de Paris, 22,SS. 145-64.
  • Meillet, A., Aperçu d' une histoire de la langue Greque, Paris, 1930.
  • Nicolai, W., Hesiods Erga, Beobachtungen zum Aufbau, Heidelberg 1964.
  • Ohler, Über den Gebrauch des Duals bei Homer, Mainz 1884.
  • Plank, F., Humboldt über den Dualis, in: Humboldt-Grimm-Konferenz, Protokollband, Teil I Berlin, 1986, SS. 231-245.
  • Rzach, A., Der Dialekt des Hesiodos, Leipzig, 1876.
  • Schlieben-Lange, B., Soziolinguistik, Eine Einführung, Stuttgart, 1958.
  • Sellschopp, I., Stilistische Untersuchungen zu Hesiod, Hamburg 1934.
  • Thumb, A., Handbuch der griechischen Dialekten, Heidelberg, 1959.
  • Troxler, Sprache und Wortschatz Hesiods, Zürich, 1964.
  • Viti, C., The use of the dual number in Homeric Greek, in Akten der 13. Fachtagung der Indogermanischen Gesellschaft, Wiesbaden, 2008, SS. 595-604.
  • Vottéro, G., Le dialecte Béotien, L' écologie du dialecte, Nancy, 1998.

1 While in the Theogony only three duals occur, there are forty-seven negligences in this poem. The percentage of employment of the dual in proportion to the negligences is 6 %, whereas the proportion between this value and the number of the verses amounts to 0,29%. For all the occurrences, the decision of the dual is always metrically obligatory, while the choice for the negligences is mostly open. In Works and Days there are eight occurrences, while the negligences are two times as many. The percentage of the employment in proportion to the negligences amounts to 33,33 %, while the proportion between this quantity and the number of the verses is 1,15%. Mostly, the choice for a dual is not metrically obligatory.

2The occurrences in the Theogony are at v. 475 (πεφραδέτην), v. 698 (ὄσσε) and v. 892 (φρασάτην). In all of these cases a plural would not be metrically possible:  πέφραδον and φρασάν would not be suitable in the verse. As ὄσσε exists just in dual, this passage is to be interpreted as metrically obligatory.

3In 21/50 of the cases the choice is metrically obligatory in the whole poem.

4See note 1.

5V. 13; v. 199; v. 185; v. 432.

6Only in 4/24 cases metrically obligatory.

  7See Cuny, p. 85 ff. about Attic epigraphic.

811/14 Occurrences.

9Theogony: v. 475 (πεφραδέτην) and v. 892 (φρασάτην). Works and Days: v. 438 (ἔχοντε).

10There are different opinions as well: Thumb claims e.g. on p. 211 that the poems attributed to Hesiod are in no way testimonies of the Boeotian dialect and that this dialect did not affect Hesiod´s language too much.

11W. von Humboldt, p. 18.

12Gonda 1953, p. 9.

13The dual is perceived as a numerus, which is specifically employed for the "in der Natur verbundenen Gegenstände" (Kühner, Blass, Gerth, Teil II, Band II, p. 68), constituting a unity. Although the concept of duality tells us much about the vocabulary assortment of the dual forms, it does not seem to be a valuable criterion to surely determinate where a dual should be expected und where not (cf. Illek, p. 100). The first reason is that often the same subjects occur not just in dual, but also in plural form; the second reason is that the concept of duality/unity is too generally formulated (a critic against the formulation of the duality theory of Humboldt can be found in Plank, p. 242): to rightly claim that two subjects constitute an inseparable duality, it is enough to observe that they appear in the text as a couple. Although the distribution and the quantity of the occurrences is surely useful for the understanding of the linguistic register, it is not the same with the analysis of singular occurrences, for eventually some randomness plays a role in this sense. Nevertheless it has been tried to legitimate singular occurrences by means of grammatical rules, but most negligences were still not explained or ignored: in legitimizing a dual it is also necessary to explain why in other passages with a similar constellation a plural appears (see e.g. Cuny p. 501, who convincingly explains the plurals in Hes. erg. vv. 432-441 as exceptions, but ignores the – two times higher! – quantity of negligences in the same poem. Another example is Gonda who theorizes that the employment of the dual depends on the presence or absence of the word δύο. Although these are clever analyses, they do not give fully satisfying explanations).

14Concerning the theory of “Diglossie ohne Zweisprachigkeit" see Ferguson, 1959; Cf. e.g. also the process of dwindling of the Italian dialects in Loporcaro, M., Profilo linguistico dei dialetti italiani, Roma-Bari, 2013, p. 176 ff.

15Cf. Kühner, Blass, Gerth, Teil II, Band I, p. 285.

16The connection between the recession of a specific primitive culture and the dwindling of the dual has been already observed in the Indo-European studies (e.g. Cuny, p. 5 or Gonda, pp. 9-11).

17About the "collective meaning" of the dual see Malzahn, p. 292 ff.

18For the Greek language cf. Kühner, Blass, Gerth, Teil II, Band I, p. 285.

19The employ of the dual as well as colloquial expressions (cf. e.g. Troxler p. 113 about Hes.erg., v. 453) could be an indicator of the influence of the Boeotian oral form on the language of Works and Days, as in this dialect, which Hesiod surely knew, the dual has been preserved for a longer time (see Buck, p. 87): In fact Hesiod asserts in Works and Days (Hes. erg., v. 640) that he spent most of his life in Askra in Boeotia. Furthermore the fact that in the Boeotian region the dual was used in rural expressions suits to what is known about Boeotia: that it was for a long time a mostly rural area (cf. Vottéro, p. 205).

20E.g. Cuny who even assumes that Theogony is not by Hesiod (see Cuny, p. 502). For a different opinion see Illek, p. 102.

21Cf. Schlieben-Lange, p. 79: "Jeder Sprecher verfügt über mehrere Varianten seiner Muttersprache, zwischen denen er umschalten kann (...) Solch ein Umschalten geschieht in Abhängigkeit der Situation, dem Auditorium, dem Thema usw. Es kann sich abspielen zwischen regionalen Varianten (häufig beherrscht ein Sprecher einen Dialekt und die regionale Form der Hochsprache), zwischen sozialen und zwischen stilistischen Varianten. Dabei sind alle mögliche Überschneidungen zwischen diesen Arten von Varianten möglich".


Managing Common Ground. Greek particles as grounding operators

Free University Amsterdam

Greek particles are usually seen as communicative devices serving as ‘sign-posts’ in the interaction between speaker and addressee, used to express the interlocutors’ stance and to organize the discourse.  This paper takes a closer look at particles in their role as ‘signposts’. Key to my approach will be the notion of common ground (e.g.  Stalnaker 1978, Clark 1996, Verhagen) and the closely related cognitive linguistic notion of current discourse space [CDS] (Langacker 2001). According to Langacker, ‘[b]esides the context of speech, the CDS includes a body of knowledge presumed to be shared and reasonably accessible. It also includes the speaker’s and hearer’s apprehension of the ongoing discourse itself: a series of previous usage events, as well as subsequent events that might be anticipated’ (Langacker 2001: 145). In Langacker’s approach, the CDS is a cognitive domain that can be evoked as the conceptual base for the meaning of linguistic elements and it plays a crucial communicative role in coordinating the interlocutor’s focusing of attention and perspectives (cf. Verhagen’s 2005 notion of intersubjective coordination).  According to Clark (1996), the interlocutor’s common ground can be based on shared perceptual experiences, joint actions and current discourse (personal common ground) and on shared membership of a community (communal common ground).  A more specific component  of the common ground which is relevant to the semantics of discourse markers is the topos, a general default rule shared by the members of some community or culture which states what is normally the case (Anscombre & Ducrot 1983, Verhagen 2005.

            In this paper, I will discuss a number of particles that have been classified as‘modal’, ‘attitudinal’ or ‘interactional’ (such as δή, μήν, τοι) as they are used in dialogical texts (especially drama), and I will analyze them as grounding operators: as instructions to the addressee how to relate the host utterance to an element in the common ground, for example by specifying the element of the common ground to which the attention should be directed,  by invoking the common ground (and especially topoi) as a basis for argumentation, and by pointing out possible conflicts with the common ground.  An analysis of particle meaning in terms of common ground management may help us to better understand some of the functional differences between particles in dialogical texts.

Allan, R.J. 2017. ‘Ancient Greek Adversative Particles in Contrast’, in Pragmatic Approaches to Latin and Ancient Greek, Denizot, C. & Spevak, O. (eds.). Amsterdam/Philadelphia, 273- 301.

Allan, R.J., 2017, ‘The Grammaticalization of Greek Particles: A Functional Discourse Grammar Approach’, in  Ancient Greek Linguistics: New Approaches, Insights, Perspectives. Logozzo, F. & Pocetti, P. (eds.). Berlin/Boston: De Gruyter, 103-118.

Anscombre, J.C. & O. Ducrot. 1983. L’argumentation dans la langue, Brussels.

Bakker, S.J. & G.C. Wakker (eds.). 2009. Discourse Cohesion in Ancient Greek, Brill, Leiden/Boston.

Bonifazi, A., A. Drummen & M. de Kreij. 2016. Particles in Ancient Greek discourse: Five volumes exploring particle use across genres. Hellenic Studies Series 74. Washington, DC: Center for Hellenic Studies. Particles_in_Ancient_Greek_Discourse.2016

Clark, H.H. 1996. Using Language, Cambridge.

Langacker, R.W. ‘Discourse in Cognitive Grammar’, Cognitive Linguistics 12, 143-188.

Rijksbaron, A. (ed.). 1997. New Approaches to Greek Particles. Proceedings of the Colloquium held in Amsterdam, 4–6 January 1996, to honour C. J. Ruijgh on the occasion of his retirement, Amsterdam.

Sicking, C.M.J. & J.M. Van Ophuijsen. 1993. Two Studies in Attic particle usage, Leiden/New York/Cologne: Brill.

Stalnaker, R.C. 1978. ‘Assertion’, in P. Cole (ed.), Syntax and Semantics 9: Pragmatics, New York, 315-332.

Verhagen, A. 2005. Constructions of Intersubjectivity: Discourse, Syntax, and Cognition, Oxford.

Wakker, G.C., 1997, ‘Emphasis and Affirmation. some aspects of μήν in tragedy’, in Rijksbaron, A. (ed.), 1997,         New Approaches to Greek Particles. Proceedings of the Colloquium held in Amsterdam, 4–6 January 1996, to honour C. J. Ruijgh on the occasion of his retirement, Amsterdam.

Myth of the raising vowels. The spellings <Y> and <OY> for <O> and the alleged high short o in Ancient Greek dialects



The spelling <Y> for <O> is occasionally attested in inscriptions of various Greek regions:

Boeot. ὐκτάς = Att. ὀκτάς (ca. 396 BCE)

Boeot. Βυΐλλε̄, cf. Βοΐδιον (410-400 BCE)

Boeot. Διυδότω = Att. Διοδότου (210-200 BCE)

Boeot. Νιυμείνιος for Νιομείνιος = Att. Νουμήνους (220-200 BCE)

Thess. Ττυλίχνας for Πτολίχνας (3rd c. BCE)

Cret. Χαριμύρτω = Att. Χαριμόρτου (2nd c. BCE)

Cyr. Πτυλεμαῖος = Att. Πτολεμαῖος (1st c. CE)

According to a common belief, short o was a close mid vowel which in some dialects could easily be mistaken for short u and hence be accordingly spelled with <Y>. The spelling <Y> for <O> purportedly triggered the reverse phenomenon (i.e., <Ο> for <Υ>):

Boeot. σκόφοι = Att. σκύφοι (ca. 396 BCE)

Maced. Εὐροδίκη = Att. Εὐρυδίκη (3rd c. CE)

Allegedly, the raising of o is further confirmed by the spelling <OY> for <O> in Boeotian documents, in which <ΟΥ> for short and long u was used since the 4th c. BCE:

[Ἀ]πολλούδωρος = Att. Ἀπολλόδωρος (ca. 235-230 BCE)

Βουίσκος , cf. Βοΐσκος (3rd c. BCE)

ποιιούμενος for ποιιόμενος = Att. ποιούμενος (late 3rd-early 2nd c. BCE)

Διουκλεῖς = Att. Διοκλῆς (2nd c. BCE)

νιουμεινίη for νιομεινίη = Att. νουμηνίᾳ (222-200 BCE)

However, the hypothesis of a close mid short o which could justify the above spellings is unlikely, since it runs counter to typological evidence: cross‑linguistically, in languages with vowel systems which contrast both length and height, short vowels tend to be lower than long ones. As a matter of fact, the spelling <Ο> for <Υ> must be taken at face value: it represents the lowering of short u. In this paper, I will focus on the spellings <Y> and <OY> for <O> and will offer an alternative interpretation of the data.


Allen, W. S. Vox Graeca. A Guide to the Pronunciation of Classical Greek. Cambridge, 1987 (3rd edition).

Bile, M. Le dialecte crétois ancien. Paris, 1988.

Brixhe, Cl. “Dialecte et koinè à Kafizin”. In: Karageorghis, J. and O. Masson (ed.), The History of the Greek language in Cyprus. Larnaca, 1988: 167-180.

Phonétique et phonologie du grec ancien I. Quelques grandes questions. Louvain-la-Neuve, 1996.

- “Un ‘nouveau’ champ de la dialectologie grecque: le macédonien”. In: Cassio, A. C. (ed.), KATA DIALEKTON. Atti del III colloquio internazionale di dialettologia greca. Napoli, 1999: 41–71.

Dobias-Lalou, C. Le dialecte des inscriptions grecques de Cyrène. Paris, 2000.

Egetmeyer, M. Le dialecte grec ancien de Chypre. Berlin, 2010.

Lejeune, M. Phonétique historique du mycénien et du grec ancien. Paris, 1972.

Méndez Dosuna, J. “Ancient Macedonian as a Greek dialect: A critical survey on recent work”. In: G. Giannakis (ed.), Ancient Macedonia. Language, History, Culture. Thessaloniki, 2012: 133-145.

Schwyzer, E. Griechische Grammatik I Allgemeiner Teil, Lautlehre, Wortbildung, Flexion. Munich, 1939.

“Entwining Greek with Asian Speech”:

An Example of Secondary Foreigner Talk in Timotheus of Miletus’ Persae

University of California, Los Angeles | Sorbonne University, Paris

            After the conquest of the Lydian capital Sardis by Cyrus the Great in 546 BC, the complex geopolitical situation of Anatolia was reduced for the first time to political and administrative unity. However, the ethnolinguistic context remained heterogeneous thanks to the tolerant linguistic policy of the Persians (Basello 2013). The linguistic repertoire of Anatolia during the Achaemenid Era included many varieties: the dominators’ languages, old Persian and Aramaic; the epichoric languages, one of which was Phrygian; and Greek, whose penetration into the intermediate zone between the western coast and the Anatolian hinterland was promoted by the Achaemenid administration (Asheri 1983: 15–17). In this multilingual context, the scene represented by Timotheus of Miletus in his nome Persae (late 5th century BC) is not implausible. In one of the direct speeches describing the Battle of Salamis (480 BC), the poet decides to give the floor to a Phrygian soldier from Kelainai, engaged in the Persian army. After having been seized by his Greek aggressor and dragged by the hair (140–149), the Phrygian soldier begs him to spare his life, speaking in broken Greek (150–161).

            The incomplete linguistic competence of a non-native speaker in a target language is known as a linguistic register called broken language (Ferguson – DeBose 1977), which can be easily imitated by a native speaker through a register called secondary foreigner talk (Hinnenkamp 1982: 40–41). In a literary context, this secondary foreigner talk is a very precise technique that aims to obtain different effects, from mimetic realism to humorous parody, according to the author’s needs (Traugott – Pratt, 1980: 358–397). Through his attempt to reproduce the overall effect of the Phrygian soldier’s imperfect knowledge of Greek, Timotheus connects himself to a tradition already established by Ancient Comedy (cf. Aristophanes: Scythian archer, Thesmophoriazousae; Pseudartabas, Acharnians; Triballic God, Birds). Several studies have been devoted to these passages, including from a sociolinguistic point of view (Brixhe 1988, 2012; Willi 2003: 198–225), but this has never been done before for Timotheus’ Phrygian soldier.

Commentators (among the most recent ones: Janssen 1984; Hordern 2002; Sevieri 2011; Lambin 2013) have often focused on the “grammatical mistakes” present in the Phrygian’s speech, considering their analysis as an end in itself. However, it seems much more interesting to situate these deviations from the norm of Greek language in the framework of secondary foreigner talk, in order to understand the strategies used by Timotheus to reproduce in a credible way the type of Greek spoken by the Phrygian. In light of the latest knowledge of Phrygian, it is possible to see the extent to which Timotheus pushed his mimesis, allowing us to identify the elements of the Phrygian’s speech that would actually be compatible with an Ionic Greek dialect learned by a non-native speaker of Phrygian origin, and to distinguish them from those attributable only to the poet’s linguistic creativity in the literary secondary foreigner talk.


Asheri, David. 1983. Fra ellenismo e iranismo: studi sulla società e cultura di Xanthos nella età achemenide. Bologna: Pàtron.

Basello, Gian Pietro. 2013. “Le unità amministrative dell’impero achemenide (satrapie): il potere percepito dai popoli sottomessi e le immagini di ritorno”. Ricerche storico bibliche 25. 37–97.

Brixhe, Claude. 1988. “La langue de l’étranger non grec chez Aristophane”. L’étranger dans le monde grec: actes du colloque organisé par l’Institut d’Études Anciennes, Nancy, mai 1987 (ed. Lonis, Raoul). Nancy: Presses Universitaires de Nancy. 113–137.

—. 2012. “Le pseudo-pidgin de l’étranger non grec chez Aristophane”. Folia graeca in honorem Edouard Will: Linguistica (eds. Brixhe, Claude, Vottéro, Guy). Nancy: Association pour la Diffusion de la Recherche sur l’Antiquité (A.D.R.A.). 65–81.

Ferguson, Charles A., DeBose, Charles E. 1977. “Simplified Registers, Broken Language, and Pidginization”. Pidgin and Creole Linguistics (ed. Valdman, Albert). Bloomington/London: Indiana University Press. 99–125.

Hinnenkamp, Volker. 1982. Foreigner Talk und Tarzanisch: Eine vergleichende Studie über die Sprechweise gegenüber Ausländern am Beispiel des Deutschen und des Türkischen. Hamburg: Helmut Buske Verlag.

Hordern, James. 2002. The Fragments of Timotheus of Miletus. Oxford: Oxford University Press.

Janssen, Tittje H. 1984. Timotheus, Persae. A Commentary. Amsterdam: Adolf M. Hakkert.

Lambin, Gérard. 2013. Timothée de Milet. Le poète et le musicien. Rennes: Presses Universitaires de Rennes.

Sevieri, Roberta. 2011. Timoteo. I Persiani. Milano: La vita felice.

Traugott, Elizabeth C., Pratt, Mary L. 1980. Linguistics for Students of Literature. New York: Harcourt Brace Jovanovich.

Willi, Andreas. 2003. The Languages of Aristophanes. Aspects of Linguistic Variation in Classical Attic Greek. Oxford: Oxford University Press.


Zurück zum „Pelasgischen“? Ein neuer Interpretationsvorschlag

Department of Greek and Latin Studies, Faculty of Arts, Charles University, Prague

Die Sprachkontakte des Griechischen im vormykenischen Zeitalter, für das man keine vollständig und sicherlich gedeutete schriftliche Quellen zur Verfügung hat, stellen für die Forschung nach wie vor ein recht problematisches und umstrittenes Thema dar. Als fast allgemein akzeptierte Meinung könnte aber zur Zeit gelten, dass die Benutzer des Griechischen (als die Muttersprache) im Griechenland mit den Sprechern zumindest einer nichtindogermanischen „vorgriechischen“ Sprache im Kontakt gelebt haben, wovon u.a. die griechischen Lexeme offensichtlich nichtindogermanischen Ursprungs zeugen.

Eine vor allem in der dritten Viertel des 20. Jahrhunderts lebendige Forschungsrichtung (V. Georgiev, A. J. Van Windekens, W. Merlingen...) hat für die Vorgeschichte des Griechischen auf dem Boden des Griechenlands auch (oder: ausschließlich) eine, vorwiegend als Pelasgisch bezeichnete, indogermanische Kontaktsprache (oder aber sogar zwei indogermanische Kontaktsprachen; dann tritt auch ein sog. „Psisprache“/„Psigriechisch“ dazu) postuliert. Die Grundlage dafür waren ebenfalls die Wörter des altgriechischen Korpus, in diesem Fall allerdings solche indogermanisch anmutende, phonetisch jedoch offenbar ungriechische.

Obwohl das von den erwähnten Forschern vorgebrachte Material allzu oft kaum überzeugend wirkte sowie für die Sicherung der ehemaligen Existenz „vorgriechischer“ indogermanischen Sprachen ungenügend war und obwohl sich das Thema seit fast einem halben Jahrhundert keiner ernstlichen Diskussion mehr erfreut, gibt es im Griechischen doch wenigstens ein paar Wörter, bei denen sich über eine indogermanische, zugleich aber keine problemlose griechische, Etymologie nachzudenken lässt, wenn nicht sogar aufdringt, wie ab und zu zugegeben worden ist (der Paradebeispiel wäre σῦς – in diesem Fall sogar mit einem „echtgriechischen“ Gegenstück ὗς; weiter etwa πύργος, τύμβος u.a.).

Es gibt eine Möglichkeit, diesen Befund zu interpretieren, die eine Art Kombination der „pelasgologischen“ und traditionellen Sichtweise ist und die ziemlich naheliegend ist, an die aber erstaunlicherweise, wie es scheint, bisher nicht gedacht wurde: Die Griechen haben die fraglichen Lexeme von der ursprünglicheren Bevölkerung Griechenlands übernommen, deren Sprache(n), wie die überwiegende Meinung lautet, wirklich nicht indogermanisch war (waren). Die Sprecher dieser Sprachen haben aber seinerseits diese – etymologisch natürlich indogermanische, wie die Pelasgologen behaupteten – Wörter schon früher von den Griechen (aus dem Griechischen bzw. seiner Vorform) übernommen. Es geht also im Griechischen um Rückentlehnungen.

Ich möchte diese grundsätzliche Überlegung zur Diskussion stellen. Sollte dieser Vorschlag akzeptiert werden und sich etymologisch als zumindest einigermaßen nützlich erweisen, würde er hoffentlich auch einige neue Einsichten in die Entwicklung des Griechischen sowie in die Kontaktsituationen der Sprecher dieser Sprache erlauben.

Modality and Injunctive in Homeric Greek:

the case of counterfactual and epistemic constructions

Università di Palermo, Universität zu Köln

Homeric unaugmented aorists and imperfects are the oldest verbal forms attested in Greek, which continue the so-called Indo-European ‘injunctives’. The latter were inflectionally underspecified as regards verbal categories such as tense or mood (Hoffmann 1967; Kiparsky 1968). Thus, the question arises as to how the attitude of the speaker towards the content of his utterance was expressed. The aim of this paper is to investigate the role of epistemic particles co-occurring with injunctives in the Iliad and the Odyssey, focusing in particular on past counterfactual constructions. Crosslinguistic studies have shown that such modal constructions reflect the universal semantic distinction between realis and irrealis (Wierzbicka 1997: 38). Specifically, the data show epistemic particles like ἄρα, δή, που, etc. (cf. Denniston 1954) occurring in the if-clause or protasis, which is usually made of εἰ μὴ + injunctive or past indicative and refers to an actual event in the past for which the outcome is already known. Differently, the main clause or apodosis is always (lexically) marked by the irrealis particle κέν + injunctive or past indicative, as also expected in a typological perspective (Elliott 2000), and refers to a potential event in the past, which in fact never happened (see also Hettrich 1998). The analysis of all the occurrences of such complex constructions shows a not random distribution of those epistemic particles, whose frequency significantly decreases when the protasis has an indicative rather than an injunctive. It might be argued that the use of epistemic particles was initially the only (lexical) means to express the speaker’s commitment to the truth of a proposition, while the more recent indicative tensed forms rendered them redundant at a later stage, since the verb was already inflected according to modality. Another piece of evidence in favour of this hypothesis comes from the use of the epistemic verb μέλλω that develops into a periphrastic marker for future tense, especially as a future in the past (cf. Allan 2017). The Homeric poems show most instances of the unaugmented 3SG occurring with an epistemic particle, while there is variation with the augmented form. The remainder of the paper will discuss the development of the various readings of μέλλω in the epic language both in combination with and without epistemics.


Allan, Rutger J. 2017. ‘The History of the Future: Grammaticalization and Subjectification in Ancient Greek Future Expressions’. In Lambert F., Allan R. J., and Markopoulos T. (Eds), The Greek Future and Its History = Le Futur Grec et Son Histoire, Bibliothèque Des Cahiers de l’Institut de Linguistique de Louvain, 139. Leuven, Peeters, pp. 43–72.

Denniston, J. D. 1954. The Greek particles. 2nd ed. Oxford, Clarendon Press.

Elliott, J. R. 2000. Realis and irrealis: Forms and concepts of the grammaticalisation of reality, «Linguistic typology» 4 (1), 55-90.

Kiparsky, P. 1968. Tense and mood in Indo-European syntax, «Foundations of Language» 4, 30-57.

Hettrich, H. 1998. ‘Die Entstehung des homerischen Irrealis der Vergangenheit’. In Jasanoff J., Melchert H. C., Oliver L. (Eds), Mír Curad. Studies in honor of Calvert Watkins, Innsbruck, Innsbrucker Beiträge zur Sprachwissenschaft, pp. 261-270. 

Hoffmann, K. 1967. Der injunktiv im Veda: eine synchronische Funktionsuntersuchung. Heidelberg, Carl Winter Universitätsverlag.

Wierzbicka, A. 1997. ‘Conditionals and counterfactuals: conceptual primitives and linguistic universals’. In Athanasiadou, A. & Dirven, R. (Eds), On Conditionals Again, Amsterdam, John Benjamins, pp. 15-59.

Some new old u‑stem adjectives in Greek

Università di Bologna, Dipartimento di Filologia Classica e Italianistica

Even after De Lamberterie’s (1990) masterful study, the morphology and semantics of u‑stem adjectives in Ancient Greek are still a fruitful field for linguistic research, as I try to show in three separate case studies, applying a combined methodology of philological investigation and comparative Indo-European reconstruction. Inner-Greek and comparative evidence enables us to reconstruct an adjective in ‑ús which was lost early in its simple form.

An adj. *tr̥sú‑ (root *ters‑ ‘dry’) is attested in several IE languages (Ved. tṛṣú‑, Avest. taršu‑, Goth. þaursus); Wackernagel (1897:17) and De Lamberterie (1990:701-14) argued that it is also preserved in Gr. τραυλός ‘lisping, stammering’, continuing *tr̥s‑u‑ló‑ with a ‘complex Caland suffix’ (cf. δαυλός ‘thick, shaggy’ < *dn̥s‑u‑ló‑ besides δασύς < *dn̥s‑ú‑). Accepting their phonetic derivation, I propose to derive this word from an homophonous *tr̥sú‑ (root *tres‑ ‘tremble’) which would fit better the semantic nuances of τραυλός in ancient sources. An adj. *tr̥sú‑ ‘moving quickly, quivering’ could also lie behind some occurrences of tṛṣú‑ in the Rigveda (Debrunner 1954:566), thus projecting this form back to (Late?) PIE.

An adj. *agú‑ can be tentatively reconstructed on the basis of fem. plur. ἀγυιαί ‘streets’, a morphological archaism as shown by its oxytone accentuation (Lundquist 2017:23-60), possibly cognate with Armenian acu ’garden bed’. I argue that additional evidence for this adjective can be found in ἄγυρις ‘gathering, assembly’, comparing θαμύς ‘thick, packed, crowded’, attested in plur. m. θαμέες, f. θαμειαί (cf. ἀγυιαί) and in several derivatives (De Lamberterie 1990:664-675, Anttila 2000:44) including a fem. noun θάμυρις ‘assembly’ (an i-stem abstract from θαμυρός ’crowded’), which presents us with a near-perfect formal and semantic match for ἄγυρις.

Lastly, a Gr. adj. *λαφύς ’greedy’ (root *labh‑ ’take, seize, grasp’, Ved. rabh‑/labh‑; possibly also in εἴληφα ’I took’, cf. Ringe 1984) is tentatively posited on the basis of neut. plur. λάφυρα ’spoils’ and of the epithet Λαφύστιος ’devourer’, which I propose to derive from *labhu‑h1d‑ ’eating greedily’. Possible cognates inside and outside Greek are discussed, including the obscure divine epithet Λάφριος/-α.


Anttila, R., Greek and Indo-European Etymology in Action, Amsterdam 2000.

Debrunner, A., Altindische Grammatik. Band II.2: die Nominalsuffixe, Göttingen 1954.

de Lamberterie, Ch., Les adjectifs grecs en ‑υς, Louvain-la-Neuve 1990.

Lundquist, J., Archaisms and Innovations in the Songs of Homer, PhD diss., UCLA, 2017.

Ringe, D.A., ‘εἴληφα and the aspirated perfect’, «Glotta» 62 (1984), 125-141.

Wackernagel, J., Vermischte Beiträge zur griechischen Sprachkunde, Basel 1897.

Deictic shifts in Greek contractual writing

Ghent University

Studies of Greek contractual writing typically distinguish between ‘objectively’ and ‘subjec­tively’ styled contracts (see e.g. Mitteis 1963; Palme 2009). As traditional terminology indicates, these two types of contracts have different vantage points or ‘deictic centers’ (cf. Zubin & Hewitt 1995), and accordingly also different personal references: with the former type, the scribe or someone else forms the deictic center, so references to the contracting parties are in the third person. In subjectively styled contracts, on the other hand, the initiating contrac­ting party forms the deictic center, so references are in the first and second person. Contrast, for example, the first lines of the following two contracts:

  1. μισθώσασθαι παρὰ σοῦ τὰς ὑπα[ρχού]σας σοι περὶ τὴ(ν) αὐτὴν κώμην ὁρι[οδ]ικ[τί]ας Κερκεσούχω(ν) σιτεικὰς ἀρούρ[ας] τέ[σ]σαρας ἢ ὅσας ἐὰν ὦσι κτλ. (P.Cair.Isid.100, ll. 4-8 (296 AD))

I wish to lease from you the four arouras sown in grain, or however many they may be, which belong to you in the vicinity of the same village, in the horiodeiktia of Kerkesoucha etc.”

  1. [σι] ἀλλήλοις Ὡρίων Ὡρίωνος τοῦ Μενχείους … καὶ Ὧ[ρος Ὥρου] τοῦ Πετεύριος …  διειρῆσθαι πρὸς ἑαυτοὺς κτλ. (P.Mich.3.187, ll. 2-4 (75 ΑD))

“Horion, son of Horion, the son of Menches … and Horos son of Horos, the son of Peteuris … acknowledge that they have divided between themselves etc.”

In this paper, I will show that the distinction between these two major types was not always strictly upheld: in many contracts, one finds subtle changes of perspective, so-called ‘deic­tic shifts’ (see e.g. McIntyre 2006:91-121), as reflected in divergent personal references. These typically take the form of subtle intrusions – either objective intrusions in subjectively styled contracts, or subjective intrusions in objectively styled contracts. Consider the following example, where the use of the third person reflexive pronoun ἑαυτῆς intrudes in an otherwise subjectively written phrase:

  1. [όη] Ἡρακλείδου μετὰ κυ̣ρ̣ί̣ο̣υ̣ ἐμοῦ τοῦ ἑαυτῆς ἀν[δ]ρὸς Λυσιμάχου τοῦ Λυσιμάχου ὁμολ̣ο̣γ̣̣ ἐξίστασθαι κτλ. (P.Mich.5.350, l. 22 (37 AD))

“I, Arsinoe, daughter of Herakleides, with her guardian, my husband Lysimachos, son of Lysimachos, acknowledge that I have released etc.”

A number of reasons can be suggested for the occurrence of such deictic shifts: objective intrusions, which involve considering the contracting parties from a third-party point of view, can be related to the fact that it is unnatural for the scribe to write in the first person, and not assume himself to be the deictic center. Subjective intrusions, in which one of the contracting parties becomes the deictic center, are less frequent and more com­plex to explain: possible factors contributing to such intrusions include one of the contracting parties acting as the scribe, or the introduction of persons triggering a different point of view.

In the final part of my paper, I will show that deictic shifts are not necessarily limited to one or two subtle objective or subjective intrusions. Some documents display more elaborate deictic alternations, in which the point of view is shifted multiple times. That such documents exist, testifies to a broader confusion between the two major types of stylization.


McIntyre, D. 2006. Point of view in plays. A cognitive stylistic approach to viewpoint in drama and other text-types. Amsterdam.

Mitteis, L. & U. Wilcken. 1963. Grundzüge und Chrestomathie der Papyruskunde. Zweiter Band: juristischer teil. Erste hälfte: Grundzüge (von L. Mitteis). Leipzig & Berlin.

Palme, B. 2009. The range of documentary texts: Types and categories. In: R.S. Bagnall (ed.), Oxford handbook of papyrology, 358-394. New York.

Zubin, D.A. & L.E. Hewitt. 1995. The deictic center: A theory of deixis in narrative. In: D.A. Zubin & L.E. Hewitt (eds.), Deixis in narrative, a Cognitive Science perspective, 129-158. Hillsdale, N.J.

Καὶ τί ἂν εἰπεῖν τις ἔχοι; Expression(s) of Focus in questions in Demosthenes

Université Côte d’Azur, CNRS, BCL, France

It is now firmly established that the word order in Ancient Greek (AG) expresses information structure (Bertrand 2010; Dik 1995; Matić 2003). Specifically, two focus positions have been identified in declarative clauses: one immediately before the verb (1), and one at the end of the clause, when the communicative context licences a narrow reading of the focus domain consisting of the verb followed by other focal elements (2).

  1. Non-ratified topics — Focus — Verb — Ratified topics — Other presupposed elements
  2. Non-ratified topics — [ Verb — Ratified topics — Focus elements ]Focus domain

However cross-linguistically, the word order in interrogative clauses is often different from the canonical declarative word order, e.g. subjects are postverbal in English. So, what about the word order in AG questions?

Since WH-words are narrow focus constituents (Lambrecht and Michaelis 1998; Rochemont 1986), they are expected to be located in one of the two focus positions: Final (in situ) or preverbal.

An exhaustive survey of the constituent questions in Demosthenes’ political speeches shows that both strategies are used (3–4):

3. In situ:

ταῦτα      δ’         ἐστὶ     τί ;                               (9.39)
NRTop    CONJ   Verb    WHFoc

4. Preverbal:

ἀλλὰ      Θετταλία          πῶς                 ἔχει;    (9.26)
CONJ      NRTop             WHFoc             Verb

Nevertheless, we claim that a different analysis of those examples is preferable.

First, there are instances where the WH-word is separated from the verb by one or more constituents. Some of them can be explained away, since they fall into a class of postpositive elements, i.e. constituents that form a unit with their prosodic host: clitics, parenthetics, vocatives, or even Ratified Topic (RTop) expressions, which have been shown to be postpositive too (Bertrand 2009) (5).

5. δ’ ἐκεῖνα Φίλιππος λάβῃ, τίς        αὐτὸν κωλύσει δεῦρο βαδίζειν; (1.25)

                                                  WHFoc RTop  Verb

However, we also found elements between the WH-word and the verb which cannot be postpositive: non-ratified topics (6) or contrastive (potentially focal) expressions (7).

6. ἢ             πότεροι           τοὺς ἱππέας    προὔδοσαν […]; (9.56)
    CONJ      WHFoc             NRTop             Verb

7. ποῖ         γὰρ      αὐτὸς.  τρέψεταιμετὰ  ταῦτα;            (14.31)
    WHFoc. CONJ   FocR?    Verb.                  Adverbial

Second, the position of clitics in many instances (about one quarter of the corpus) shows that in WH-interrogatives, the WH-word constitutes an independent prosodic domain (Goldstein 2015: 200–14) (8).

8. (τί)φ (ἐποίησεν =ἄν)φ;                     (31.09)

Last, the WH-constituent is sometimes split (9):

9. τίν'         ἂν        οὗτος  ἀξίαν τῶν πεπραγμένων      ὑπόσχοι           δίκην;      (54.22)
    WHFoc  PTC      TopR    FocR                                           Verb               TopR

We argue that there is a dedicated slot at the beginning of the clause for WH-constituents (WHFoc), and that the WH-word rises stepwise to the left periphery: It first stops in the preverbal focus position before moving up to the WH-focus position. Instances like (9) would then be a case of stranding, with parts of the constituent being left at every step: δίκην in its original position, ἀξίαν τῶν πεπραγμένων in FocR, and τίν' in WHFoc. Accordingly, cases like (4) where the WH-word is apparently in the preverbal focus position are amenable to such an analysis, since everything that precedes the WH-word can be viewed as extraclausal (circumstancials, themes, etc.). Even in seemingly unproblematic examples, the position of clitics results from such stranding, as in (10) where εἰπεῖν is stranded in the FocR position.

10. (καὶ=     τί           =ἂν )φ   ( εἰπεῖν    =τις      ἔχοι; )φ                  (3.29)
       CONJ   WHFoc  PTC        FocR      PRO     Verb

Bertrand, Nicolas (2009), 'Les pronoms postpositifs dans l’ordre des mots en grec: Domaines syntaxiques, domaines pragmatiques', Lalies. Actes des sections de linguistique et de littérature [d’Aussois], 29, 227–252.

— (2010), L’ordre des mots chez Homère: Structure informationnelle, localisation et progression du récit, Unpublished PhD dissertation (

Dik, Helma J. M. (1995), Word order in ancient Greek: A pragmatic account of word order variation in Herodotus, (Amsterdam studies in classical philology; Amsterdam: J.-C. Gieben).

Goldstein, David M. (2015), Classical Greek Syntax: Wackernagel's Law in Herodotus (Brill's Studies in Indo-European Languages & Linguistics; Leiden: E. J. Brill).

Lambrecht, Knud and Michaelis, Laura A. (1998), 'Sentence accent in information questions: default and projection', Linguistics and Philosophy, 21 (5), 477–544.

Matić, Dejan (2003), 'Topic, focus, and discourse structure: Ancient Greek word order', Studies in Language, 27 (3), 573–633.

Rochemont, Michael S (1986), Focus in Generative Grammar (Studies in Generative Linguistic Analysis, Amsterdam & Philadelphia: John Benjamins Publishing Company).

Anatolian Lexicon in Mycenaean Greek

University of Oxford

Mycenaean Greek, just like first-millennium alphabetic Greek, contains a substantial number of lexical items which are not directly inherited, and which are commonly defined as “non-Greek”. Francisco Aura Jorro’s Diccionario Micénico (which does not include the newer data from Thebes, Iklaina and Hagios Vasileios) reports more than fifty items that some scholars identified as borrowings from an Anatolian language (Hittite and Luwian are traditionally used as the main comparanda). Most of these supposed isoglosses are controversial, at least to a certain extent, and a case-by-case approach is needed in order to disentangle borrowings from items inherited from PIE into Greek, or to identify possible independent developments of the two branches.

In this paper, I will present the results of a re-examination of the (supposed) Anatolian lexicon in Mycenaean Greek, including onomastic material (e.g. mo-qo-so in KN De 1381, ru-wa-ni-jo in KN X 7706 + 8108, and the anthroponyms in pi-ja°: e.g. pi-ja-ma-so, pi-ja-mu-nu, pi-ja-se-me, pi-ja-si-ro), and some ‘classical’ (e.g. di-pa, ku-wa-no) and less discussed (o-nu in KN Ld 584 +, po-ro-wi-to in PY Fr 1221 +, wo-no-wa-ti-si in PY Vn 48) case studies. The goal is to offer an overall reflection based on a systematic review of the available evidence, which could shed some light on the linguistic interactions in the Aegean and in the Eastern Mediterranean in the Late Bronze Age, and on the degree of language contact which took place between speakers of the earliest form of Greek we possess and the speakers of the second-millennium Indo-European languages of Anatolia.


Aura Jorro, F. 1995, Diccionario micénico, Madrid : Consejo Superior de Investigaciones Científicas, Instituto de Filología.

Evangelisti, E. 1966, Graecia Asianica II, «ASGM» 15-16: 6-8.

Gasbarra, V. – Pozza, M. 2012, Fenomeni di interferenza greco-anatolica nel II millennio A.C.: l’ittito come mediatore tra mondo indoeuropeo e mondo non indoeuropeo, «AION» 1 (n.s.): 165-214.

García Ramón, J.L. 2011, Mycenaean Onomastics, in Y. Duhoux – A. Morpurgo Davies (edd.) A companion to linear B. Mycenaean Greek texts and their world, Louvain-la-Neuve / Walpole: Peeters: 213-251.

Hajnal, I. forthcoming (2018), Graeco-Anatolian Contacts in the Mycenaean Period, in Fritz, M.A. –Joseph, B.D. – Klein, J.S. (eds.), Comparative Indo-European Linguistics. An International Handbook of Language Comparison and the Reconstruction of Indo-European, Berlin/New York.

Melchert, H.C. 2008, Greek mólybdos as a loanword from Lydian, in B.J. Collins – M.R. Bachvarova – I.C. Rutherford (edd.) Anatolian Interfaces: Hittite, Greeks and their Neighbours. Proceedings of an International Conference on Cross-Cultural Interaction (September 17-19, 2004, Emory University, Atlanta, GA), Oxford: Oxbow: 153-157.

Milani, C. 2001, Onomastica micenea e onomastica anatolica, in O. Carruba – W. Meid (hrsg.) Anatolisch und Indogermanisch. Anatolico e Indoeuropeo. Akten del Kolloquiums der Indogermanischen Gesellschaft. Pavia, 22.-25. September 1998, Innsbruck: Institut für Sprachen und Literaturen der Universität Innsbruck: 273-290.

Morpurgo Davies, A. 1986, The linguistic evidence: is there any?, in G. Cadogan (ed.) The end of early Bronze Age in the Aegean, Leiden: Brill: 93-123.

Negri, M. 1981, Miceneo e lingua omerica, Firenze: La Nuova Italia.

Oettinger, N. 2008, The Seer Mospus (Muksas) as a Historical Figure, in Collins, B.J. – Bachvarova, M.R. – Rutherford, I. (edd.) Anatolian Interfaces. Hittites, Greeks and their Neighbours, Oxford: Oxbow: 63-66.

Szemerényi, O. 1977, Review of: P. Chantraine Dictionnaire étymologique de la langue grecque. Histoire des mots. Tome III: Λ-Π, «Gnomon» 49/1: 1-10

Watkins, C. 2007, The Golden Bowl: thoughts on the New Sappho and its Asianic Background, «CA» 26/2: 305-324.

Widmer, P.  2006, Mykenisch RU-WA-NI-JO ,Luwier’, «Kadmos» 45: 82-84.

Discourse segmentation in Herodotus and Thucydides: Suggestions from medieval punctuation

Universität zu Köln/University of Cologne

This paper compares punctuation in the OCT editions and in two medieval manuscripts of Herodotus and Thucydides’ Histories. When we read an OCT edition of either historian, we process intuitively not only words but also punctuation markers. We may stop reading when we find a full stop, for example, or we might infer from a comma the boundary of a certain phrase or clause. 

However, we (and our students) tend to forget that those full stops and commas have not been transmitted directly from antiquity. They represent the result—an interpretive result—of a complex cultural and philological operation where readability, writing rules, syntactic clarity, and the comparison with previous editions (sometimes including pre-print editions) are considered at once. 

            The first part of the paper zooms in on two passages from each historian, and illustrates the differences between the medieval punctuation adopted in Laurentianus A, the oldest extant manuscript of Herodotus (dating back to the 10th century CE) as well as in the manuscript β M of Thucydides (British Museum 11727, dating back to the 11th century CE), and the modern punctuation adopted in the two respective OCT editions. The theoretical implication of this part is what we can infer about the different position, size, and the nature of the discourse boundaries being indicated. The practical implication consists in facing possible shifts of meaning and of syntactical arrangement that depend on which segmentation is “read”.     

            The second part complements the analysis by adding input from three related topics: punctuation in antiquity and in early print editions; the macro-segmentation reflected in section and chapter divisions; and, finally, contemporary thoughts about prose colometry, intonational phrases, and discourse acts—all of them actually matching medieval more than modern segmentation strategies. The overall picture is a variety of criteria followed by readers, scribes, teachers, and philologists across centuries. Such a variety questions the idea of “the” right punctuation to be adopted (and published), and, perhaps more remarkably, it questions assessments about the length and shape of periods and sentences. 

The paper concludes by suggesting how much more reliable than punctuation are the words and their order in guiding us through the articulation of the text, and which linguistic constructions or patterns are likely to indicate a minor or major discourse boundary.

Bonifazi, A. 2016. Particle Use in Herodotus and Thucydides. Volume IV (142.064 words) in Particles in Ancient Greek Discourse. Five Volumes Exploring Particle Use Across Genres, A. Bonifazi, A. Drummen, and M. De Kreij. Washington D.C.: Center for Hellenic Studies.

Fraenkel, E. 1933. “Kolon und Satz II. Beobachtungen zur Gliederung des Antiken Satzes.” Nachrichten der Göttinger Gesellschaft der Wissenschaften, Philosophisch-Historische Klasse, 319–354.

Goldstein, D. M. 2016. Classical Greek Syntax: Wackernagel’s Law in Herodotus. Leiden.

Habinek, T. N. 1985. The Colometry of Latin Prose. University of California Publications. Classical Studies 25. Berkeley.

Hemmendinger, B. 1955. Essai sur l’histoire du texte de Thucydide. Collection d’études anciennes. Paris.

———--------------. 1981. Les manuscrits d’Hérodote et la critique verbale. Pubblicazioni dell’Istituto di filologia classica e medievale 72. Genoa.

Parkes, M. B. 1992. Pause and Effect. An Introduction to the History of Punctuation in the West. Aldershot.

Scheppers, F. 2011. The Colon Hypothesis: word order, discourse segmentation and discourse coherence in Ancient Greek. Brussels: Vubpress.

Exclamative nominatives and nominatives pro vocatives in Ancient Greek: a possible distinction?

“La Sapienza”, University of Rome

The absolute uses of the nominative case, namely its secondary extrasyntactical functions, consist in lists, quotations, anacolutha and exclamations.

In this contribution we are interested in exclamations, in particular those expressing evaluations, generally referred to a first (Ὦ πόλλ’ἐγὼ μοχθηρός ‘oh unhappy me!’ S. Ph., 254) or a third person (κατέκτανεν ᾦ ἐνὶ οὶκῳ, σχέτλιος ‘he killed him in his house, how mad!’ Od., XXI, 27-28).

Between the examples collected by grammarians, however, there are sentences with a so-called ‘exclamative nominative’ in contexts referred to a second person too, without any distinction (Δημοβόρος βασιλεύς, ἐπεὶ οὐτιδανοῖσιν ἀνάσσεις ‘People-devouring king, since you rule over nobodies’ Il., 1, 231).

We wonder if, in these latter circumstances, it is possible to effectively distinguish between an exclamation or a real case of nominative pro vocative.

Firstly, we will then try to define the concepts of neutralization and substitution in order to apply them to the categories of nominative and vocative cases.

We are going to show how the phenomenon of nominative pro vocative does not correspond to a neutralization, neither in terms of markedness nor if we intend neutralization as the total lack of syntactic relevance of the feature (case) value within a given context (Baerman, Brown, Corbett 2005: 30).

The use of a nominative instead of a vocative is not a systematic phenomenon and it does not imply the neutralization of the distinction between the two cases. It only implies the expression of the feature value in an unexpected way. The real neutralization of the case value, instead, is observable within the secondary uses of the nominative.

Our first aim is to demonstrate how the main contact point between nominative and vocative has to be seen not in all the absolute uses of the nominative, for the extrasyntactical feature in common with the vocative, but rather it originates in those exclamative contexts where the line between neutralization and substitution is quite fine.

It is possible to prove this with couple of sentences where the same terms occur in similar contexts, both in the nominative and vocative:

a) Τὶ δ’ ὦ τάλας, σε τοῦδ’ ἔχει πλέκους χρέος;

‘Why needest thou that wicker, thou poor wretch’ (Aristoph., Acharn., 454).

b) Μὴ σκῶπτέ μ’ ὦ τάλαν, ἀλλʹἕπου δεῦρ’ ὡς ἐμέ

‘Stop making fun of me, you wretched man! […]’ (Aristoph. Eccl., 1005).

In order to investigate the problem, we made a systematic analysis on Aristophanes’ comedies, by collecting and classifying all the exclamations and addresses in the nominative and vocative, excluding ambiguous forms, in three groups.

Examples like (a) and (b) can be put in a particular area -that we defined phatic-expressive one- where there is a contact with a second person as well as the expression of an emotion or a feeling towards the interlocutor.

Nominative and vocative forms alternate indistinctly within this area, starting from where the phenomenon of nominative pro vocative should have occurred in all its typologies.

Essential bibliography:

Baerman, M., Brown, D., Corbett, G. G. The Syntax Morphology Interface. A Study of Syncretism, Cambridge University Press, 2005.

Belardi W. L’opposizione privativa. Appendice: le occlusive del coreano in “Quaderni della sezione linguistica degli annali”, VII, Napoli, 1970.

Belardi W. Linguistica e poetica di Roman Jakobson, in Linguistica generale, filologia e critica dell’espressione, Bonacci, Roma, 1990, pp. 357-430.

Blake, B. J. Case, Cambridge, Cambridge University Press, 2001.

Brugmann K., Thumb A. Griechische Grammatik, Lautlehre, Stammbildungs und Flexionslehre, Syntax, München, 1913, IV ed.

Calboli, G. La linguistica moderna e il latino. I casi, Bologna, Pàtron, 1972.

Cantineau J. Principes de Phonologie, Paris, Klincksieck, 1957 (French transl. of Trubetzkoy, N. S. Grundzüge der Phonologie, 1939).

Chantraine, P. Grammaire homérique, Tome II, Syntaxe, Paris, Klincksieck, 1953.

Ciancaglini, C. A. Per una valutazione dei fondamenti teorici della marcatezza, in Cipriano P., Di Giovine P., Mancini M. (eds.) Miscellanea di studi linguistici in onore di Walter Belardi, Il Calamo, Roma, 1994, vol. II, 811-845.

Croitor, B., Hill, V. Vocatives, in Dobrovie-Sorin, C. e Giurgea, I. (eds.), A Reference Grammar of Romanian. 1: The noun phrase, John Benjamins Publishing Company, Amsterdam / Philadelphia, 2013, pp. 801-826.

D’Avis, F., Meibauer, J. Du Idiot! Din idiot! Pseudo-vocative Cconstructions and Insults in German (and Swedish), in Sonnenhauser, B., Noel Aziz Anna, P. (eds.) Vocative! Addressing Between System and Performance, De Gruyter Mouton, Berlin-Boston, 2013.

De Groot, A. W. Classifications of Cases and Uses of Cases, in Halle, M. (a cura di), For Roman Jakobson, Mouton & Co., The Hague, pp. 187-194, 1956.

Dickey, E. Greek Forms of Address From Herodotus to Lucian, Oxford, Clarendon Press, 1996.

Donati, M. Il vocativo nel processo identitario dell’interazione linguistica, Mo­naco, Lincom Europa, 2013.

Dunbar, H. A Complete Concordance to the Comedies and Fragments of Aristophanes, George Olms Verlag, Hildesheim, New York, 1973.

Ernout A., Thomas F. Syntaxe latine, 2e éd., Paris, Klincksieck, 1959.

Fink, R. O. Person in Nouns: is the Vocative a Case?, in “American Journal of Philology”, 93, n. 1, 1972, pp. 61-68.

Gildersleeve, B. Syntax of Classical Greek. From Homer to Demosthenes, Groningen, 1980.

Haspelmath, M. Understanding Morphology, Arnold, London, 2002.

Hill, V. Vocatives. How Syntax meets with Pragmatics, Brill, Leiden-Boston, 2014.

Hoffman, J. B., Szantyr, A. Lateinische Syntax und Stilistik, München, Beck’sche Verlagsbuchandlung, 1965.

Jakobson, R. The gender pattern of Russian, in Halle, M., Waugh, L. Russian and Slavic Grammar, Berlin-New York, Mouton, 1984, pp. 141-143.

Jakobson, R. Shifters, Verbal Categories, and the Russian Verb (1957), in Selected Writings II, 1971, pp. 130-147.

Jakobson, R. Signe zéro (1939), in Selected Writings II, 1971, pp. 211-219.

Kiparsky, P. Tense and Mood in Indo-European Syntax, in “Foundation of Language” 4, 1968, pp. 30-57.

Kühner, R., Gerth, B. Ausfühurliche Grammatik der griechischen Sprache, vol. II/1, Ahnsche, Hannover-Leipzig, 1898, III ed.

Langslow, D. Lectures on Syntax with Special Reference to Greek, Latin and Germanic, Oxford, 2009 (trad. ingl. di Wackernagel J. Vorlesungen über Syntax, mit besonderer Berücksichtigung von Griechisch, Lateinisch und Deutsch, 2 voll., Emil Birkhäuser & Cie., Basel).

Lazzeroni, R. Il nominativo esclamativo latino: un εἴδωλον scholae? In “Incontri Linguistici”, 40, 2017, pp. 77-89.

Meillet, A., Vendryes, J. Traité de grammaire comparée des langues classiques, Paris, Honoré Champion, 1979, V ed.

Michaelis, L. Exclamative constructions, in Haspelmath, M., König, E., Oesterreicher, W., and Raible, W. (eds.) Language Typology and Language Universals, Berlin, Mouton de Gruyter, 2001, pp. 1038-1058.

Riemann, O. D’après les principes de la Grammaire Historique, Paris, Klincksieck, 1927.

Schwyzer, E. Griechische Grammatik. Syntax und syntaktische Stilistik, vol. II, München, Beck, 1950.

Serbat, G. Grammaire fondamentale du latine. Tome VI: L’emploi des cas en latin. Vol. 1: Nominatif, Vocatif, Accusatif, Génitif, Datif, Peeters, 1996.

Sommerstein, A. H. The Comedies of Aristophanes, I, Acharnians, Warminster, 1980.

Sommerstein, A. H. The Comedies of Aristophanes, II, Knights, Warminster, 1981.

Sommerstein, A. H. The Comedies of Aristophanes, III. Clouds, Warminster 1982.

Sommerstein, A. H. The Comedies of Aristophanes, IV, Wasps, Warminster, 1983.

Sommerstein, A. H. The Comedies of Aristophanes, V, Peace, Warminster, 1985.

Sommerstein, A. H. The Comedies of Aristophanes, VI, Birds, Warminster 1987.

Sommerstein, A. H. The Comedies of Aristophanes, VII, Lysistrata, Warminster, 1990.

Sommerstein, A. H. The Comedies of Aristophanes, VIII, Thesmophoriazusae, 1994.

Sommerstein, A. H. The Comedies of Aristophanes, IX, Frogs, Warminster, 1996.

Sommerstein, A. H. The Comedies of Aristophanes, X, Ecclesiazusae, Warminster, 1998.

Sommerstein, A. H. The Comedies of Aristophanes, XI, Wealth, Warminster, 2001.

Svennung, J. Anredeformen: vergleichende Forschungen zur indirekten Anrede in der dritten Person und zum Nominativ für den Vokativ, Almqvist & Wiksells Boktryckeri, Uppsala, 1958.

Trnka, B. On Some Problems Of Neutralization, in Omagiu lui Iorgu Iordan cu prilejul ímplinirii a 70 de ani, Bucuresti, Editura Academiei Republicii Populare Romine, 1958, pp. 861-876

Vairel, H. The Position of the Vocative in the Latin Case System, in “American Journal of Philology”, CXX, 1981, pp. 438-447.

Waugh, L. Marked and Unmarked, in “Semiotica” 1982, 38, pp. 300-318.

Zwicky, A. M. Hey, Whatsyourname!, in “Proceedings from the 10th Regional Meeting of the Chicago Linguistic Society”, X, 1974, pp. 787-801.

Ancient Greek -ln- clusters and 1st Compensatory Lengthening

Cornell University

An issue that remains unsettled in the study of Ancient Greek historical phonology is the exact development of intervocalic *-ln- clusters (henceforth simply *-ln- clusters).

The majority of scholars (e.g. buck1933:148-9, lejeune1972:153-4, slings1975, rix1992:61, minamimoto2012, beek2013:305-12, batisti2014:256, somewhat more cautiously schwyzer1939:284) maintain that *-ln- sequence yielded a geminate *-ll- (preserved as such in Lesbian, Thessalian and Orchomenian Arcadian) and that *-ll- later simplified to -l- in all but the aforementioned dialects with compensatory lengthening of the preceding vowel (henceforth CL). Since the quality of the long vowel is identical to the one resulting from 1st CL, this change has been considered a special case of 1st CL. This position was championed by ruiperez1972 (ruiperez1972), who, on the basis of the alleged 1st CL behavior of *-ln- clusters, proposed a global reformulation of the 1st CL in terms of generalized total assimilation of the relevant clusters (a situation still preserved in Lesbian, Thessalian, and Orchomenian Arcadian) and a later across-the-board degemination with 1st CL of the preceding vowel.

Not all scholars, however, have been convinced by this account of the facts: in the literature one can encounter explicitly dissenting voices (e.g. sihler2008:213-4), authors that tacitly assume a different development for this cluster (e.g. willi2012), as well as others that take no explicit stance on the matter but do not include this sequence among 1st CL sequences (e.g. cassio2017).

In view of the lack of consensus in the literature regarding the fate of *-ln- clusters and its importance for our understanding of 1st CL as a whole, a new look at this topic is in order. In this talk I survey all of the forms that can be used to establish the fate of *-ln- clusters, paying special attention to cognates forms in other ancient Indo-European languages and epigraphical evidence in the dialects, in line with one of the main themes of this conference.

The findings of this survey are that:

1.         *-ln- clusters seem to have been targeted by at least one sound change because few traces of them can still be observed (only analogically restored forms, glosses and names, mostly of foreign origin).

2.         None of the Ancient Greek forms cited as evidence for a 1st CL from *-ln- unambiguously points to *-ln- in the first place. The cluster is most often circularly reconstructed: alternative phonological preforms are equally possible and usually preferable.

3.         There are pan-Greek forms displaying -ll- clusters that point to a preform containing *-ln- on comparative grounds. This suggest that *-ln- simply assimilated to *-ll-.

4.         By further enlarging the dossier with forms not cited in the literature, one finds out that there are Ancient Greek forms, especially ‘thematized’ nasal presents with clear cognates in other Indo-European languages, with a geminate -ll- but no CL where -ll- can only go back to *-ln- because simplification of other clusters would not yield the attested outcomes.

I conclude that there is no unambiguous evidence for a 1st CL behavior of *-ln- clusters. Moreover, if we assume that *-ln- clusters were eliminated with 1st CL, we are forced to assume that several pan-Greek forms with medial -ll- were either (a) borrowed in all dialects from Lesbian, Thessalian, or Orchomenian Arcadian or (b) that analogy applied vacuously to these forms, i.e. *-ln- assimilated to *-ll-, *-ll- was then analogically restored to *-ln-, but assimilation applied again yielding surface observed -ll-. It can be shown that none of these solutions is really tenable.

Thus, for the time being, we should simply assume *-ln- assimilated to *-ll- with no CL. A typologically common sound change that will be motivated on phonetic gestural grounds (as already briefly pointed out by mendez1994).

The Linguistic Landscape of late Roman Sicily: Interferences and Resistances.

Oriental University of Naples

Although Sicily in the late Roman Empire is clearly represented by ancient authors as a multilingual environment (e.g. Expositio totius mundi et gentium, 4th century), the 20th-century scientific debate has proposed two divergent descriptions of the Sicilian linguistic landscape: while some scholars denied a deep latinization under the Roman Empire, the increasing evidence of Latin inscriptions led others to hypothesise the decline of Greek. In the last decades, new approaches to bilingualism and linguistic contact, applied to antiquity, have demonstrated that the presence of a language does not imply the collapse of another one previously attested: many languages frequently coexist for long time.

Multilingualism has always characterised Sicily, but during Hellenism all minority languages gradually disappeared, and the diatopic and dialectal variation of Greek converged towards a Doric κοινά. This is the situation the Romans found after the Carthaginian war.

My aim is to demonstrate, both by epigraphic evidence and historical sources, that Roman Sicily was fully Greek-Latin bilingual until the end of the 5th century, and that the two languages influenced each other: Latin and Greek epigraphs show similar onomastic material and phonological and morphological features (e.g. drop of final -s and -r), as well as a number of shared set phrases (mostly from Latin). Furthermore, I argue that the persistence of Greek - especially its Doric nuance - was not a fortuitous phenomenon, but a mechanism of "koinaization" due to Sicilian pride: the Romans unified Sicily from an administrative point of view, but this new unification (the first one if we exclude Ducetius’) fostered a nationalistic pride, which built the narration of a (unreal) common "Doric Golden Age" whose language was the κοινά. The resilience of a slightly Doric Greek in Sicily was, in fact, a political reaction to the Romanization of the island and, ultimately, a result of Roman politics.


Adams J.N. 2003 Bilingualism and the Latin Language. Cambridge: Cambridge University Press.

Aikhenvald A.Y. & Dixon R.M.W. (eds.) 2006 Grammars in Contact: a Cross-Linguistic

Typology. Oxford: Oxford University Press.
BARTONĚK, A. 1973 Greek Dialects of Archaic Sicily: Their Integration Tendencies. «Graecolatina et Orientalia» V: 71-89 .

Campbell, L. 2006 Areal Linguistics: A Closer Scrutiny, in Matras, Y., McMahon, A., Vincent, N. (Eds.) Linguistic Areas. Convergence in Historical and Typological Perspective. Basingstoke: Palgrave MacMillan. 1-31.

Consani 2006 Il greco di Sicilia in età romana: forme di contatto e interferenza. In Bombi et alii (eds.) Studi linguistici in onore di Roberto Gusmani. Alessandria: Edizioni dell’Orso: 467-481.

Fanciullo, F. 2001 On the Origins of Modern Greek in Southern Italy. In Proceedings on the first international conference of Modern Greek Dialects and Linguistic Theory. Patras: University of Patras.

Hickey, R.(ed.) 2010 Handbook of Language Contact. Chichester/Malden (MA): Wiley-Blackwell.

Hutchinson, G.O. 2013 Greek to Latin: Frameworks and Contexts for Intertextuality. Oxford: Oxford University Press.

Korhonen, K. 2011 Language and Identity in the Roman Colonies of Sicily, in Sweetman (ed.) Roman Colonies in the First Century of Their Foundation.

McDonald, K. 2015 Oscan in Southern Italy and Sicily: Evaluating Language Contact in a Fragmentary Corpus. Cambridge: Cambridge University Press.
MULLEN, A. 2015 In both our languages. «Language and Literature» 24/3.
PARLANGELI, O. 1959 Contributi allo studio della grecità siciliana in «Kokalos» 5: 211ff.

Rochette, B. (2008), Le bilinguisme gréco-latin dans les communités juives d’Italie d’après les inscriptions (IIIe-VIe S.), in Biville, F., Decourt, J.-Cl. et Rougemont, G. (Eds.), Bilinguisme gréco- latin et épigraphie. Actes du colloque organisé à l’Université Lumière-Lyon 2, le 17, 18 et 19 mai 2004, Lyon: Maison de l’Orient et de la Méditerranée-Jean Pouilloux: 273-304.

Romaine, S. 1995 Bilingualism [2nd edition]. Oxford: Blackwell. 
ROHLFS, G. 1924 Griechen und Romanen in Unteritalien, «Archivum Romanicum» II/7. ROUGÉ, J. 1969 Expositio totius mundi et gentium. Paris: Collection Sources Chrétiennes.

Thomason, S. G. (ed.) 1997, Contact Languages: A Wider Perspective. Amsterdam/Philadelphia: John Benjamin Publishing Company.

Tribulato, O. (ed.) 2008 Language and Linguistic Contact in Sicily. Cambridge: Cambridge University Press.

Willi, A. 2008 Sikelismos. Sprache, Literatur und Gesellschaft im Griechischen Sizilien. Basel.

Measuring ambiguity and the invention of vowel-writing in Greek

University of Cambridge

As is well known, the Greek alphabet was adapted from the West Semitic (WS) abjad. This abjad is, however, characterised especially early on by an almost complete absence of the denotation of vowel phonemes. Yet on the script’s adaptation for Greek vowel phonemes are consistently represented. Indeed, there are no extant examples of the WS abjad without vowels being used to write Greek (Powell 1991:10). Should the Greek alphabet be seen as a mere refinement of the alphabetic principle, where vowel writing emerges naturally from a Greek speaker hearing and applying the WS abjad to Greek (Jeffery 1961; Faber 1992), or as a radical new development, the creation of one man (Powell 1991:12), whereby ‘for the first time it is possible to puzzle out the sound of written language without being a speaker of that language’ (Powell 1989).

The present paper aims to shed light on this question by looking for structural reasons for why there might have been very few, if any, attempts to write Greek without vowels. In an experimental approach, model abjads are created for Greek on the basis of WS models. These abjads are applied to represent various Archaic and Classical Greek texts, including sections of Homer, Tragedy and Classical historians. The level of ambiguity vis-à-vis the Greek alphabet is measured by comparing the number of alphabetic readings for given abjad forms (= types), e.g. the abjad type πλν can be read as πυλῶν, πηλόν, πολύν, πολλῶν, πολλήν, πόλιν, πόλεων, πλήν, πλέον etc. The results are compared with results from the same exercise conducted on fully vocalised and non-vocalised Biblical Hebrew texts, serving as a proxy for NWS dialects in general.

Preliminary results (see Table) comparing samples of 9,000 words of Genesis, Judges, Herodotus Book 2, and Xenophon’s Anabasis show that while the mean number of solutions in the vocalised / alphabetic texts for types in the non-vocalised / abjad texts is very similar (1.14 and 1.12 for Genesis and Judges, 1.15 and 1.12 for Herodotus and Xenophon), the standard deviation is higher for Greek (0.52 and 0.46 for Herodotus and Xenophon vs. 0.43 and 0.38 for Genesis and Judges). Indeed for one abjad type in the Greek abjad texts there are as many as 10 solutions in the alphabetic text, and high token-frequency types are among those with many readings. It is argued that an important reason why Greek is never observed written with an abjad because the ambiguity for some types was simply too high.

Number of alphabetic readings / vocalisations of abjad / non-vocalised types attested in 9,000 word text samples of Greek and Hebrew

# of readings





























































Faber, Alice. 1992. Phonemic segmentation as epiphenomenon. In Pamela Downing, Susan D. Lima & Michael Noonan (eds.), The Linguistics of Literacy, 111–134. John Benjamins.

Jeffery, Lilian H. 1961. The local Scripts of Archaic Greece: a Study of the Origin of the Greek Alphabet and its Development from the eighth to the fifteenth centuries B.C. Oxford: Clarendon Press.

Powell, Barry B. 1989. Why was the Greek Alphabet invented? The epigraphical evidence. Classical Antiquity 8(2). 321–350.

Powell, Barry B. 1991. Homer and the Origin of the Greek Alphabet. Cambridge: Cambridge University Press.

From pronoun to discourse marker: the rising of ἀλλά ‘but’ and τοι ‘let me tell you’

Universidad Autónoma de Madrid

This study, which is part of a research project on classical Greek discourse markers, has the primary purpose of reconstructing the process by which the old pronouns ἄλλα 'other things' and τοι 'to you' evolved into the coordinating conjunction ἀλλά 'but' and into the discourse marker of appeal τοι '(your) attention!', respectively.

My point of departure is that these pronouns did not evolve into discourse markers because of a sudden and direct shift, but rather through an evolution that went through several phases that are not documented. To reconstruct these phases, we will take as a model similar evolutions observed in the usages that some adverbs show (e.g. οὕτως ‘in this way’ or ‘therefore’) and in the differential characteristics of connective particles (which combine with a coordinating conjunction) and coordination conjunctions (which do not combine with another coordinating conjunction).

I will be paying attention to developments such as that of the ancient adverb καί ‘also, even’ into the coordinating conjunction ‘and’, that of the adverb νῦν ‘now’ which expanded its meaning into ‘as it is (or was)’, and that of the adverbs πρῶτον μέν... ἔπειτα (δέ)… which can refer either to the order of events in a world or to the order of units within the discourse. By reconstructing the evolution of ἄλλα and τοι we hope to be able to construct a pattern for the grammaticalization of lexical elements into discourse markers.


Traugott, Elizabeth Closs & Dasher, Richard B. (2002), Regularity in Semantic Change,


Bowern, Claire & Evans, Bethwyn (eds.) (2015), The Routledge Handbook of Historical

Linguistics, London – New York

The dialect and geographical origin of the inquirers

in the Dodona lead-tablet corpus

Universitad Autónoma de Madrid

 Aristotle University of Thessaloniki

The recent edition of more than 4200 sheets of lead containing oracular texts brought to light at the Sacred House of Dodona —mainly inquiries addressed to Zeus’ oracle— constitutes an important corpus of sources which increase our knowledge about different aspects of religion and the history of the ancient Greek speaking world from the end of the sixth century BC until the middle of the third century BC, period in which these documents are dated.

From the linguistic viewpoint, such lamellae, both those published in 2013 and those edited at previous dates, offer ample evidence illustrating various aspects of the ancient Greek language. In particular, there are numerous data on the geographical origin, the name of the inquirers and the dialect(s) in which they consulted the oracle, either for personal matters or on behalf of an institution or state.

The communication has the objective of drawing a map that reflects the area of influence of the sanctuary of Dodona in the different periods for which we have oracular lamellae. To this end, we will base our investigation on a database that includes the date attributed to each inquiry in the editions, the dialect in which each query was written down and the geographical origin of each inquirer, distinguishing the occasional information explicitly provided by the text of the consultations and the conclusions that can be drawn from the analysis of the dialect and the personal name of each inquirer.


Δάκαρη, Σ. - Βοκοτοπούλου, Ι. - Χριστίδη, Α.-Φ. 2013. Τα χρηστήρια ελάσματα της Δωδώνης των ανασκσφών Δ. Ευαγγελίδη, Τόμοι Ι-ΙΙ. Επιμέλεια Σωτήρη Τσέλικα, Αθήνα, Βιβλιοθήκη της εν Αθήναις Αρχαιολογικής Εταιρείας αρ. 285, 286.

Ελευθεράτου, Σ and Κ. Σουέρεφ (eds.) 2016. Δωδώνη. Το Μαντείο των  Ήχων. Κατάλογος, Αθήνα, Εκδόσεις Μουσείου Ακρόπολης.

Filos, P. 2018. “The dialectal variety of Epirus”, in Studies in Ancient Greek Dialects. From Central Greece to the Black Sea, ed. by G. K. Giannakis, E. Crespo and P. Filos, Berlin/Boston, De Gruyter, 215-247.

Lhôte, É. 2006. Les lamelles oraculaires de Dodone, Genève, Droz.

Méndez Dosuna, J. 2008. “Novedades en el oráculo de Dodona. A propósito de una reciente monografía de Éric Lhôte”, Minerva 21, 51-79.

Méndez Dosuna, J. 2018. “The language of the Dodona oracular tablets: The non-Doric inquiries”, in Studies in Ancient Greek Dialects. From Central Greece to the Black Sea, ed. by G. K. Giannakis, E. Crespo and P. Filos, Berlin/Boston, De Gruyter, 265-296.

Tselikas, Sotirios, 2018. “The Doric dialects in the corpus of the oracular tablets from Dodona”, in Studies in Ancient Greek Dialects. From Central Greece to the Black Sea, ed. by G. K. Giannakis, E. Crespo and P. Filos, Berlin/Boston, De Gruyter, 249-263.

Diglossia and language choice in the bilingual (Greek-Latin) wall-inscriptions from Pompeii.

Universidad de Salamanca

Bilingualism was a widespread phenomenon in Antiquity. Because of that, it is not difficult to find an ancient city, society or culture that had a mixed population, as is the case of Pompeii. Located in a privileged position in the Vesuvian area and close to the mouth of the river Sarno, this city brought together different cultures and, therefore, languages. During the Roman domination, the supremacy of the Latin language above others was undeniable. Nevertheless, the Greek language played an important role in the education of Roman elites, so it never lost its status as a language of culture. At the same time, Greek was considered the language of slaves and lower classes, used by merchants around the Mediterranean and regarded as lingua franca in the relations with the Eastern part of the Empire.

Our main interest in these documents is to study the phenomenon of diglossia or usage of the same language through two different roles or functions, being one of them more prestigious than the other. In this paper, we will discuss some examples from the Pompeian wall-inscriptions that present the Greek language in both the High (H) and the Low (L) levels. The choice of one language or another could be also conditioned by the use of a certain language within a given domain out of tradition or convention. We will provide some examples of specific fields that were commonly written in Greek: the erotic-love inscriptions, including the ones that show the practice of isopsephism (CIL IV 1462, CIL IV 4839, CIL IV 4861, Giordano nº 31), the cult of Isis and other non-roman divinities (CIL IV 4138, CIL IV 4189, CIL IV 5262) and the ἐμνήσθη expressions (CIL IV 4839, CIL IV 4189, CIL IV 6828) among other examples. Since the alternation in use of the Greek and Latin languages and alphabets shown in the wall-inscriptions comes as a result of their improvised nature and unexpected preservation, it is an important aspect when studying these inscriptions.

We will try to show how the Greek language plays a crucial role in the Roman world and to explain the relationship between Greek and Latin language in a Roman city like Pompeii. Due to the variety of linguistic levels available in these documents, we can understand better the situation, use and conception of the Greek language in a very specific period and location. The final goal will be to answer some key questions like to what extent was the knowledge of the Greek language spread among the population of Pompeii.

Adams, J. N. Janse, M. Swain, S. (2002) Bilingualism in Ancient Society: Language Contact and the Written Text, Oxford: University Press.

Adams, J. N. (2003) Bilingualism and the Latin Language, Cambridge: University Press.

Benefiel R. - Keegan P. (2016), Inscriptions in the Private Sphere in the Greco-Roman World, Leiden-Boston: Brill.

Biville, F. (2003) “Le latin et le grec `vulgaires´ des inscriptions pompéiennes” in H. Solin, M. Leiwo and H. Halla-aho (eds), Latin vulgaire-latin tardif VI. Actes du VIe colloque international sur le latin vulgaire et tardif, Helsinki, 29 août - 2 septembre 2000, 219-35, Hildesheim: Olms-Weidmann.

Clackson, J. (2015) Language and Society in the Greek and Roman Worlds, Cambridge: University Press.

Kaimio, J. (1979) The Romans and the Greek Language, Helsinki.

Mullen, A. (2011) “Latin and other languages: societal and individual bilingualism” in Clackson, J. (ed.) A Companion to the Latin Language, Oxford: Blackwell Publishing.

Solin, H. (2012) “On the use of Greek in Campania”, in Leiwo, M. et al., Variation and Change in Greek and Latin, 97-114, Helsinki: Foundation of the Finnish Institute at Athens.

Varone, A. (2017) “Amare. Il linguaggio greco dell’amore” in M. Osanna, C. Rescigno, Pompei e i Greci (cat. mostra Pompei 11 apr.-27 nov. 2017), Electa, 201-207, Milan.

Syntax and semantics of Ancient Greek δεῖνα

University of Bergamo

As well known, the pronoun δεῖνα, whose meaning is ‘such an one, so-and-so, un tel, un tale, der und der’, remains thus far morphologically unclear and etymologically unexplained. The fact that it is always accompanied by the definite article has made scholars think that it could be derived from the wrong division of an unattested and unexpectedly contracted phrase *τάδε *ἔνα > *ταδεενα > *ταδεῖνα, finally reanalyzed as τὰ δεῖνα (Schwyzer 1990: 612). This process is supposed to be parallel to that at the origin of the demonstrative adjective *ἐκε-ενος > ἐκεῖνος. It is also interesting to observe that ἐκεῖνος can be used instead of δεῖνα, even though very infrequently: see Aristophanes, 

However, a careful attempt at analyzing semantics and syntax of this pronoun is still a desideratum.

The fact that this pronoun is frequently described as corresponding to Lat. quidam is only partially correct: whereas “only by quidam is made clear that the speaker is able to identify the referent of the indefinite pronoun” (Bertocchi, Maraldi, Orlandini 2010: 55), this is not always true for δεῖνα, as shown in the two following examples:

1. Aristophanes, Thesm. 619-622

Τίς ἐστ’ ἀνήρ σοι;

                                    τὸν ἐμὸν ἄνδρα πυνϑάνει;

τὸν δεῖνα γιγνώσκεις, τὸν ἐκ Κοϑωκιδῶν;

τὸν δεῖνα; ποῖον;

                                    ἐσϑ’ ὁ δεῖν’, ὃς καί ποτε

τὸν δεῖνα, τὸν τοῦ δεῖνα –

                                                ληρεῖν μοι δοκεῖς


2. Demosthenes, 13, 5

τοὺς στρατηγοὺς κρίνετε, καὶ περίεσϑ’ ὑμῖν ἐκ τῶν πραγμάτων “ὁ δεῖνα τοῦ δεῖνος τὸν δεῖνα εἰσήγγειλεν”, ἄλλο δ’ οὐδέν.

In these two examples, it is clear that the pronoun itself is not sufficient to establish a relation based on a shared knowledge between the speaker and the hearer.

Even in late Greek δεῖνα is attested but its meaning dramatically changes according to patterns of change that are not clear.

In this paper I want investigate in detail the semantic and syntactic conditions under which δεῖνα is used and how it could develop its late meaning. Some attempts will be made to possibly envisage the pragmatic and stylistic value associated to it.


Bertocchi Alessandra, Maraldi Mirka, Orlandini Anna. 2010. “Quantification”. In: Baldi Philip and Cuzzolin Pierluigi (eds.), New Perspectives on Historical Latin Syntax. Vol. 3: Constituent Syntax: Quantification, Numerals, Possession, Anaphora. Berlin and New York, de Gruyter: 19-174.

Schwyzer, Eduard. 1990. Griechische Grammatik. Erster Band. Sechste, unveränderte Auflage.   München, Beck.

Accusative of respect in Homeric Greek as evidence for language contact

Università per Stranieri di Siena

One of the most famous features of Homeric Greek is the use of a construction including an intransitive predicate and a noun in the accusative case that restricts the force of the predicate to a part or attribute of the subject. This accusative is mainly used to express an inalienable possession as is the case with body parts:

ὃ δὲ φρένα τέρπετ᾽ ἀκούων. (Il. 1.474)

“and his heart was glad, as he heard”

The accusative of respect may have originated in the double accusative construction of the whole and the part (σχῆμα καθ᾽ ὅλον καὶ μέρος). When the double accusative construction undergoes passivization, the whole (the person) becomes the subject, while the part (the body part) remains in the accusative, producing the accusative of respect. Once the category was established, it was extended beyond the passive construction and it was added also to adjectives.

In Homer the double accusative construction of the whole and the part is limited to inalienable possession (mostly body parts):          

Τρῶας δ᾽ ἄχος ἔλλαβε θυμόν. (Il. 14.475)

“and sorrow seized the hearts of the Trojans

The fact that passivization is not possible for both accusative arguments shows that they have different behavioral properites: the body-part noun which in the active is in apposition with the accusative person noun remains accusative in the passive instead of continuing to share the case of the (now nominative) person noun.

            Also in Hittite the accusative of respect arises through the passivization of the double accusative construction and occurs with predicates denoting a state or a change of state (middle verbs, participles and adjectives):

VBoT 24 III 11-13

11        namma ANA UDUḪI.Aištarna paimi nu=kan kui[š]

12        UDUiyanza IGI[ḪI].A-wa dUTU-i neanza

13        nu=šši=kan SÍGḫuttulli ḫūittiyami

 “Then I go among the sheep and I pluck a tuft of wool from what sheep is turned with its eyes toward the sun”

The construction with the accusative of respect NOM. šakuwa nai- (middle) ‘to be turned as far as the eyes’ can be compared with the double accusative construction ACC. šakuwa nai- (active) ‘to turn someone, (his) eyes’:

KUB 23.72 rev. 62

[n=an=kan IGI]ḪI.A-ŠU ḪUR.SAG-i le naištani

“Do no turn his eyes to the mountain!”

Based on the evidence of Hittite, Luvian and Homeric Greek, the aim of this paper is to suggest that that the accusative of respect was an areal feature of some languages spoken in the area of eastern Anatolia in the second and first millenia B.C.E.


R. Blümel, “Der Ursprung des griechischen Bereichsakkusativs und anderes”, IF 33, 1913-14, 1-96.

K. Brugmann, “Der sogenannte Akkusativ der Beziehung im Arischen, Griechischen, Lateinischen, Germanischen”, IF 27, 1910, 121-151.

P. Cotticelli-Kurras, “"Schema", Akkusativ der Beziehung oder (sekundäre) Prädikation?”, in: H. Marquardt et al. (eds.), Anatolica et indogermanica. Studia linguistica in honorem Johannis Tischler septuagenarii dedicata, Innsbruck 2016, 27-42.

B. Delbrück, Vergleichende Syntax der indogermanischen Sprachen, I. Band, Straßburg 1893.

E. A. Hahn, “Partitive Apposition in Homer and the Greek Accusative”, TAPhA 85, 1954, 197-289.

P. Högemann, “Das ionische Griechentum und seine altanatolische Umwelt im Spiegel Homers”, in: M. Witte – St. Alkier (Hrsg.), Die Griechen und der Vordere Orient. Beiträge zum Kultur- und Religionskontakt zwischen Griechenland und dem Vorderen Orient im 1. Jahrtausend v. Chr. (OBO 191), Freiburg (Schweiz)-Göttingen 2003, 1-24.

B. Jacquinod, “Analyse syntaxique de la mise au même cas du complément du tout et du complément de la partie en grec ancien”, in: A. Rijksbaron (ed.), In the footsteps of Raphael Kühner. Proceedings of the International Colloquium in Commemoration of the 150th anniversary of the publication of Raphael Kühner’s Ausführliche Grammatik der griechischen Sprache. II. Theil: Syntaxe, Amsterdam 1986, 135-145.

B. Jacquinod, Le double accusatif en grec d'Homère à la fin du Ve siècle avant J.-C., Louvain-la-Neuve, 1989.

D. Romagno, “The Accusative of Respect in Ancient Greek: Animacy Hierarchy, Semantic Roles and Event Type”, SSL 55/1, 2017, 65-95.

C. Rosen, “‘Accusativus graecus’: a relational analysis”, in: R. L. Brown et al. (eds.), Proceedings of the 1976 Mid-America Linguistic Conference, Minneapolis, Minnesota 1977, 285-296.

S. G. Thomason – T. Kaufman, Language Contact, Creolization, and Genetic Linguistics, Berkeley 1988.

C. Watkins, “An Indo-European linguistic area and its characteristics”, in: A. Aikhenvald – R. M. W. Dixon (eds.), Areal Diffusion and Genetic Inheritance. Problems in Comparative Linguistics, Oxford 2001, 44-63.

Grassmann’s Law as “weak phonetic change”

University of Messina/University ”La Sapienza”, Rome/University ”La Sapienza”, Rome

1. According to Ohala (1989; 1993), dissimilation processes, especially involving non-adjacent segments, are highly unnatural, in the sense that the frequently invoked principles of speech production are unable to explain such a type of phonetic change. Rather, it represents the outcome of  the listener’s mis-application of corrective processes. Acoustic-perceptual cues of some features (glottalization, pharyngalization, retroflexion, nasalization, aspiration) spread over beyond the immediate “hold” of the segment they are distinctive on. Therefore, the listener can interpret the original presence of one of these features on the segment in which it was distinctive as it were the outcome of an erroneous speaker’s production; consequently, it undoes such a feature through a mechanism of hypercorrection. For example, an original form *phath- could be misinterpreted by the listener as the result of an erroneous speech production, originated by the spread of the second segment [+aspirated] feature (h-) on the first one (*p-).

2. Such an explanation could shed new light on the well-known dissimilation process of Ancient Greek, the so-called Grassmann’s Law (= GL), the phonetic change in virtue of which in an original diaspirate root a (regressive) dissimilation process takes place: τίθησι < *θίθησι, τριχóς < θριχóς (cf. nom. θρίξ), τρέφω  < *θρέφω and so on.

Greek inscriptions from different dialectal areas show in some cases the preservation of the (presumably) original diaspirate roots (see e.g. Att. hέχει, καθέχει; καταθιθέναι; θρεφθε‚ς etc.); furthermore, diaspirate non original roots are also attested, in which original forms such as *path-/*phat- are both represented as phath- (see e.g. Att. hαρ[ιθμóν; θρέφος;  χαλχί[ον] etc.). Such evidence points the speaker’s entropic trend to spread over the laryngeal specifications beyond the affected segment. In contrast with this, dissimilated forms such path- < *phath, with a regular application of GL, represent the listener’s syntropic trend, which aims to (hyper)correct all the diaspirate forms (both the original ones and those “wrongly” produced in the speech).

3. The interplay between the entropic speech production and the syntropic listener’analysis produces a long term syncronic variation between diaspirate and dissimilated forms, until the dissimilated forms imposes the template path- as a rule. The delay with which this type of linguistic change spreads out in the lexicon could be due to the lack of an articulatory basis of the dissimilatory processes, which are rather grounded on the listener’s inability to parse the acoustic signal appropriately. The same motivations could explain the likely retention of the original diaspirate roots in Mycenaean and in Homeric Greek, where both (-)h- and aspirated voiceless obstruents seem not to undergo dissimilation.    

4. Lastly, dissimilation as listener’s mis-perception could enlighten the instability in the dissimilation processes in some compounds (e.g. ἐκεχειρία alongside ἐχέθυμος and ἐχέφρων), where GL is anything but systematic. One could bring into play the presence of a morpheme-boundary in these forms between the two [+aspirated] segments as an actual obstacle towards the listener’s effort to (hyper-)correct through a dissimilatory process.


Abramson, Arthur S., Whalen, Doug H. 2017, Voice Onset Time (VOT) at 50: Theoretical and practical issues in measuring voicing distinctions, «Journal of Phonetics» 63: 75-86.

Albano Leoni, Federico 2001, Il ruolo dell’udito nella comunicazione linguistica. Il caso della prosodia, «Rivista di Linguistica» 13, 1: 45-68.

Beekes, Robert, Etymological Dictionary of Greek (Leiden Indo-European Etymological Dictionary), Leiden, Brill, 2009.

Belardi, Walter, L’etimologia nella storia della cultura occidentale, 2 voll., Roma, Il Calamo, 2002.

Bettarini, Luca 2017, Lingua e testo di Ipponatte (“SYNCRISIS. Biblioteca di studi e ricerche sull’antichità classica”), Pisa, Serra.

Bile, Monique 1988, Le dialecte crétois ancien. Étude de la langue des inscriptions. Recueil des inscriptions postérieures aux IC, Paris, École Française d’Athènes.

Blevins, Juliette and Garrett, Andrew 1998, The origins of consonant-vowel metathesis, «Language» 74, 3, 508-556.

Blümel, Wolfgang 1982, Die aiolischen Dialekte. Phonologie und Morphologie der inschriftlichen Texte aus generativer Sicht, Göttingen, Vandenhœck & Ruprecht.

Clackson, James 2002, The writing of χσand φσ for andy, «Glotta» 78: 22-35.

Collinge, Neville E. 1985, The laws of Indo-European, Amsterdam-Philadelphia, Benjamins.

De Mauro, Tullio 1994, Intelligenti pauca, in P. Cipriano, P. Di Giovine, M. Mancini (a c. di), Miscellanea di studi linguistici in onore di W. Belardi, vol. 2, Roma, Il Calamo: 865-875.

Dubois, Laurent 1988, Recherches sur le dialecte arcadien. I Grammaire. II Corpus dialectal. III Notes – Index – Bibliographie, Louvain-La-Neuve, Peeters.

De Decker, Filip 2015, Another attempt at a chronology for Grassmann’s Law in Greek, «Journal of Indo-European Studies» 43, 1-2: 140-177.

Garrett, Andrew 2015, Sound change, in C. Bowern and B. Evansin (eds.), The Routledge handbook of historical linguistics, London, Routledge: 227-248.

Grassmann, Hermann 1863, Ueber die Aspiraten und ihr gleichzeitiges Vorhandensein im An- und Auslaute der Wurzeln, «Kuhns Zeitschrift» 12: 81-138.

Hock, Hans Henrick (19912), Principles of historical linguistics, Berlin-New York, de Gruyter.

Hoenigswald, Henry M. 1965, A Property of “Grassmann’s Law” in Indic, «Journal of the American Oriental Society» 85, 159-60.

Kiparsky, Paul 1973, On comparative linguistics: the case of Grassmann’s Law, in Th. E. Sebeok (ed.), Current Trends in Linguistics. Vol. 11: Diachronic, Areal, and Typological Linguistics, The Hague-Paris, Mouton: 115-134.

Lehmann, Winfred P. 19923, Historical Linguistics, London-New York, Routledge.

Lejeune, Michel 1972, Phonétique historique du mycénien et du grec ancien, Paris, Klincksieck.

Levine, Saul 1969, Grassmann’s “Law” in the early semitic loan-word χιτών, κιθών, «Studi Micenei ed Egeo-Anatolici» 8: 66-75.  

Malkiel, Yakov 1983, Multiple versus simple causation in linguistic change, in Id., From particular to general linguistics. Vol. 3, Essays 1965-1978, Amsterdam-Philadelphia, Benjamins: 251-268.

McMahon, April M. S. 1994, Understanding language change, Cambridge, CUP.

Mendez Dosuna, Julian 1985, Los dialectos dorios del noroeste. Gramatica y estudio dialectal, Salamanca, Ed. Universidad de Salamanca.

Miller, D. Gary 1977, Was Grassmann’s Law reordered in Greek ?, «Historische Sprachforschung» 91: 131-158.

Ohala, John J. 1981, The listener as a source of sound change, in C. S. Masek, R. A. Hendrick, M. F. Miller (eds.), Papers from the Parasession on Language and Behavior, Chicago, Chicago Linguistic Society: 178-203.

Ohala, John J. 1989, Sound change is drawn from a pool of synchronic variation, in L. E. Breivik and E. H. Jahr (eds.), Language change: Contributions to the study of its causes, Berlin, de Gruyter: 173-198.

Ohala, John J. 1993, The phonetics of sound change, in Charles Jones (ed.), Historical Linguistics: Problems and Perspectives, London: Longman: 237-278.

Ohala, John J. 2003, Phonetics and Historical Phonology, in B. D. Joseph and R. D. Janda (eds.), The Handbook of Historical Linguistics, Malden (MA)/Oxford, Blackwell: 669-686.            

Pozza, Marianna 1998, La legge di Grassmann riconsiderata alla luce delle nuove acquisizioni dell’indoeuropeistica, tesi di laurea, Roma, Università degli Studi di Roma “La Sapienza”.

Ruijgh, Cornelius J. 1967, Études sur la grammaire et le vocabulaire du gréc mycénien, Amsterdam, Hakkert.

Rix, Helmut (hg.) 20012 ([19981]), Lexikon der indogermanischen Verben (LIV). Die Wurzeln und ihre Primärstammbildungen. Unter Leitung von Helmut Rix und der Mitarbeit vieler anderer bearbeitet von Martin Kümmel, Thomas Zehnder, Reiner Lipp, Brigitte Schirmer, Wiesbaden, Dr. Ludwig Reichert.

Salmons, Joseph 2010, Segmental Phonological Change, in S. Luraghi and V. Bubenik (eds.), Continuum Companion to Historical Linguistics, London-New York, Continuum: 106-116.

Sanchez Garrido, M. del Carmen (1988), Norma y uso y método generativo: el caso de la ley de Grassmann, «Revista española de lingüística» 18,1: 149-167.

Schwyzer, Eduard 1939, Griechische Grammatik auf der Grundlage von Karl Brugmanns Griechischer Grammatik. I Bd. Allgemeiner Teil, Lautlehre, Wortbildung, Flexion, München, Bek’sche.

Threatte, Leslie 1980, The grammar of Attic inscriptions. Vol. I. Phonology, Berlin-New York, de Gruyter.

Thumb, Albert/Kieckers, Ernst (19322), Handbuch der griechischen Dialekte, Vol. 1, Heidelberg, Winter.

Trask, Robert Lawrence 1996, Historical Linguistics, London, Arnold.

Wackernagel, Jacob, 1916, Sprachliche Untersuchungen zu Homer, «Glotta» 7, 2-3: 161-319.

The augment in Epic Greek.

Universiteit Gent - FWO Vlaanderen

Greek past tense forms of the indicative (imperfect, aorist and pluperfect) are marked by the addition of a prefix - to the verbal form. While mandatory in Classical Greek prose, the prefix appeared facultative in earlier Greek poetry and this absence is often explained as a poetic licence and/or a metrical tool. I argue that the use in the oldest Greek literature (Homer, Hesiod, the Homeric Hymns and the fragments of the Epic Cycle) was not random but explainable by morphometric, syntactic and semantic reasons. This prefix gradually evolved into the mandatory marker of past tense and the earlier rules governing the use were no longer understood. As a result, the absence in epic Greek was reinterpreted as a poetic licence and was imitated by later (epic) poets. In this presentation, I focus on the earliest epic texts. First, I treat the transmitted forms and by using metrical rules, bridges and caesurae (such as those posited by Varro, Hermann 1805:692-693, Giseke 1864:128-138, Meyer 1884:980 - see the chapters in Maas 1960, Snell 1982, West 1982 and Oswald 2014) I decide whether the forms are secured by the metre or not. When they are not, I check if their value can be determined by an internal reconstruction (i.e. comparing the insecure forms to the secure forms of the same paradigm and looking at the position in the verse and at the word(s) preceding the verb form, as was done in Taida 2007, 2010 and De Decker 2016, 2017, based on Barrett 1964:361-362). Then, I proceed to the actual analysis of the established. Starting from earlier scholarship on the issue that described the augment as a deictic marker that marked the completion of the action in the presence of the speaker (Platt 1891:227; Bakker 2005:147), I show that and explain why the augment is used/preferred in

a)   verbal forms that would otherwise be monosyllabic (contrary to Wackernagel 1916 and Strunk 1967, 1987 argued, this rule also applies to long monosyllables);

b)   similia and gnomic aorists (Platt 1891, Bakker 2005:131-135, Faulkner 2005:68-69);

c)   speech introductions with an addressee (De Decker 2015);

d)   actions in the immediate past (Platt 1891; Drewitt 1912a,b; Bakker 2005:114-153);

e)    in speeches (Koch 1868, Bottin 1969:110-128, Basset 1989);

f)     when new and/or important information was communicated (Mumm 2004);

and it is avoided in verb forms

a)  referring to repeated actions in the past, often with αἰεί;

b)  with the -σκ-suffix, referring to repeated actions or one action repeated by many (Pagniello 2007);

c)  occurring in narrative and mythical stories (Koch 1868, Bottin 1969:110-128, Basset 1989);

d)  describing a remote past, mythical stories or timeless activities (Hoffmann 1967, West 1989);

e)  describing background actions, especially in subordinate clauses (De Decker 2016:298-299);

f)  followed by a clitic, such as δέ, γάρ, μέν, τε, ἄρα and ῥα (Drewitt 1912b:104);

g)  appearing in a series of past tense forms and preceded by an augmented verb form                                          (Kiparsky 1968, Luraghi 2014);

h)  in negative sentences, especially in narrative passages (Bakker 2005:126, De Decker 2017:144-146);

i)  in the beginning of a verse or sentence (as had been noted already as early as the Byzantine scholars).

An example will make the distinction between foreground and background clear:

ὃς τότ' ἐν ἀκροπόλοις ὄρεσιν πολυπιδάκου Ἴδης

βουκολέεσκεν βοῦς δέμας ἀθανάτοισιν ἐοικώς.

τὸν δὴ ἔπειτα ἰδοῦσα φιλομμειδὴς Ἀφροδίτη

ἠράσατ', ἐκπάγλως δὲ κατὰ φρένας ἵμερος εἷλεν.

"(Ankhises, who) pastured his cattle on the top of mountains of Ida, rich in fountains, in stature he resembled the gods. Then lovely-smiling Aphrodite saw him, longed deeply for him and a strong desire for him took mercilessly control over her mind." (Homeric Hymn to Aphrodite, HH 5,54-57).

This passage taken from the Homeric Hymn to Aphrodite clearly shows the use and absence of the augment: the unaugmented βουκολέεσκεν describes the background and habitual action of Ankhises's being a shepherd: he has always been herding his cattle on the mount Ida; the augmented forms ἠράσατ' and εἷλεν describe something new and unexpected, namely his sudden erotic desires for Aphrodite as soon as he notices her.


Bakker, E. 2005.  Pointing at the past: from formula to performance in Homeric poetics. Cambridge, MA.

Barrett, W. 1964. Euripides Hippolytos. Edited with Introduction and Commentary. Oxford.

Bartolotta, A. & Buijs, M. & Kölligan, D. 2017. Modality and Injunctive in Ancient Greek. Presentation during the Conference "Charting the Semantic Space of Ancient Greek Modality", on May 6th, 2017 in Paris.

Basset, L. 1989. L'augment et la distinction discours/récit dans l'Iliade et l'Odyssée. In: Casevitz, M. Études homériques. Lyon. 9-16.

Bottin, L.  1969.           Studio dell’aumento in Omero. Studi Micenei ed Egeo Anatolici 10.69-145.

Chantraine, P. 1948.   Grammaire homérique. Paris.

Chantraine, P. 1953.   Grammaire homérique. Tome II : Syntaxe. Paris.

De Decker, F. 2015. The Augment in Homer, with Special Attention to Speech Introductions and Conclusions. JournaLIPP 4. 53-71, accessible online:

De Decker, F. 2016. The Augment Use in Iliad 6: an Evidential Marker? Les Études Classiques 84. 259-317.

De Decker, F. 2017. Ὅμηρον ἐξ Ὁμήρου σαφηνίζειν: an analysis of the augment use in Iliad 1. Journal of Indo-European Studies 45. 58-171.

De Lamberterie, C. 2007. L’augment dans le texte arménien de l’Évangile. Revue des Études arméniennes 30. 31-57.

Drewitt, J. 1912a. The Augment in Homer. Classical Quarterly 6. 44-59.

Drewitt, J. 1912b. The Augment in Homer (continued). Classical Quarterly 6. 104-120.

Drewitt, J. 1913.          A Note on The Augment. Classical Philology 8. 349-353.

Faulkner, A. 2005. Aphrodite’s Aorists: Attributive Sections in the Homeric Hymns. Glotta 81. 60-79.

García Ramón, J. 2012.           TAM, Augment and Evidentiality in Indo-European. Unpublished Handout from 2012.

Giseke, B. 1864. Homerische Forschungen. Leipzig.

Hajnal, I. 2016. Induktive versus abduktive Rekonstruktion: das Beispiel des griechischen Augments. Indogermanische Forschungen 121.435-453.

Hermann, G. 1805. Orphica. Leipzig.

Hoffmann, K. 1967.     Der Injunktiv im Veda. Heidelberg.

Kiparsky, P. 1968. Tense and Mood in Indo-European Syntax. Foundations of Language 4. 30-57.

Koch, K. 1868. De augmento apud Homerum omisso. Braunschweig.

Luraghi, S. 2014. Conjunction Reduction. In: Giannakis, G. The Encyclopaedia of Ancient Greek Language and Linguistics. Leiden. 362-363.

Maas, P. 1923. Griechische Metrik. Leipzig.

Maas, P. 1960. Greek Metre. Translated by Hugh Lloyd Jones. Oxford.

Meyer, G. 1884. Zur Geschichte des griechischen und des lateinischen Hexameters. Sitzungsberichte der Bayerischen Akademie der Wissenschaften, Philosophisch-Historische Klasse 4. 979-1089.

Mumm, P. 2004. Zur Funktion des homerischen Augments. In: Krisch, T. Analecta homini universali dicata. Festschrift für Oswald Panagl. Stuttgart. 148-158.

Oswald, S. 2014. Metrical Laws. In: Giannakis, G. (ed). The Encyclopaedia of Ancient Greek Language and Linguistics. Leiden. 419-423.

Pagniello, F. 2007. The past-iterative and the augment in Homer. Indogermanische Forschungen 112. 105-123.

Pelliccia, H. 1985. The Structure of the Archaic Greek Hymns. PhD Thesis Yale.

Platt, A. 1891. The Augment in Homer. Journal of Philology 19. 211-237.

Rzach, A. 1876. Der Dialekt des Hesiodos. Leipzig.

Snell, B. 1982. Griechische Metrik. Göttingen.

Strunk, K. 1967. Wortstruktur und Pronomen im Altpersischen. Kuhns Zeitschrift für Vergleichende Sprachforschung 81. 265-275.

Strunk, K. 1987. Ergänzende Beobachtungen zu Wortumfang und Wortform. Kuhns Zeitschrift für Vergleichende Sprachforschung 100. 323-338.

Taida, I. 2007. Elision and Augment in the Homeric Hymn to Demeter. Exemplaria Classica 11. 3-12.

Taida, I. 2010. Augment in the Homeric Hymn to Hermes. Hermes 138. 250-258.

Troxler, H. 1964.         Sprache und Wortschatz Hesiods. Zürich.

Wackernagel, J. 1892. Über ein Gesetz der indogermanischen Wortstellung. Indogermanische Forschungen 1. 333-437.

Wackernagel, J. 1916. Wortumfang und Wortform. Göttingische Gelehrte Nachrichten 1906. 147-184.

West, M. 1966. Hesiod. Theogony. Oxford.

West, M. 1978. Hesiod. Works and Days. Oxford.

West, M. 1982. Greek Meter. Oxford.

West, M. 1989. An Unrecognized Injunctive Usage in Greek. Glotta 67. 135-138.

West, M. 2003. Greek Epic Fragments from the Seventh to the Fifth Centuries BC. Cambridge, MA.

West, M. 2013. The Epic Cycle: a Commentary on the Lost Troy Epics. Oxford.

Willi, A. 2007. Of aspects, augments, aorists – or how to say to have killed a dragon. In: George, C. & Nielsen, B. & Ruppel, A. & Tribulato, O. Greek and Latin from an Indo-European Perspective. Cambridge. 34-48.

Willi, A. 2018. The Origins of the Greek Verb. Cambridge.

Lexical and syntactic constrictions for the derivation of the verbal nouns in –τις / -σις

Universidad Autónoma de Madrid

The traditional description of verbal nouns formed with the suffix -τις / -σις claims that it forms de-verbal formations with the meaning of action (e.g. Schwyzer 1953: 504). However, in some cases, other meanings have been recognized for these formations, as instrument, in examples such as those of (1), and result, in examples such as (2); in this last case, the meaning of the formations seems to overlap with those formed with the suffix –μα, the prototypical form to derive result nouns from verbal stems.

(1)       δόσις “gift” <  δίδωμι “to give”

            σίτησις “food” < σιτέω “to feed”

            πόσις  “drink” cf. πίνω “to drink”

            ἄρυστις “spoon” < ἀρύω “to draw”

            μάστις “whip” < μάω “to seek, to pursue”


(2)        ἄροσις “ground” < ἀρόω “to plough”

            γύμνωσις “defenceless side of the warrior” < γυμνόω “to strip”

            περίστασις “surrounding” cf. περιίστημι “to place round”

            τάξις “order, organization” <  τάσσω “to arrange, to put in order”

            τρῆσις “perforation” cf. τετραίνω “to pierce”

            On the basis of these cases, it was suggested that, perhaps originally, the suffix did not form only action nouns, but that it could have other meanings (e.g. Chantraine 1933: 275, Holt 1940). Nevertheless, many of those derived forms are not old formations, but were created all along the history of the Greek language. Therefore, the existence of forms as those of (1) and (2) cannot be explained as an archaism. On the contrary, there must be a historical linguistic procedure that connected the main meaning of action with that of instrument or result. Moreover, there must be a reason why the formations in -τις / -σις can be, apparently, sometimes used for instruments and other times for result. 

            In this paper, I want to explore how the lexical and relational characteristics of the verbal basis from which -τις / -σις nouns are derived condition the cognitive procedures that allow the different metonymic reinterpretation of action nouns. I will argue that in all those cases there is a single procedure by which the action is reinterpreted as the object of the action. The pretended difference between instrument-like formations and result-like formations can be reduced to a single semantic metonymic phenomenon.


Chantraine, Pierre (1933): La formation des noms en grec ancien. París: Klincksieck.

Holt, Jens (1940): Les noms d’action en –σις. Aarhus : Universitetsforlaget.

Schwyzer, Eduard (1953): Griechische Grammatik I. Munich : C.H. Beck

The Greek suffix -ινδα within the Micro-Asiatic multilingual context

Università degli Studi di Milano

The Greek suffix -ινδα, which characterises adverbs denoting games (such as βασιλίνδα “king of the castle”, κρυπτίνδα “hide and seek”, on these adverbs see Dedè 2016), has a phonological shape difficult to explain within Greek and Indo-European phonology, as already observed by Pierre Chantraine (1933:277): “l’origine de ces adverbes est claire, mais le suffixe qui s’y trouve impliqué presente une structure singulière. Aucune analyse ne permet d’expliquer le groupe -νδ-”. In the same context Chantraine, also quoting a statement by Herodotus, suggested that the suffix has its ultimate origin in Asia Minor, particularly in Lydia.

A fact which deserves consideration, although it is in itself not enough to demonstrate the Asiatic origin of the suffix, is the presence in the area of several place names ending in -ινδα (e.g. Πιγίνδα in Caria, Σίνδα in Pisidia). Another element, which encourages to elaborate on this perspective is the hypothesis – in our view correct – recently put forth by Paola Dardano (2011), according to which the suffix -ίδᾱς, which in the Homeric poems forms patronymics and is later refunctionalised as a suffix for deriving proper names, would have entered Greek via the Lydian language.

In my talk I will take into account and discuss the hypothesis of a Micro-Asiatic origin of the suffix -ινδα, evaluating its position within the context of linguistic contact between Greek and the languages of Asia Minor.


Chantraine 1933 = P. Chantraine, Notes sur les adverbes en -ινδην, -ινδα, -ινδον désignant des jeux, «Révue des Études Grecques» 46 (1933), pp. 277-283.

Dardano 2011 = P. Dardano, I patronimici in -ίδᾱς del greco antico tra conservazione e innovazione, «Res Antiquae» 8 (2011), pp. 41-62.

Dedè 2016 = F. Dedè, Ludonimia e classi lessicali: lo statuto degli avverbi di gioco in -ινδα del greco, in F. Dedè (ed.), Categorie grammaticali e classi di parole. Statuto e riflessi metalinguistici, Roma, Il Calamo 2016, pp. 139-156.

Verba rogandi in the Greek documentary papyri of the Roman and Byzantine periods

University of Cologne

I present a study of the verba rogandi (i.e. ἀξιόω, δέομαι, ἐρωτάω and παρακαλέω) based on the Greek documentary papyri of the Roman and Byzantine Periods (1st c.– 8th c. AD) with two goals:

i)      to document the syntactic patterns found in the documentary texts;

ii)     to examine the pairing of these verbs in request formulaic expressions and their usage in relation to specific sociolinguistic situations.

The documentary papyri belong to the most relevant and copious sources for the study of the ancient everyday language. They contain several constructions different from those of the literary language and allow us to detect some tendencies of the Greek language that would become standard many centuries later. One of these concerns the construction of the verba rogandi with completive clauses introduced by a conjunction in place of an infinitival or participial construction, foreshadowing the disappearance of non-finite complementation in later Greek (Horrocks, 2007). Taking into consideration the evidence from the documentary texts of the Roman and Byzantine periods, I will firstly illustrate the complementation patterns of each relevant verb: they occur with a subordinate construction introduced by the declarative, final and consecutive conjunctions ὅτι, ἵνα, ὅπως and ὥστε respectively (e.g. P. Oxy. IV 744, private letter, 1st c. BC, l. 13: ἐρωτῶ σε οὖν ἵνα μὴ ἀγωνιάσῃς - I urge you therefore not to worry), and in the paratactical use with the imperative (e.g. P.Oxy. XXXIV 2731, private letter, 4th/5th c. AD, l. 17-19: π[αρα]κα̣λῶ δέ, ἀντιγράψατέ μοι ὑπὲρ τῆς ὑμῶ[ν] ὑγιείας - I implore you, write back to me about your health).

Secondly, I will specifically analyse the case of the verb ἐρωτάω, whose syntax can be explained not only as an effect of the general tendency mentioned above, but also as a consequence of semantic extension. In this respect, I will discuss Dickey’s (2010) thesis about the Latin influence on the semantic and syntax of ἐρωτάω, when used alone and when paired with the verb παρακαλέω (i.e. ἐρωτάω καὶ παρακαλέω). On the one hand, the pairing of ἐρωτάω and παρακαλέω could be explained as a result of the influence of the correspondent Latin formula rogo atque oro, as Dickey argued. On the other hand, the pairing of these two verbs could be explained as an instance of the more general phenomenon in the papyri to build formulaic expressions by combining two verba rogandi (e.g. ἀξιόω καὶ δέομαι, δέομαι καὶ παρακαλέω) in order to emphasize the tone of a request. This phenomenon can be also observed in relation to the verba iubendi (i.e. βούλομαι καὶ κελεύω, ἐντέλλομαι καὶ ἐπιτρέπω). Furthermore, the use of these formulas in the papyri is related to specific sociolinguistic contexts: they are found only in private letters and in petitions.

Finally, I will explore the hypothesis that these formulas in the context of private letters behave like directive expressions, in other words they occur as fixed expressions in paratactic structures to introduce a request (cf. Leiwo, 2010 and for Latin Risselada, 1993).

Selected References

Cristofaro S. (1996), Aspetti sintattici e semantici delle frasi completive in greco antico, Firenze.

Dickey E. (2010), “Latin Influence and Greek Request Formulae”, The Language of the Papyri, ed. by T.V. Evans and D.D. Obbink, Oxford, 208-220.

Horrocks G. (2007), “Syntax: from Classical Greek to the Koine”, A History of Ancient Greek from the Beginnings to Late Antiquity, ed. by A. Christidis et al., Cambridge, 618- 631.

Leiwo M. (2010), “Imperatives and Other Directives”, The Language of the Papyri, ed. by T.V. Evans and D.D. Obbink, Oxford, 97-119.

Mayser E. (1934), Grammatik der griechischen Papyri aus der Ptolemäerzeit II.3, Berlin – Leipzig.

Palme B. (2009), “The Range of Documentary Texts: Types and Categories”, The Oxford Handbook of Papyrology, ed. by R. Bagnall, Oxford, 149-169.

Risselada R. (1993), Imperatives and other Directive Expressions in Latin, Amsterdam.

Contrastive study of praedicativa in Greek and Latin

University of Santiago de Compostela

The traditional description of the Greek praedicativum has focused largely on the frequent —albeit extremely specific— type of adjective which modifies the subject of an intransitive verb. Both the highly inflectional character of Ancient Greek and the sensitive-to-morphology approach of descriptive grammars (Kühner-Gerth 1966: 273-6; Schwyzer-Debrunner 1966: 26-7, 178-9; Goodwin 1965: 201) have thus led to the praedicativum being addressed mainly as the mere non-syntagmatic counterpart of the attributive adjective. Recent typological studies, however, have sought to downplay significantly the importance of agreement as a main feature here, and to highlight instead the syntactic phenomenon of secondary predication as the core of the praedicativum (also labelled depictive), cf. Denizot (2014: 132).

Henceforth, the parts of speech liable to function as depictive broadly include, in addition to the well-known qualifying adjectives of physical or mental state, a semantically varied range of adjectives, participles, pronouns in their adjectival use (quantifiers, ordinals, emphatic pronouns), preposition phrases, nouns, and even free-from-agreement forms like noun phrases such as genitives of description (Pinskter 1998: 143-6). As a result, the traditional semantic characterization of the praedicativum as expressing a transient property which temporally overlaps with the event is to some extent challenged. This controversial issue is seen particularly well both in temporal adjectives (as ἑσπέριος or vespertinus) and in participles, which, in as far as they are endowed with tense and aspect, may have a temporal reference which does not overlap with the temporal interval of the main predication (Geuder 2002:182-91, Himmelmann & Schultze-Berndt 2005: 40-2). 

On the other hand, optional depictives (He left angry) are often compared with a broad range of adverbials, which are also participant-oriented adjuncts (He left angry/angrily). The distribution of the two constructions has been seen as a typological criterion (He walked at evening // ἑσπέριος / vespertinus). On this note, even if a broad use of praedicativum has been deemed a common trait of both Classical languages, it has been claimed (Kühner-Gerth 1966: 273) that its use in Greek is far more frequent than in Latin. Thus, a corpus-based contribution to the contact between both languages will be conducted here, focussing on the use of depictives (including the much discussed ‘obligatory’ depictives or complementatives: He looked angry, He called him ‘coward’; cf. van der Auwera & Malchukov 2005: 406-10, Pinkster 2015: 30). Attention will be paid primarily to the use of co-predicative participles, as far as such use reflects major differences in tense and aspect parameters between the two languages. Findings from this contrastive analysis will be expressed not only in terms of frequency rates but also as contrastive semantic maps.

The original Greek depictives in the Greek Pseudo-Aristotle’s On the Cosmos (ca. II a.D.) will be traced through its translation into Latin by the Pseudo-Apuleius (ca. II a.D.); conversely, the reflection of the original Latin depictives in the Res Gestae Divi Augusti (ca. I a.D.) will also be examined in its Greek translation as preserved in the Ancyranum inscription. The notorious diachronic and stylistic differences between these two texts will serve to enrich the analysis.


Denizot, C. 2014: “Predicative Constituents” in G. K. Giannakis ed.), Encyclopedia of Ancient Language and Linguistics, 3, Leiden – Boston: 130-3.

Geuder, W. (2002): Oriented Adverbs, Issues in the Lexical Semantics of Event Adverbs (Ph. D. Universität Tübingen), Konstanz.

Goodwin, W.W. (1965): Greek Grammar, New York.

Himmelmann, N. P. & E. Schultze-Berndt (2005), “Issues in the syntax and semantics of participant-oriented
 adjuncts: an introduction”, N. P. Himmelmann and E. Schultze-Berndt (eds.), Secondary Predication and Adverbial Modification, Oxford – New York: 1-67.

Kühner, R. & B. Gerth, 1966: Ausführliche Grammatik der Griechischen Sprache, Darmstadt (third edition).

Nichols, J. (1978): “Secondary Predicates”, Proceedings of the Annual Meeting of the Berkeley Linguistics Society, 4(4): 114-127.

Pinkster, H. (1990): Latin Syntax and Semantics, London–New York.

Pinkster, H. (2015): The Oxford Latin Syntax, Oxford.

Schwyzer E. & A. Debrunner 1966: Griechische Grammatik, München (third edition).

van der Auwera J. & A. Malchukov (2005), “A semantic map for depictive adjectivals”, N. P. Himmelmann and E. Schultze-Berndt (eds.), Secondary Predication and Adverbial Modification, Oxford – New York: 393-421.

Etas in Archaic Attica? The pre-Euclidean use of <H> as a long vowel

Jesus College, University of Cambridge

It is generally assumed that the epichoric Attic script lacks graphemes for the long vowels, thus using epsilon and omicron to write their short, long-open and long-closed allophones respectively. This situation didn’t change until the Ionic script was ultimately adopted in Athens after Euclides’ reform in 403/2 BCE, when both eta and omega were finally introduced. This model of explanation, however, overlooks a few instances where we might recognise the letter eta in archaic inscriptions from Attica. The earliest examples might be an ostrakon from the Sanctuary of Zeus in Mount Hymettos with the dedication ΣΗΜΙΟΙ ΔΙ (Langdon 1976, no.2) and three other inscriptions in this corpus where the grapheme <H> is thought not to render an aspiration (Langdon 1976, nos.3, 9 & 73). Nevertheless, no satisfactory explanation has been given to these inscriptions since there has been no comprehensive research on the subject.

A study of these characteristics is inevitably affected by the difficulties that the fragmentary epigraphic samples available present. Some of the inscriptions are preserved in such conditions that it is very difficult to assess the real phonetic value behind a certain sign, especially in a case like this, where the grapheme <H> can be used for two very distinct values such as the initial aspiration and /ɛ:/. Moreover, it could be the case that <H> as /ɛ/ is underrepresented in the inscriptions found in the Attic corpora. This cannot, however, be the explanation for this phenomenon, since there are multiple instances of <E> rendering /ɛ:/ in archaic Attic inscriptions. Not to mention the additional complications given the context where the inscriptions mentioned above have been found. It is not unheard of that sanctuaries are places where foreign people may deposit votive offerings using their own script (cf. the Semitic inscriptions in the Sanctuary of Apollo Daphnephoros in Eretria and Kommos: Kenzelmann Pfyffer, Theurillat, and Verdan 2005, no.66; Csapo, Johnston, and Geagan 2000, no.1).

Having these difficulties in mind, the aim of this paper is to examine the inscriptions that include these and other examples of possible etas –rendered by the aforementioned grapheme or a different allograph for eta– that appear in Attica before the Euclidean script reform. Thus we can assess whether the previous explanations given to the inscriptions of Mt. Hymettos are plausible and clarify the reasons behind this grapholinguistic phenomenon.


Bartoněk, A. (1966) Development of the long-vowel system in ancient Greek dialects. Spisy University J.E. Purkyně v Brně, Filosofická fakulta 106. Praha.

Csapo, E., Johnston, A.W. and Geagan, D. (2000) “The Iron Age Inscriptions”. In Kommos IV. The Greek Sanctuary, part 1, ed. by Joseph W. Shaw and Maria C. Shaw, 101–134. Princeton.

D’Angour, A.J. (1999) “Archinus, Eucleides and the Reform of the Athenian Alphabet”. Bulletin of the Institute of Classical Studies 43, 109–130.

Guarducci, M. (1995) Epigrafia greca. Vol.1 Caratteri e storia della disciplina.La scrittura Greca dalle origini all’età imperiale. Roma.

Immerwahr, H.R. (1990) Attic Script. A survey. Oxford.

Jeffery, L.H. and Johnston, A.W. (1990) The local scripts of archaic Greece: a study of the origin of the Greek alphabet and its development from the eighth to the fifth centuries B.C. Revised ed. Oxford.

Kenzelmann Pfyffer, A., Theurillat, T. and Verdan, S. (2005) “Graffiti d’époque géométrique provenant du sactuaire d’Apollon Daphnéphoros à Erétrie”. Zeitschrift für Papyrologie und Epigraphik 151, 51–83.

Langdon, M.K. (1976) A Sanctuary of Zeus on Mount Hymettos. Hesperia: Supplement XVI.

Threatte, L. (1980) The Gammar of Attic Inscriptions. Vol.I Phonology. Berlin.

Young, R.S. (1940) ”Excavation on Mt. Hymettos, 1939”. American Journal of Archaeology 44, 1-9.

A corpus-based approach to language contact:  Greek and Latin in late antiquity

Univeristy of Genoa; Università degli studi del Piemonte Orientale

The aim of this talk is to examine different contact phenomena between Greek and Latin as evidenced from an annotated corpus, named Textual bilingualism in Latin, which includes Late Latin literary texts (from the 3rd to the 7th century AD) containing different cases of switches into Greek. This corpus, which has been specifically created by our research group for the study of Greek-Latin contact, is already available online as a new tool for historical sociolinguistics focusing on bilingualism in the ancient Mediterranean world and on historical code-switching.

In the first part, we will discuss our methodology for the linguistic analysis of language contact between Greek and Latin, which implies a systematic assessment of the quality of data at our disposal and of the specific characteristics of language contact as attested in ancient texts. To get at the heart of the specific nature of language contact in our corpus, we will discuss the individual features of written code-switching as distinct from conversational code-switching. As is known, written code-switching has to be considered as a special instance of language mixing: since it does not provide uncontroversial direct representations of speech, its study requires a number of methodological cautions (cf. Adams, Janse & Swain 2002 for important remarks on written code- switching). In particular, it needs to be described and analyzed within the larger scenario of the literacy practices of which it is a part, also including graphical and philological issues.

Moreover, we will deal with the characteristics of what we have recognized as two distinct products of language contact, namely code-switching and single-word switches. In our approach, code-switching refers to cases of switching from Latin to Greek in the form of a sentence, whereas single-word switches correspond to the use of single Greek words into Latin texts. Obviously, this approach to ancient data requires some methodological caution about the distinction between the notion of switch itself and the notion of loanword, an issue which has received much attention with a focus on Latin borrowings in Greek (cf. Dickey 2012, forthcoming). Although a significant body of research has been carried out on the use of Greek in Latin texts (e.g., Biville 2008, Rochette 2010 among many others), corpus-driven data and methodological reflections have something new to bring to the field and can contribute a perspective that is lacking in studies of Greek-Latin contact.

In the second part, we will illustrate the development of the multi-layered tagset specific to contact phenomena between Greek and Latin worked out for our corpus. Drawing on a case study on forms, functions and textual distribution of Greek in our literary texts, we will show how this tool can be used for various types of qualitative and quantitative research on contact phenomena in the past.

The originality of this research lies on the development of a new resource for historical sociolinguistics which permits a corpus-based methodology on a wide selection of ancient texts, also promoting networking between scholars interested in language contact between Greek and Latin.


Adams J. N., Janse M. & Swain S. (eds.) 2002. Bilingualism in ancient society: language contact and the written text. Oxford: Oxford University Press.

Biville F. 2008. Situations et documents bilingues dans le monde gréco-romain. In: F. Biville, J.C. Decourt and G. Rougemont (éds.), Bilingualisme gréco-latin et épigraphie, 35-53. Lyon: Maison de L’Orient.

Dickey E. (2012) Latin loanwords in Greek: a preliminary analysis. In: M. Leiwo, H. Halla-aho & M. Vierros (eds.), Variation and Change in Greek and Latin, 57-70. Helsinki: Papers and Monographs of the Finnish Institute at Athens, vol. XVII.

Dickey E. forthcoming. What is a loanword? The case of Latin borrowings and codeswitches in Ancient Greek. In “Lingue e Linguaggio” 2018/1.

Rochette B. 2010. Greek and Latin bilingualism. In Bakker E.J. (ed.), A companion to the Ancient Greek language, 281-293. Oxford: Oxford University Press.

Anatolian Languages Moulding Greek

Some Intriguing Cases of Heteroclisis in Hipponax’ Fragments

Trinity College Dublin

The question of the influence of Anatolian languages on Ancient Greek has been widely debated in the field of Indo-European linguistics, with scholars such as Beekes, Janse, and Melchert. In particular, the Ionian cities on the coast of Asia Minor constitute the ideal melting pot of Greek and Anatolian languages, as shown e.g. by Cassio, Smyth, and Woodard. An optimal case-study of this linguistic interaction is found in ancient Greek literature, namely in the language of the iambographer Hipponax, which has been deeply analysed from this perspective by scholars such as Degani, Kearns, Masson, Rapallo, Tedeschi, and more recently Dale and Hawkins. However, these works have never adequately addressed the morphological issue of heteroclisis. For this reason, my paper gives attention to the phenomenon of heteroclisis in Hipponax’ fragments as a result of Lydian and Lycian influence on Greek morphology.

Specifically, I will be looking at some cases of heteroclitic endings which appear in Hipponax’ lines. Scholarship on Greek heteroclisis focuses on analogical changes limited to Greek paradigms, as shown by the studies of Coker and Stump. However, the heteroclitic forms exhibited in the fragments of the Ephesian iambographer are not explicable within the Greek morphological system. The majority are nouns which switch from feminine forms in the singular to neuter forms in the plural. In order to explain the exceptionality of this phenomenon, I will employ a twofold approach: 1) analysing the phonological developments in Lydian and Lycian languages through which a three-gender system (masculine, feminine, and neuter) collapses into a two-gender system (animate and inanimate), and the feminine gender fades by merger; 2) investigating the etymology of the heteroclitic words. Respectively, I argue that 1) the resulting uncharacterized ending of feminine nouns blends with the ending of neuter plural nouns, and that 2) most of the examples of heteroclitic plural are Lydian borrowings which entered the Greek lexicon. Furthermore, I will introduce evidence of collective plural in Lycian and pluralia tantum in Lydian in order to corroborate my thesis.

In conclusion, this paper, by closely examining the morphology of both Anatolian languages and Hipponax’ fragments, shows that the heteroclitic forms in Hipponax’ lines have their explanation outside Greek morphology, and sheds new light on the wide-open debate of the influence of the languages of Asia Minor on Greek language and culture.


Beekes, R. S. P. 2003, “Luwians and Lydians,” Kadmos 42: 4-7.

— 2010, Etymological Dictionary of Greek (2 vols.), Leiden - Boston.

Coker, A. 2009, “Analogical change and grammatical gender in ancient Greek,” Journal of Greek Linguistics 9: 34-55

Collins, B. J., Bachvarova, M. R., and Rutherford, I. C. 2008 (eds.), Anatolian Interfaces: Hittite, Greeks and their Neighbours, Oxford.

Degani, E. 1991, Hipponax: Testimonia et Fragmenta, Stutgart - Leipzig.

— 1995/96, “La lingua dei barbari nella letteratura greca arcaica: esotismi Ipponattei,” Studi orientali e linguistici 6: 157-164.

Garçía Ramón, J. L. 1991, “Reconstrucció indoeuropea y anomalies morfologiques en grec antic II,” in Treballs en honor de Virgilio Bejarano, Barcelona: 513-522.

Hawkins, S. 2010, “Greek and the Languages of Asia Minor to the Classical Period,” in E. J. Bakker (ed.), A Companion to the Ancient Greek Language, Chichester - Malden: 213-227.

— 2013, Studies in the Language of Hipponax, Bremen.

Heine, B., and Kuteva T. 2005, Language Contact and Grammatical Change, Cambridge.

Janse, M. 2001, “Morphological Borrowing in Asia Minor,” in Y. Aggouraki, et al. (eds.), Proceedings of the 4th International Conference of Greek linguistics, Thessaloniki: 473-479.

Karatsareas, P. 2009, “The loss of grammatical gender in Cappadocian Greek,” Transactions of the Philological Society 107.2: 196-230

— 2011, “Neuter Heteroclisis in Asia Minor Greek: Origin and Development,” Neoelliniki Dialektologia 6: 111-135.

Kearns, J. M. 1994, “A Greek Genitive from Lydia,” Glotta 72: 5-14.

Melchert, H. C. 1992, “Relative chronology and Anatolian: the vowel system,” in R. Beekes, A. Lubotsky, J. Weitenberg, et al. (eds.), Rekonstruktion und relative Chronologie. Akten der VIII. Fachtagung der Indogermanischen Gesellschaft, Innsbruck: 41–53.

— 1994, “The feminine gender in Anatolian,” in G. Dunkel, G. Meyer, S. Scarlata, et al. (eds.), Früh-, Mittel-, Spätindogermanisch, Wiesbaden: 231–244

— 2004, A Dictionary of the Lycian Language, Ann Arbor.

Myers-Scotton, C. 2002, Contact Linguistics: Bilingual Encounters and Grammatical Outcomes, Oxford.

Rapallo, U. 1976, “Influssi anatolici sulla grammatica di Ipponatte,” Studi italiani di filologia classica 48, Firenze: 200-243.

Rizza, A. 2001, “Ipponatte e i Lidii: note sul framm. 7 Degani,” Syngraphe 3: 7-14.

Stump, G. T. 2006, “Heteroclisis and Paradigm Linkage,” Language 82.2: 279-322.

Tedeschi, G. 1978, “Lingue e culture in contatto: il problema della lingua in Ipponatte,” Incontri Linguistici 4, Pisa: 225-233.

—  1981, “I prestiti linguistici nei frammenti ipponattei e il problema dell’interferenza culturale ad Efeso,” Quaderni di filologia classica 3, Trieste: 35-48.

Watkins, C. 2000, Indo-European and Indo-Europeans. The American Heritage Dictionary of the English Language (4th, revised edition). Boston.

Woodard, R. D. 2008 (ed.), The Ancient Languages of Asia Minor, New York.

Phrasal verbs in a corpus of post-classical Greek letters from Egypt

Lady Margaret Hall, University of Oxford

Phrasal verbs are a cross-linguistic phenomenon. They consist of a simple verb like ‘to go’ and a particle or adverb that contributes to the meaning as ‘up’ in ‘to go up’. Phrasal verbs may be intransitive as ‘to go up’ or transitive like ‘to look something up’. Alternatively, a simple verb can be modified by prefixing a particle. Classical and post-classical Greek, unlike English, prefer this option. For instance, ἀνέρχομαι ‘to go up’ consists of a particle ἀν(α)- and the simple verb ἔρχομαι.

            However, in post-classical Greek letters from Egypt, several phrasal-verb constructions appear. An example is αἴρω ἔξω (P. Lond. 6 1914) for expected ἐξαίρω. Given the linguistic situation in Egypt, these phrasal-verb constructions may have resulted from language contact with Egyptian (Coptic). Coptic, unlike Greek, operated exclusively with phrasal verbs rather than compound verbs. However, the issue is actually more complicated. In the post-classical period, the prefix of many compound verbs seems to have lost its semantic significance. As a result, double compounds such as συνκατέρχομαι ‘to accompany’ were created. In these, only the first prefix (here συν-) contributed to the meaning whereas the second prefix (here κατ-) was semantically bleached. Hence, in order to explain the phrasal-verb structures in post-classical Greek texts, we must consider at least four factors: (1) external factors, i.e. language contact with Egyptian (Coptic), (2) internal diachronic developments (similar to English, cf. Hiltunen 1983), (3) internal synchronic variation, i.e. register-related variation (cf. Biber and Conrad 2009), and (4) the possibility that the phrasal verb and the corresponding compound verb developed a nuanced meaning and were thus not interchangeable (cf. Storrer 2007).

I will seek to shed light on the reasons for the appearance of phrasal-verb patterns in Greek starting from instances in a corpus of post-classical Greek texts from Egypt. The corpus consists of all private letters that belong to bilingual papyrus archives dating from the fourth to mid-seventh centuries. The phrasal-verb patterns that appear in this corpus will then be checked against classical and post-classical literary sources as well as post-classical documentary sources to test whether the use of phrasal verbs was limited to Egypt or represents a more widespread development in later Greek.

Biber, D., and Conrad, S. (2009). Register, genre, style. Cambridge.

Hiltunen, R. (1983). The decline of the prefixes and the beginnings of the English phrasal verb: the evidence from some Old and Early Middle English texts. Turku.

Layton, B. (2011). A Coptic grammar: Sahidic dialect. Porta linguarum orientalium N. S. 20, 3rd edition. Wiesbaden.

Storrer, A. (2007). ‘Corpus based investigations on German support verb constructions’, in Fellbaum, C. (ed.). Idioms and collocations: Corpus-based linguistic and lexicographic studies. Research in corpus and discourse. London, 164–187.

Thim, S. (2012). Phrasal verbs: The English verb-particle construction and its history. Topics in English Linguistics 78. Berlin.

Wild, K. (2011). ‘Phrasal verbs: ‘a process of the common, relatively uneducated, mind’?’, English Today 27/4: 53–57.

Analyse linguistique du "Bronze d'Idalion". Observations phonétiques et morphologiques.

Universidad Nacional de Córdoba - Argentina

               L'unité linguistique du monde hellénique est une abstraction, dans le sens où  'il n'y a pas un seul système qu'on puisse appeler «le grec», mais une série de variantes locales, parfois avec des différences très prononcées. Cette "abstraction" n'est pas, cependant, une création moderne, car les Grecs eux-mêmes ont reconnu cette unité en lui attribuant un nom unique: ἑλληνίζειν (τῇ φωνῇ) est de "parler grec", et les dialectes doivent avoir été mutuellement intelligibles (par opposition à βάρβαρος).

Pour la connaissance de la diversité dialectale, le grec du premier millénaire avant JC c'est dans une situation particulière (pas très commune dans les langues anciennes) pour plusieurs raisons.

Il a un usage très répandu de l'écriture, ce qui permet d'avoir des documents épigraphiques locaux de pratiquement toute son extension géographique. Bien sûr, la répartition est irrégulière, et il y a des écarts très importants surtout lorsqu'on considère le facteur historique: dans de nombreuses régions l'adoption de l'écriture était tardive, à une époque où le dialecte local commençait à laisser sa place au κοινή, de sorte que dans les textes le dialecte local est souvent vu avec des éléments incorporés de ce dernier. En outre, on doit toujours à l'esprit que toute langue écrite a un certain degré de fossilisation par rapport à la langue parlée.

Certains dialectes, ou du moins certaines caractéristiques identifiées avec des groupes dialectaux, se sont cristallisés en tant que langues littéraires de genres spécifiques. Ces "dialectes littéraires" ne sont pas identiques à ceux historiques, et dans certains cas (comme le "dorique" des choeurs tragiques) ils atteignent à peine quelques particularités, mais indiquent une conscience de la différence dialectale, une connaissance des formes spécifiques dans le reste du monde hellénique, et ils ont exercé une influence permanente sur la façon d'étudier et de comprendre la dialectologie grecque.

Le résultat est que nous avons une multitude de particularités locaux attestés dans les documents de la plupart des villes de langue grecque.

Le présent travail vise à mettre en évidence les caractéristiques dialectales qui se distinguent dans la plus ancienne et la plus importante des inscriptions chypriotes, connue sous le nom de "Bronze d'Idalion". Il s'agit d'une plaque opisthographe en bronze, composée de trente et une lignes écrites dans le syllabaire chypriote. Il est daté de 475 av. J.C. et a été trouvé dans un sanctuaire d'Athéna à Idalion, près de Nicosie.

Dans l'analyse de l'inscription nous nous occuperons tout d'abord de l'explication des signes dans lesquels le document est écri (on doit se rappeler que le chypriote était le seul dialecte non écrit en alphabet grec), et nous spécifierons ensuite des différentes caractéristiques, phonétiques et morphologiques, qui caractérisent ce dialecte.


BUCK, C.D. (1955) The Greek Dialects. Chicago: University of Chicago Press.

CHADWICK, J. (1987) El enigma micénico. El desciframiento de la escritura Lineal B (Versión castellana de Enrique Tierno Galván – Título original: The Decipherment of Lineal B, Cambridge 1958) Madrid: Taurus.

COLVIN, S. (2014) A Brief History of Ancient Greek. UK: Blackwell.

COLVIN, S. (2007) A Historical Greek reader: Mycenaean to the Koiné. New York: Oxford University Press.

DUHOUX, Y. (1983) Introduction aux dialectes grecs anciens: problèmes et méthodes, recueil de textes traduits. Louvain/Paris: Peeters.

EGETMEYER, M. (2010) Le dialecte grec ancien de Chypre. Tome II: Répertoire des inscriptions en syllabaire chypro-grec, Berlin: De Gruyter.

HORROCKS, G., Greek. (2010) A History of the Language and its Speakers. Second Edition. Malden & Oxford: Wiley-Blackwell.

JEFFERY L.H. (1961) The local scripts of Archaic Greece. A study of the origin of the Greek alphabet and its development from the eighth to the fifth centuries B.C. Oxford: Oxford University Press.

MILLER, D.G. (2010) Ancient Greek Dialects and Early Authors. Introduction to the Dialect Mixture in Homer, with Notes on Lyric and Herodotus. Boston & Berlin: De Gruyter.

OSBORNE, R. (Ed.) (2000) Classical Greece. New York: Oxford University Press.

PALMER, L.R. (1996) The Greek Language. London: University of Oklahoma Press.

RAMÍREZ TREJO, A.E. (2005) Manual de Dialectología Griega. México: UNAM.

RODRÍGUEZ ADRADOS, F. (1993) La democracia ateniense. Madrid: Alianza Universidad.

VAN EFFENTERRE, H. et RUZÉ, F. (1994) Nomima (vol. I). Roma: École française de Rome.

ὦτα ἀπιστότερα ὀφθαλμῶν?

Intonation Units and discourse coherence in the Gyges-episode (Hdt. 1.8-13) and beyond

Humboldt-Universität zu Berlin

This poster presents first results of my corpus research on Intonation Units (IUs) in Hdt. 1.6-94. It will give an overview over (1) what recurrent IU-types in my corpus are, (2) how they enter into coherence relations with other IUs on a local and more global level of discourse structure, (3) how an IU-based framework for discourse coherence may enhance our understanding of individual passages of Ancient Greek texts as well as more specific linguistic phenomena. – The third point will by illustrated by a case study on the famous Gyges-episode (Hdt. 1.8-13), which has figured prominently in the relevant literature since Dover 1960.


Editions, commentaries, translations

Godley                                           Godley, A. D. (1920–1925): Herodotus. With an English Translation by A. D. Godley. 4 vols. Heinemann, London & Putnam’s Sons, New York.

Hude                                              Hude, C. (1927): Herodoti historiae. 2 vols. 3rd ed. Clarendon, Oxford.

Legrand                                          Legrand, Ph.-E. (1932–1954): Hérodote. Histoires. 11 vols. Belles Lettres, Paris.

Macaulay/Lateiner                      Lateiner, D. (2004): The Histories. Herodotus. With an Introduction and Notes by Donald Lateiner. Translated by G. C. Macaulay and Revised throughout by Donald Lateiner. Barnes & Noble Classics, New York.

Rosén                                             Rosén, H. (1987/1997): Herodotus. Historiae. 2 vols. Teubner, Stuttgart–Leipzig.

Stein                                               Stein, H. (1962[1901]): Herodotos, erklärt von Heinrich Stein. Erster Band: Buch I, mit einer Einleitung über Leben, Werk und Dialekt Herodots und einer Karte. 6th ed. Weidmannsche Verlagsbuchhandlung, Berlin.

van Herwerden                            van Herwerden, H. (1884–1888): ΗΡΟΔΟΤΟΥ ΙΣΤΟΡΙΑΙ. Praesertim in usum scholarum recognovit et brevi annotatione instruxit Henricus van Herwerden. Kemink & fil., Traiecti ad Rhenum[Utrecht].

Wilson                                            Wilson, N. G. (2015): Herodoti historiae. 2 vols. Clarendon, Oxford.

General linguistic literature

Chafe, W. L. (1994): Discourse, consciousness, and time: The flow and displacement of conscious experience in speaking and writing. University of Chicago Press, Chicago.

Dik, S. C. et al. (1981): On the typology of Focus Phenomena. In: Hoekstra, T., van der Hulst, H. & Moortgat, M. (eds.): Perspectives on Functional Grammar. Foris, Dordrecht–Cinnaminson, 41–74.

Dik, S. C. (1997a): The Theory of Functional Grammar: Part I: The Structure of the Clause. 2nd rev. ed. by K. Hengeveld. Mouton–de Gruyter, Berlin–New York.

Dik, S. C. (1997b): The Theory of Functional Grammar: Part II: Complex and Derived Constructions. Ed. by K. Hengeveld. Mouton–de Gruyter, Berlin–New York.

Féry, C. (2013): Focus as Prosodic Alignment. Natural Language & Linguistic Theory 31, 683–734.

Hengeveld, K. & Mackenzie, J. L. (2008): Functional Discourse Grammar: A typologically-based theory of language structure. Oxford University Press, Oxford.

Höhle, T. N. (1992): Über Verum-Fokus im Deutschen. In: Jacobs, J. (ed.): Informationsstruktur und Grammatik. (Linguistische Berichte. Sonderheft 4/1991–92). Westdeutscher Verlag, Opladen, 112–141.

Lambrecht, K. (1994): Information structure and sentence form: Topic, focus and the mental representation of discourse referents. Cambridge University Press, Cambridge.

Matić, D. (2003b): Topics, Presuppositions, and Theticity: An Empirical Study of Verb-Subject Clauses in Albanian, Greek, and Serbo-Croat. Dissertation. Universität zu Köln, Köln.

Literature on word order and discourse structure in Ancient Greek central to the present approach

Allan, R. J. (2006): Herodotus’ historiën als sprekend leesboek: Herodotus tussen oraliteit en geletterdheid. Lampas 39, 19–32.

Allan, R. J. (2014a): Changing the Topic: Topic Position in Ancient Greek Word Order. Mnemosyne 67 (2), 181–213.

Allan, R. J. (2014b): Oral Aspects of Herodotus’ Style: A Linguistic Approach. Paper presented at the conference “Herodotus and the rise of Greek narrative prose”, Cologne, November 28th 2014.

Allan, R. J. (2016): Herodotus’ style als ‘Sprache der Nähe’: Paper presented at the conference “Language in Style”, Oxford, 18th-20th May 2016.

Bakker, E. J. (1997): Poetry in speech: Orality and Homeric discourse. Cornell University Press, Ithaca–London.

Bertrand, N. (2009): Les pronoms postpositifs dans l’ordre des mots en grec: domaines syntaxiques, domaines pragmatiques. Lalies 29, 227–252.

Bertrand, N. (2010): L’ordre des mots chez Homère: structure informationelle, localisation et progression du récit. Thèse de doctorat. Paris-Sorbonne, Paris.

Dik, H. (1995): Word Order in Ancient Greek: A Pragmatic Account of Word Order Variation in Herodotus. Gieben, Amsterdam.

Dover, K. J. (1960): Greek Word Order. Cambridge University Press, Cambridge.

Fraenkel, E. (1933): Kolon und Satz. Beobachtungen zur Gliederung des antiken Satzes II. In: Nachrichten von der Gesellschaft der Wissenschaften zu Göttingen. Philosophisch-historische Klasse 1933 (3), 319–354.

Goldstein, D. M. (2010): Wackernagel’s Law in Fifth Century Greek. PhD Dissertation. University of Berkeley, California.

Goldstein, D. M. (2016): Classical Greek Syntax: Wacker-nagel’s Law in Herodotus. Brill, Leiden–Boston.

Matić, D. (2003a): Topic, focus, and discourse structure: Ancient Greek word order. Studies in Language 27, 573–633.

Scheppers, F. (2011): The colon hypothesis: Word order, discourse segmentation and discourse coherence in Ancient Greek. VUBPRESS, Brussels.

Slings, S. R. (1992): Written and spoken language: an exercise in the pragmatics of the Greek sentence. Classical Philology 87, 95–109.

Slings, S. R. (2002): Oral Strategies in the Language of Herodotus. In: Bakker, E. J., de Jong, I. J. F. & van Wees, H. (eds.): Brill’s Companion to Herodotus. Brill, Leiden, 53–77.

Additional literature consulted for the discussion of the Gyges-episode

Aerts, W. J. (1965): Periphrastica: An Investigation into the Use of εἶναι and ἔχειν as Auxiliaries or Pseudo-Auxiliaries in Greek from Homer up to the Present Day. Hakkert, Amsterdam.

Bentein, K. (2013): Prog Imperfective Drift in Ancient Greek? Reconsidering EIMI ‘Be’ with Present Participle. Transactions of the Philological Society 111 (1), 67–107.

Bonifazi, A., Drummen, A. & Kreij, M. de (2016): Particles in Ancient Greek Discourse: Five Volumes Exploring Particle Use Across Genres. Center for Hellenic Studies, Washington, DC. (last access: 2018-01-29).

Fraser, B. (2001[2003]): The clause start in Ancient Greek: focus and the second position. Glotta 77, 138–177.

Krisch, T. (1990): Das Wackernagelsche Gesetz aus heutiger Sicht. In: Eichner, H. & Rix, H. (eds.): Sprachwissenschaft und Philologie: Jacob Wackernagel und die Indogermanistik heute. Kolloquium der Indogermanischen Gesellschaft vom 13. bis 15. Oktober 1988 in Basel. Reichert, Wiesbaden, 64–81.

Luraghi, S. (1990): Osservazioni sulla Legge die Wackernagel e la posizione del verbo nelle lingue indoeuropee. In: Conte, M.-E., Giacalone Ramat, A. & Ramat, P. (eds.): Dimensioni della linguistica. Angeli, Milano, 31–60.

Luraghi, S. (2013): Clitics. In: Luraghi, S. & Parodi, C. (eds.): Bloomsbury Companion to Syntax. Continuum, London, 165–193.

Marshall, M. H. B. (1987): Verbs, Nouns, And Postpositives in Attic Prose. Scottish Academic Press, Edinburgh.

Ruijgh, C. J. (1990): La place des enclitiques dans l’ordre des mots chez Homère d’après la loi de Wackernagel. In: Eichner, H. & Rix, H. (eds.): Sprachwissenschaft und Philologie: Jacob Wackernagel und die Indogermanistik heute. Kolloquium der Indogermanischen Gesellschaft vom 13. bis 15. Oktober 1988 in Basel. Reichert, Wiesbaden, 213–


Evidence for Linguistic Atticism in the Papyrus Fragments of Achilles’ Tatius

University of KwaZulu-Natal

Achilles Tatius’ novel, Leucippe and Clitophon, (2nd century AD) is the best preserved of the five fully extant Ancient Greek novels. His work survives in the form of 12 complete and 13 partial medieval manuscripts dating from the 12th to 16th centuries (Vilborg 1955: xviii). In addition, 7 papyrus fragments containing portions of the text have been identified, the longest of which covers the last 9 chapters of book 3, a total of 218 lines (Willis 1990: 74-76). These fragments date from the 2nd to 4th centuries AD and are therefore an invaluable source of evidence for the accuracy of the later manuscripts. The information they present is especially important in any linguistic analysis of his work, as it can reveal if and how the manuscript scribes may have edited his language based on their own practices and personal bias.

Achilles Tatius’ novel has previously been described as using “Atticist” language (Sexauer 1899: 76-77; Vilborg 1962: 13-16; Silk 2009: 22). The practice of Atticism, in which Hellenistic and Roman-era authors made use of older Attic dialectal features, is well attested, although the examination of its practice in individual authors is rarer (Kim 2014; Horrocks 2014: 173). In this paper, presenting some of the findings of my doctoral thesis, I will consider what the papyrus fragments of Achilles’ text reveal about his use of Atticist grammatical forms. My thesis examined examples of phonological, morphological and lexical Atticism in the whole of Achilles’ text based primarily on Ebbe Vilborg’s (1955) edition (which was derived from the manuscripts and papyrus fragments known to him at the time).

In this paper, I will take those of the Atticist tokens which I have identified that occur in the papyrus fragments and compare them with their corresponding representations in the manuscripts. It will be seen that many, although not all, of the Atticist forms found in the manuscripts are already present in the papyrus fragments and that there is a great deal of consistency between the earlier and later sources with respect to these examples. Similarly, Koine forms that could have been, but are not Atticised in the manuscripts, are normally not Atticised in the papyrus fragments either. This suggests little evidence of scribal error or bias having corrupted the manuscripts.  Significantly, there already seems to be evidence of the inconsistent application of Atticist forms which I identified in Achilles’ text as a whole.

This paper will show that the papyrus fragments of Achilles’ text, though covering a small portion of the total work, provide some evidence for the reliability of the later manuscripts and the confidence with which they can be used to assess Achilles’ language.


HORROCKS, G. C. (2014). Greek: A History of the Language and its Speakers (Second edition; Paperback ed.). Chichester: John Wiley and Sons Ltd.

KIM, L. (2014). The Literary Heritage as Language: Atticism and the Second Sophistic. In E. Bakker (Ed.), A Companion to the Ancient Greek Language (Paperback ed., pp. 468-482). Oxford: Wiley Blackwell.

SEXAUER, H. (1899). Der Sprachgebrauch des Romanschriftstellers Achilles Tatius. Heidelberg: Karlsruhe.

SILK, M. (2009). The Invention of Greek: Macedonians, Poets and Others. In A. Georgakopoulou & M. Silk (Eds.), Standard Languages and Language Standards: Greek, Past and Present, (pp. 3-32). Farnham: Ashgate.

VILBORG, E. (1955). Achilles Tatius: Leucippe and Clitophon (Edition). Stockholm: Almqvist & Wiksell.

VILBORG, E. (1962). Achilles Tatius: Leucippe and Clitophon (Commentary). Stockholm: Almqvist & Wiksell.

WILLIS, W. H. (1990). The Robinson-Cologne Papyrus of Achilles Tatius. Greek, Roman, and Byzantine Studies, 31(1), 73-102.

Forbidden language and the rise of linguistic taboos in Ancient Greek

Aristotle University of Thessaloniki

Linguistic taboos are part of the metaphorical usage of language and constitute a mechanism whereby the speakers avoid certain linguistic features and references out of fear lest their usage may cause some harm to the user. Such language use is driven by social, religious and cultural beliefs and attitudes which prescribe the avoidance of direct reference to certain concepts and the employment of indirect ways for their expression. Taboos may also be determined by pragmatic and situational concerns, e.g. age, gender and social position of the collocutors or the specific setting of the utterance. A taboo may be a single word, a concept, a name, an animal, a dietary item, an object or even a thought whose mention is, in the speaker’s cultural and/or personal milieu, believed to cause harm or even death and destruction.

Despite the occasional studies on individual semantic areas (e.g. Parker 1983, 328-31 for the general term for taboo in Greek; Havers 1946, Meillet 1958, Bonfante 1939, Watkins 1975 for Indo-European, including Greek), a systematic and full-fledged study of the phenomenon of taboo in Ancient Greek is still a desideratum. The present paper will deal with two such cases of taboos, namely (i) the cardinal points and points of orientation and their various significations and/or symbolic references (e.g. Gk. ἀνατολαί vs. δυσμαί, δεξιός or δεξιτερός vs. σκαιός and λαι(ϝ)ός (→ ἀριστερός/εὐώνυμος), Lat. scaeuus, laeuus, etc., or Hes. Op. 727 μηδ’ ἄντ’ ἠελίου τετραμμένος ὀρθὸς ὀμιχεῖν ‘and do not urinate standing facing the sun’, etc.), and (ii) sacred language and the various ramifications of its usage (e.g. Gk. οὐ θέμις, Hitt. natta ara ‘(it is) not right’, as in Hittite: KUB XXX 10 Ro. 13, ši-ú-ni-mi-ma-mu ku-it šu-up-pí a-da-an-na na-at-ta a-ra na-at Ú-UL ku-uš-ša-an-ka e-du-un ‘but what has been offered (šuppi) to my god <and> I am not allowed (natta ara) to eat, I have never eaten’; cf. Gk. ὅσιος as in ὁσίη κρεάων ‘sanctified portions of meat; rite of flesh-offering’, Umbrian supa/sopa ‘id.’, etc.). A sub-category of the latter group of taboos is the language of curse. The focus will be Ancient Greek, but within the comparative framework of cognate Indo-European linguistic and cultural traditions, with material drawn mainly from Greek, Italic, Sanskrit and Hittite.


Bonfante, G. 1939. “Études sur les tabou dans les langues indo-européennes”,  Mélanges de linguistique offerts à Charles Bally, Geneva, 195-207.

Cohen, Y. 2002. Taboos and the prohibitions in Hittite society. A study of the Hittite expression natta āra (‘not permitted’), Heidelberg.

Giannakis, G. K. (in preparation). Indo-European taboos.

Havers, W. 1946. Neuere Literatur zum Sprachtabu, Vienna.

Meillet, A. 1958 [1906]. “Quelques hypothèses sur des interdictions de vocabulaire dans les langues indo-européennes”, Linguistique historique et linguistique générale, Paris, 280-91. 

Parker, R. 1983. Miasma. Pollution and purification in early Greek religion, Oxford.

Watkins, C. 1975. “La désignation indo-européenne du ‘tabou’”, Langues, discours, société. Pour Émile Benveniste, Paris, 208-14. 

Differential agent marking in classical Greek

University of California, Los Angeles

Agents of passive predicates in classical Greek are realized either in the dative case or with a prepositional phrase. Standard doctrine holds that the realization of the agent phrase is conditioned by grammatical aspect (Kühner and Gerth 1898: 422, Schwyzer 1988: 150.2, Smyth 1956: §§1488, 1490, George 2005: 1, 78). Dative agents are licensed by perfect passive predicates, while prepositional phrase agents occur elsewhere:

(1) i. Dative agent with perfect

                       ἡ δὲ ὁδὸς ἡ ἡμερησίη ἀνὰ διηκόσια στάδια συμβέβληταί⸗μοι.

                        ‘A day’s journey has been calculated by me at 200 stades.’                Hdt. 4.101.3

                    ii. PP-agent with non-perfect

                        γνώμηι μέντοι ἑσσοῦται ὑπὸ σεῦ.

‘In counsel, however, he has been bested by you.’                                Hdt. 7.237.1

We have an example of a dative agent with perfect predicate in example (1i). In (1ii), the predicate is a present imperfective and the agent is realized with a prepositional phrase, as expected.

Counterexamples to this generalization run in both directions, however:

            (2) i. Dative of agent with non-perfect (present)

τότε δὲ ὡς τῷ Κλεομένεϊ ὡδώθη τὸ ἐς τὸν Δημάρητον πρῆγμα, αὐτίκα     

παραλαβὼν Λευτυχίδεα ἤιε ἐπὶ τοὺς Αἰγινήτας.

‘Then, once the affair with Demaratos was successfully concluded by Kleomenes, he along with Leutykhides straightaway advanced the Aigenetans.’                                                                                           Hdt. 6.73.1

                   ii. PP-agent with perfect

                       ἐξεληλαμένος τε ὑπὸ τοῦ πατρὸς...

‘And having been banished by my father...’                                            Hdt. 1.35.3

In (2i), the dative phrase τῷ Κλεομένεϊ occurs with a past perfective (i.e., aorist) verb. In (2ii), the perfect participle ἐξεληλαμένος occurs with a prepositional phrase agent ὑπὸ τοῦ πατρὸς.

Although the literature is unanimous in the view that aspect plays a fundamental role in the realization of passive agents, the role of other factors has received only minimal investigation. George (2005), for instance, argues that agent marking is governed by the animacy hierarchy and the morphology of the passive predicate.

By contrast, I argue that the following prominence scales play a crucial role in the realization of agent phrases:

(3) Prominence scales

       i.  Aspect

           Perfect » perfective » imperfective

       ii. Definiteness

           Pronoun » proper name » definite NP » indefinite specific NP » indefinite non-specific NP

       iii. Animacy

            Human » animate » inanimate

The following three properties strongly favor dative agents: perfect passive predicate; enclitic pronominal agent; and inanimate subject. By contrast, prepositional phrase agents preponderate among non-perfect predicates; non-pronominal agents; and sentient (i.e., human and animate) subjects. 

I motivate this distribution of agent markers on the basis of insights from the study of differential object marking (DOM). A prominent analysis of DOM is that objects that are prominent on scales such as those in example (3) are more likely to be subject to differential object marking (see, e.g., Aissen 2003). Applying this insight to differential agent marking, we expect agents to be high on the scales in example (3) and subject patients to be lower. In such contexts, passive agents are predominantly marked with the dative. When this association between semantic role and referential prominence is violated, however, prepositional phrase agents are preferred (cf. Haspelmath 2018).

To evaluate the role of the factors in (3), I fit a logistic regression model to a sample of 500 examples of passive predicates with overt agents from Herodotus’ Histories. Regression modeling reveals that the model proposed here is empirically more adequate than the traditional one.


Aissen, Judith (2003). “Different object marking. Iconicity vs. economy.” Natural Language & Linguistic Theory 21: 435–483.

George, Coulter H. (2005). Expressions of agency in ancient Greek. Cambridge: Cambridge University Press.

Haspelmath, Martin (Feb. 2018). “Role-reference associations and the explanation of argument coding splits.” MS.,

Kühner, Raphael and Bernhard Gerth (1898). Ausführliche Grammatik der griechischen Sprache. Satzlehre. 3rd ed. Vol. 1. Hanover: Hahn.

Schwyzer, Eduard (1988). Griechische Grammatik. Syntax und syntaktische Stilistik. 5th ed. Vol. 2. München: C. H. Beck.

Smyth, Herbert Weir (1956). Greek grammar. 2nd ed. Cambridge, MA: Harvard University Press.

Langue grecque et réalités romaines : l’exemple de la transposition de quaestor et de ses dérivés

Université de Liège

Les travaux de M. Mentz, D. Magie et H. J. Mason ont imposé une typologie tripartite dans le domaine du lexique institutionnel gréco-latin (1). Face à une réalité romaine, les hellénophones auraient pu librement choisir de l’exprimer par l’équivalence (comparatio), la traduction (interpretatio) ou la transcription (transcriptio). Dans ce cadre, l’exemple devenu canonique était quaestor, qui aurait pu indifféremment être transposé par ταμίας, ζητητής ou κουαίστωρ. Le but de cette intervention est de questionner le modèle général proposé par les trois lexicographes en approfondissant l’étude de la transposition en grec de quaestor et de ses dérivés. Pour ce faire, il convient de replacer chaque texte et chaque auteur dans le contexte qui était le leur : les témoignages épigraphiques grecs de la questure s’étendent sur 500 ans ; plus de sept siècles séparent Polybe et Jean le Lydien. La dimension diachronique du problème se double également d’une réflexion sur les sources et leur nature : quelle influence les pratiques de chancellerie romaines, telles qu’elles se sont développées à l’époque républicaine, ont-elles eue sur la langue des inscriptions grecques mentionnant la questure ? Pour quelles raisons les différents auteurs hellénophones ayant évoqué la magistrature ont-ils mis en œuvre tel ou tel procédé de transposition ? De la sorte, nous espérons pouvoir illustrer l’influence qu’a pu avoir la langue latine à l’occasion du processus d’appréhension des realia romains par le monde grec. Et dans ce cadre, de mettre également en lumière les variations linguistiques propres à chaque auteur concerné par l’étude.

(1) Mentz, De magistratuum Romanorum Graecis appellationibus, Iéna, 1894 ; Magie, De Romanorum iuris publici sacrique uocabulis sollemnibus in Graecum sermone conuersis, Leipzig, 1905 ; Mason, Greek Terms for Roman Institutions. A Lexicon and Analysis, Toronto, 1974.

Greek loanwords in post-Biblical Hebrew/Aramaic:

some case studies from the midrash Genesis Rabbah

University of Salzburg Department of Linguistics & Austrian Academy of Sciences, ACDH

Greek loanwords, which total over two thousand items stemming from various dialects, make up the largest group of non-native words in the totality of the Hebrew/Aramaic lexicon (in Mishnaic Hebrew, Jewish Palestinian Aramaic and Jewish Babylonian Aramaic). The transliterations of Greek words in the Rabbinic literature (mishnah, targums, talmuds, midrash) contained in the monumental work by Krauss (Lehnwörter 1898-1899), in Lieberman’s and Sperber’s important contributions (Sperber 1984, 1986, 2012), in the more recent Aramaic dictionaries by Sokoloff (2002a, 2002b, 2003, 2014) and the specialized work on Greek loanwords by Shoval-Dudai (2015) do not only constitute evidence for cultural and linguistic contact between Hebrew/Aramaic and Greek, but also make up most important secondary evidence of Hellenistic Greek.

The present study concerns the Greek loanwords in Genesis Rabbah (GenR), a text from Judaism’s classical period, which is the earliest rabbinic commentary on the Book of Genesis and was compiled during the 5th c. CE in Roman Palestine. Except for the last parashiot (chapters), it is structured as an (almost) continuous commentary. GenR is an exegetical midrash; it contains short explanations of words and sentences, often in Aramaic, but also highly difficile and subtle aggadic (narrative) explanations and interpretations of the Biblical text. GenR displays the rabbinic-Jewish worldview and maps out numerous rabbinic theological concepts, e.g. creation, Torah, Israel, the nations, Messiah. GenR also marks an important starting point in terms of its historical relationship with its Roman imperial context. More than other early rabbinic genres, GenR is characterized by its frequent use of Greek loanwords (about 400 types in total) and of concepts and metaphors from the Graeco-Roman culture. Moreover, GenR is the first work of rabbinic midrash that post-dates the Christianization of the Roman Empire.

The paper will present some case studies of Greek loanwords in Genesis Rabbah, the study of which still remains a desideratum, especially due to the interdisciplinary character of the research (cf. Sperber 2012: 55). In most of the relevant glossaries and the scholarly literature, the corresponding Greek form of a Hebrew/Aramaic lexeme is just cited from the major Greek dictionaries (LSJ, Lampe), without further consideration of Hellenistic, Early Byzantine and Modern Greek sources, and recent publications in Greek linguistics. The investigation will focus on hapax legomena and problematic cases of alleged Greek loanwords, which examine each attestation of the respective lexeme in its context and offer an up-to-date linguistic analysis, concerning the origin, the morphophonology as well as detailed fine grained semantics. Furthermore, the investigation will pursue –where possible – comparisons with the Greek loaned vocabulary in Syriac Aramaic and Coptic sources, in order to present the findings in their Eastern Mediterranean context.

Selected bibliography:

Bubenik Vit 2010: “Hellenistic Koine in Contact with Latin and Semitic Languages during the Roman Period”, In: Studies in Greek Linguistics 30, 32-54.

Bubenik, Vit 1989, Hellenistic and Roman Greece as a sociolinguistic area, Amsterdam-Philadelphia.

Butts, Aaron Michael 2016, Language Change in the Wake of the Empire. Syriac in Its Greco-Roman Context, Winona Lake, Indiana.

Clackson, James 2015. Language and Society in the Greek and Roman Worlds, Cambridge.

Dalman, Gustav 1905 [1989]. Grammatik des jüdisch-palästinischen Aramäisch. Darmstadt: Wissenschaftliche Buchgesellschaft.

Gribetz, S.K, D.M & Grossberg, M. Himmelfarb, P. Schäfer (eds). 2016. Genesis Rabbah in Text and Context. Tübingen: Mohr Siebeck.

Grossman, Eitan / Haspelmath, Martin / T. Sebastian Richter (eds.) 2015. Egyptian-Coptic linguistics in typological perspective. Typological Studies in Language. Berlin etc.

Heijmans, Shai. 2013. Greek and Latin loanwords in Mishnaic Hebrew. Lexicon and phonology [in Heb.]. Tel Aviv University PhD thesis.

Katsikadeli C. 2017. „Inwieweit ist Wortbildung entlehnbar? Bemerkungen zu den nachklassischen griechischen Komposita, in: Gieseke-Golembowski, F. / Schaaff, K. (Hrsgg.): Wörter bilden – Akten des Jenaer Mai-Kolloquiums 2015 im Gedenken an Bernd Barschel. (SHVS 9) Hamburg: Baar Verlag, 55-73.

Katsikadeli C. & I. Fykias 2017. “Etymology and Language Contact: Remarks on Greek Loanwords in Late Antiquity.”, in: Ch. Tzitzilis & G. Papanastasiou (eds.): Greek Etymology, Thessaloniki: Institute of Modern Greek Studies (Manolis Triantafillidis Foundation), 495-512 (Greek Language: Synchrony and Diachrony, vol. 1)

Krauss, Samuel. 1898. Griechische und lateinische Lehnwörter im Talmud, Midrasch und Targum, vol. 1: Grammatik. Berlin: Calvary.

Kriaras, E. 2001, 2003, Dictionary of Medieval Vulgar Greek Literature (1100-1669), from A - παραθήκη. Thessaloniki: KEG

Krivoruchko, Julia. 2012. “Greek loanwords in rabbinic literature. Reflections on current research methodology”. In: Greek scripture and the rabbis, ed. by Timothy Michael Law and Alison Salvesen, Leuven – Paris – Walpole, MA, 193-216.

Lampe, G.W.H. 1969. A Patristic Greek Lexicon. Oxford. Clarendon Press.

Lieberman, S. 1965. Greek in Jewish Palestine. New York: The Jewish Theological Seminary of America.

LSJ = Liddell, H. and R. Scott 1996. A Greek-English Lexicon, rev. H. Stuart Jones and R. Mc Kenzie. Oxford: Clarendon Press.

Shoval-Dudai, Nurit. 2015. The integration of Greek and Latin loanwords in post-biblical Hebrew, in: Scripta Classica Israelica, vol. 34, 215–225.

Shoval-Dudai, Nurit (fc.) A Glossary of Greek and Latin Loanwords in Classical Hebrew (in Hebrew). Academy of the Hebrew language.

Sokoloff, Michael 22002a. A Dictionary of Jewish Palestinian Aramaic of the Byzantine Period. Ramat Gan.

Sokoloff, Michael 2002b, A Dictionary of Jewish Babylonian Aramaic of the Talmudic and the Geonic periods. Ramat Gan.

Sokoloff, Michael 22012. A Syriac Lexicon: A translation of the Latin, Correction, Expansion and Update of C. Brockelmann’s Lexicon Syriacum, 2nd. Corrected printing, Winona Lake, Indiana.

Sokoloff 2014, A Dictionary of Christian Palestinian Aramaic, Louvain.

Sperber, Daniel 1984, A Dictionary of Greek and Latin Legal terms in Rabbinic Literature, Jerusalem.

Sperber, Daniel 2012, Greek in Talmudic Palestine, Ramat Gan.

Trapp, Erich et al. 2001-2017. Lexikon zur Byzantinischen Gräzität besonders des 9.-12. Jahrhunderts. Wien: Verlag der ÖAW.

Explorations in automatically parsing Ancient Greek papyrological and literary texts

KU Leuven

Due to the free constituent order and the highly inflected nature of Ancient Greek, current progress in automatically analyzing Ancient Greeks texts has been limited so far (see Lee et al. 2011, Mambrini and Passarotti 2012 for previous approaches), despite the presence of three extensive dependency treebanks (the Ancient Greek Dependency Treebanks (AGDT) of Perseus, the PROIEL treebanks and the Sematia project; see Robie 2017 for a short presentation and further references). This paper will discuss the results of our endeavors to make further progress in this respect. In order to come to better results, we have proceeded in a five-fold way.

  • While the AGDT and Sematia treebanks on the one hand and the PROIEL treebanks on the other hand make use of deviating annotation styles, we automatically converted them to a common format in order that both could be used as training data. We also examined which annotation style is easier to learn for an automatic parsing system.
  • These treebanks, in particular the AGDT project, result from the voluntary work of many individuals, which inevitably lead to a number of incoherent annotations. In order to come to more coherent training data, a number of adjustments are proposed. The paper will present both a set of structural and punctual corrections, which might be considered to be adopted in a new release of the treebanks, and a set of ‘temporary’ manipulations, i.e. changes in the syntactic structure that would make it easier for an automatic system to learn and parse Ancient Greek dependencies.
  • In addition, we have attempted to enlarge the training data by adding a set of manually annotated texts (both papyri and literary texts).
  • While previous approaches use ‘gold’ (viz. manually annotated) morphology and lemma information for their test data, we examined whether similar results were attainable when this information is automatically determined. We also tested and compared several syntactic parsing methods.
  • We manually annotated a small test corpus, consisting of both non-literary (documentary papyri) and literary texts, and compared syntactic parsing accuracy for the different types of texts.

In sum, this paper will discuss to what extent the modifications and increase of the training data impact on the quality of the results of the test data; to what extent the annotation format influences parsing accuracy; which parsing methods turn out to be most suitable to analyze Ancient Greek; and what types of texts yield the best results.


Lee, John, Jason Naradowsky, and David A. Smith. 2011. “A Discriminative Model for Joint Morphological Disambiguation and Dependency Parsing.” In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies-Volume 1, 885–894. Association for Computational Linguistics.

Mambrini, Francesco, and Marco Carlo Passarotti. 2012. “Will a Parser Overtake Achilles? First Experiments on Parsing the Ancient Greek Dependency Treebank.” In Eleventh International Workshop on Treebanks and Linguistic Theories, 133–144. Edições Colibri.

Robie, Jonathan. 2017. “Nine Kinds of Ancient Greek Treebanks.” Open Data for Digital Biblical Humanities (blog),

Syllabification and Sound Change in Greek Stop+Nasal Clusters


Paul Kiparsky

Stanford University


Greek eliminates sequences of oral stops + m differently at each place of articulation. Labials are obligatorily nasalized (1a), and velars and dentals are respectively nasalized and spirantized regularly in morphological paradigms, and sporadically elsewere (1b)-(1c).

  1. a.  1. P+m mm: γεγραμμένος, γέγραμμαι

             2. Pm  mm: γράμμα, ὄμμα


b.   1. K+m Nm: (δε)δεγμένος, πεπλεγμένος, κεκήρυγμαι, κήρυγμα, δόγμα, σμῆγμα

       2. Km = Km: κέκμηκα, ἀκμή, ἄκμων, τέκμαρ, τέκμωρ, δοχμή, δόχμη, αὐχμός

       3. Doublets: δράγμα ‘handful, sheaf’, δραχμή ‘drachm(a)’


c.    1. T+m sm: πέπεισμαι, πεπεισμένος, ἔψευσμαι, ἐψευσμένος, ἤνυσμαι

       2.  = Tm: ἄδμητος, κέδματα, πότμος, ἐφετμή, ἀτμός, πορθμός, ἀριθμός, σταθμός

       3.  Doublets: Κάδμος, inscr. Κάσσμος; ὀδμή (Hom., Hdt., Pind.), ὀσμή (Attic) ‘odor’

I argue that these are instances of a single phonological change of coda weakening. They are not morphologically conditioned, and (1c) is not analogical, as usually thought. Rather, the syllabifica-tion of muta cum liquida clusters is variably sensitive to morphology, and coda weakening is cate-gorically conditioned by syllable structure. Thus tautomorphemic VKmV escapes coda weakening because it is syllabified as V.KmV (e.g. ké.kmē.ka), while heteromorphemic VK-mV is syllabified as VK.mV and undergoes coda weakening: /pe-plek-ménos/ > pe.pleŋ.ménos.

Metrics and morphology support this syllabification. In Attic poetry, stop+liquid and stop+nasal clusters, except -gn-, -gl-, -bl-, but sometimes including mn-, do not make position. It reveals that morphological boundaries induce onset syllabification in proportion to their strength: the sequence V##kmV > V#kmV > Vk+mV shows increasing frequency of onset syllabification.

Morphology provides convergent evidence (Steriade 1982). Verbs beginning with single con-sonants and stop+sonorant clusters (except gn-, gl-, bl-, but sometimes including mn-) form their perfect stem by reduplication: gègrafa, kèkmhka. Verbs beginning with other clusters form per-fects with e-: êyeus ai, êgnwka. Reduplicating them would have to violate either the CV- template (*pep.seud-), or the syllable structure constraints. The temporal augment avoids both violations.

The reason labials assimilate obligatorily is that *pm-, *phm-, *bm- are not possible onsets. Therefore the labial in VPmV cannot be syllabified as an onset (as V.PmV). It must be syllabified as a coda (VP.mV) regardless of morpheme boundaries. Being always a coda, it always assimilates.

On the pattern of the nasal assimilation in /ge-graph-ménos/ > gegrammènoc, /pe-plek-ménos/ > peplegmènos [-ŋ-], we might have expected /pe-peith-ménos/ > *pepeinménos. But *nm is an impermissible cluster in Greek. The closest phonotactically acceptable outcome is pepeisménos, with the apparently unnatural replacement of n by s. I reject the traditional explanation that invokes paradigmatic analogy on the pattern 2.Sg. /pé-peith-sai/ pèpeissai → pèpeisai, 3.Sg. /pé-peith-tai/ → pèpeistai, 1. Sg. /pé-peith-mai/ > pèpeismai, because there is no analogical basis for n → s in the perfect middle, nor outside paradigms, as in /phan-ma/ φάσμα ‘specter’, /khan-ma/ χάσμα ‘chasm’, /khan-mē/ χάσμη ‘yawning’. Κάσσμος (inscr.), ῥυσμός (Ionic), Ἄσμητος (inscr.), Ἀγαμέσμων (Att.inscr.)

Given the rest of Greek phonotactics, /T+m/ sm is in fact the optimal repair of the phonotactic violation created at the morpheme boundary. In a robust crosslinguistically widespread pattern that has been explained as avoidance of gestural overlap, Greek allows only ŋn, ŋm, mn, and not *nŋ, *mŋ, *nm. Since the closest replacement *nm is not an acceptable consonant cluster in Greek, it is blocked, and spirantization emerges the next best solution. I demonstrate this by a formal Optimality-theoretic analysis.

This case illustrates how a single change can have multiple and indirect manifestations through its interaction with the rest of the constraints of the phonological system. The A B / C___D schema is a not an appropriate representation of sound change in such cases. We see again that changes don’t wear their identities on their sleeves. When put under the microscope, what looked like analogical changes may turn out to be sound changes, and vice versa.

Infinitives and imperatives in Greek votive and prescriptive texts

Institute of linguistics, Department of comparative linguistics

0. Infinitive is a verbal noun meaning the action as an abstract item: a process without any nominative subject. It is derived from several fixed case forms of verbal nous with or without noun suffixes. The most archaic and vary system containing 17 arts is attested in Rig Veda. Dorothy Disterheft (1997) meant that this category is here not formed. But it is not right: several infinitives exist as specific verbal nouns. Most infinitives have suffixes originated from flexion of Locative, Dative, Genitive or Accusative. The locative infinitive describes the existing level:

(1) yásya trásanti śávasā saṃcákṣi śátravo bhiyā́ (RV VI 14, 4)

‘Enemies are trembling from fear at the sight of whose power’

Controversially, the Dative infinitive means the virtual, possible or owing event:

(2) r̥bhukṣánam ná vártave (VII 45, 29)

‘One cannot to win a wise man’

The Locative and Dative infinitives can occur in the same pada and form opposition of real and possible event:

(3) imá ā́ yātam índavaḥ sómāso…

utá vām uṣáso budhí sākám sū́ryasya raśmíbhiḥ

sutá mitrā́ya váruṇāya pītáye (I 137, 2)

‘Let come drops of Soma, for you both in time of wakening of Dawn, with the rays of sun, pressed (Soma) to Mitra and Varuṇa for drinking’ Cf. (Sgall 1959)

1. The system of Greek infinitives is more compact. It contains two suffixes –en and –men. The first one is connected with the root in weak grade; by athematic verb the suffiх –ai is added. This suffix is facultative by -men. Some researchers compare it with Vedic Dative flexion: doῦnai, Cypr. doϝenai, Homer. dόmenai = dā́vane/ dā́mane (Brugmann 1893). According Delbrück (1892) this affix is an old Locative of –ā-stems. Benveniste (1935) and Aalto (1953) meant, that –ai was originally an independent particle with hortative meaning. But one can suggest that it is a Dative of laryngeal stems (less probably ― Locative). Its presence corresponds with the character of Greek infinitive, what has ever a modal meaning: debt, possibility, order.

2. The meaning of infinitive in Greek is similar with imperative. It can depend on indicative verb:

(4) δῶρον καὶ ἐγώ, τέκνον φίλε, τοῦτο δίδωμι…

σῆ  ἀλόχῳ φορέειν… (Od. XIV: 125-127)

‘I give a this present, dear child (in order) to bring it to your wife’

Infinitive with the imperative meaning can be also independent:

(5) τῷ νῦν μηποτε και σὺ γυναικί περ ἤπιος εἶναι

μή οἱ μῦθον ἅπαντα πιφαύσκεμεν ὅν κ᾽ εὒ εἰδῃς

ἀλλὰ τὸ μὲν φάσθαι, τὸ δὲ κεκρυμμένον εἶναι (Od. XI: 441-443)

‘Don’t be every blunt with your wife; don’t tell her anything, what you know well. But one must be said, other concealed’.

Infinitive and imperative can be similar members of sentence:

(6) εἰ μέν κεν ἐμὲ κεῖνος ἕλῃ ταναήκει χάλκῳ

Τεύχεα συλήσας φερέτω κοιλὰς ἐπὶ νῆμας

σῶμα δ᾽ οἴκαδ ἐμὸν δόμεναι… (Il. VII: 77-79)

‘If he’ll kill me with long-spiked copper weapon, let him, taking my armor bring it to the ships; but he must return my body home’.

The infinitive, however, denotes in this parallel utterances a debt, and imperative ― permission.

Another context shows the same difference now more evident:

(7) εἰ μέν κεν Μενέλαον Ἀλέξανδρος καταπέφνῃ

αὐτὸς ἐπεῖθ  Ἑληνην ἐχέτω καὶ κτήματα πάντα

ἡμεῖς δ᾽ ἐν νήεσσι νεώμεθα ποντοπροῖσιν

εἰ δὲ  Ἀλέξανδρον κτείνῃ ξανθὸς Μενέλαος,

Τρῶας ἐπεῖθ᾽ Ἑλήνην καὶ κτήματα πάντ᾽ ἀποδοῦναι (Il. III: 281-285)

‘If Alexander (Paris) will win and kill Menelaus, let him possess Helena and all the property; if Menelaus will kill Alexander, ― Helena and all the property must be returned’.

3. The structure of conditional sentences in several Greek dialectal votive and law inscriptions is similar: protasis has the predicate in subjunctive, apodosis in imperative or infinitive. In some texts both form are attested, e.g. in Gortyn Laws:

(8) ὅς κ᾽ ἐλευθέροι ε δολοι μέλλει ἀνπιμολεῖν, πρὸ δίκας με ἄγεν. αἰ δε κ᾽ ἄγει, καταδικακσάτο το ἐλευθέρο δέκα στατερανς το δολο πέντε, ὅτι αγει, καὶ δικακσάτο λαγἀσαι έν τρισὶ ἀμεραις. αἰ [δέ] κα με λαγάσει καταδικάδδετο το μεν ἐλευθέρο στατερα, το δολο δάρκναν τας αμερας ϝεκαστας, πρίν κα λαγάσει (I 3-10)

‘Whoever is about to bring suit in relation to a free man or a slave, shall not make seizure before the trial. If he make the seizure, (the judge) shall condemn him to a fine of ten staters in the case of a free man, five in case of a slave, because he seizes him, and shall decree that he releases him in three days. But if he does not release him, (the judge) shall condemn him to a fine of a stater in the case of free man, a drachma in case of a slave, for each day until he releases him’.

Some researchers mean that the infinitive pro imperativo and imperative itself have the same meaning. Garcia Ramos (2001) considers infinitive as an unmarked form, and imperative as a marked one. The first mode expresses impersonal prescription, the second one personal. I consider infinitive as a marker of general prescription, and imperative as a marker of special prescriptions, following the general one.

I.e. infinitive in Greek can denote most strong order. This function can be presented also in other languages (Russian Molčat’ ‘Be silent!’ vs. Molčite! ‘Silent!’)


Aalto 1953 ― Aalto Pentti. Studien zur Geschichte des Infinitives im Griechischen. Helsinki: Acta Antiqua Academiae scientiarum Fennicae, 1953

Benveniste 1935 ― Benveniste Emil. Origines de la formation des noms en indo-européen. Paris : Maisonneuve, 1935

Brugmann 1893 ― Brugmann Karl. Grundriss der vergleichenden Grammatik der indogermanischen Sprachen. Strasbourg, 1893. Bd. 2

Delbrück 1893 ― Delbrück Bertold. Vergleichende Syntax der indogermanischen Sprachen, Strasbourg, 1893

Disterheft 1997 ― Disterheft Dorothy. The evolution of Indo-European Infinitives // Roger Pearson (ed.) Studies in honor Jaan Puhvel. Washington: Institute of man, 1997

García Ramón 2001 ―  García Ramón José Luis. Impératif et infinitif pro imperativo dans les texts grecs dialectaux : les Lois de Gortyne // Verbum, 2001, v. XXIII, № 3 : Les modes dans les dialects grecs anciens

Sgall 1959 ― Sgall Petr. Studien zu Infinitiven in RigVeda. Praha: Academia, 1997