The Chronological Development of Adjectives in –aya up to 725

In a couple of earlier contributions I have indicated the potential interest of certain word formations, and in particular adjectival forms in –aya, as providing some useful criteria for dating undated texts.1 In the earlier of the two, my focus was primarily on the increasing use of certain word formations employing the suffixes –ana, -anuta, -ana’it, and adjectival forms in -aya only received mention at the end.2 With the collection of further materials, however, it became evident that these adjectival forms in –aya in fact provided much the best evidence for diachronic developments, and it was with these that the second study was exclusively concerned.

Anyone who is at all familiar with Syriac literature is likely to have observed that earlier Syriac texts are sparing in their use of adjectives, whereas later ones often have a profusion. Having now collected a large amount of material from texts both published and unpublished for the period up to AD 725, it seems possible to offer with a certain amount of confidence an outline of the development in the use of these adjectival forms, presented in the form of a basic alphabetical inventory of them in which, for each entry, an indication is given for the earliest attestation (the century is indicated by a Roman numeral), followed by (usually) single examples from later writers, in particular the poets Narsai, Jacob of Serugh and ‘Isaac of Antioch’, as well as the main prose writers of the fifth/sixth – early eighth century (the death of George of the Arab tribes in 724 and the dated (721) translation of Sophronios’ Letter to Arcadius have provided the cut-off date of 725). Since many adjectives in –aya will have first been introduced into usage through Syriac translations of Greek texts, considerable attention has been paid to those Greek authors who were clearly influential, above all Athanasius, Cyril of Alexandria, Gregory of Nazianzus, John Chrysostom and Theodore of Mopsuestia and Severus, the Syriac translation of several of whose works can be approximately dated. Since a considerable number of these Syriac translations have not yet been published, I have made use of the rich holdings of early Syriac manuscripts in the British Library which contain many of these unpublished translations. Even with such an extensive range of source materials from which the lexicon below is drawn, it is of course likely that future work on editing further texts will bring to light earlier and/or additional examples of these adjectival forms; accordingly the present contribution should be regarded as being only preliminary in character.

It is important to draw attention to two important features which the inventory does not address, namely frequency in any given author, and semantics. I hope to address the former in particular in some future studies; here, however, it is only for fourth-century texts that more than a single reference is given for each author.

Needless to say, for purposes of dating undated texts this inventory should not be used in isolation, but alongside a whole number of other criteria which are helpful for this purpose.3

The Sources Used

The materials upon which the inventory is based consist of:

(1) Peshitta Old and New Testaments.

(2) Writings earlier than c.400, thus predominantly Aphrahat, Ephrem and the Liber Graduum, for all of whom several references (where they exist) are given.

(3) All the published texts of Narsai (thus including those in the edition by the Patriarchal Press), likewise all the published texts of Jacob of Serugh, both prose and verse (including the 160 further memre published by Akhrass and Syryany), and the various works under the name of Isaac of Antioch published by Bickell and Bedjan; here it should be noted that a large number of works by ‘Isaac’ remain to be published.

(4) The main published prose writers of the fifth/sixth century, notably (and in roughly chronological order) John of Apamea, Philoxenus, Cyrus and Thomas of Edessa, Daniel of Salah, and John of Ephesus; and those of the sixth/seventh century, notably Barḥadbeshabba, Sahdona, Babai, Isho‘yahb III, Dadisho‘, Isaac of Nineveh, Jacob of Edessa, George of the Arab tribes. These are supplemented, primarily when they attest additional forms, by a number of other sources such as Gregory of Cyprus, John of the Sedre, Athanasius of Balad, the Synodicon Orientale, and others.

(5) Syriac translations, introduced by TR in the inventory, of the main influential Greek authors, as noted above. Several of these translations can be at least approximately dated, and in some cases exist in two versions, an earlier and a later one, usually a revision. Particularly important for the present purposes are those translations which can be approximately dated; details of these are given below under Dating. Since it appears that around the turn of the fifth/sixth century was the time when a number of important new adjective forms came into use, special attention is paid to the Canons of Nicaea and subsequent Synods, preserved in a manuscript dated 500/1, the original translations of Gregory of Nazianzus’ Discourses and of Severus’ Homilies, and polemical works against Julian of Halicarnassus.

(6) Since hagiographical texts are usually anonymous and for the most part difficult to date, these are only rarely cited (primarily when they provide witness for a rare entry).

It should be noted that, in the case of authors listed in (4) and (5), not all surviving works covered in my materials have been cited in the Inventory; thus, for example Basil’s on the Holy Spirit, Theodore of Mopsuestia’s Against the Macedonians, and the texts in W. Strothmann’s Johannes von Apamaea (1972), to mention only the most obvious, do not feature since they provide no further information beyond what is found in the other works of these authors which are cited.

Needless to say, the number of the adjectival forms in –aya continues to increase considerably after the cut-off date of 725; simply as a pointer in this direction I have included a limited selection of these drawn from liturgical texts, in particular anaphoras. These have the indication ‘post VII’.

PROBLEMATIC CASES

Adjectives in –aya originated as adjectival forms of numbers (ordinals), and to denote ethnic origin. Both these categories are excluded, along with the points of the compass. Certain of these forms in –aya function primarily as nouns, and only rather rarely as adjectives; this applies to cases such as aksnaya, barbraya, iḥidaya, nukraya and ‘almaya (‘layman’); in the repertory below it is only instances of the adjectival use of such words that are included.

In unvocalized texts is it impossible to tell whether a feminine ending should be vocalized –ita (attached to words ending –ana, e.g. pagrana), or –ayta (thus belonging to an adjective ending –anaya). This ambiguity can remarkably be seen in Bedjan’s occasional vocalization shmayanayta (thus e.g. in his edition of Jacob of Serugh, v, 439), rather than shmayanita, the former being the feminine singular of shmayanaya, whereas the latter (certainly correct in this case) is from shmayana. Nevertheless one should note an isolated case of shmayanaya in Narsai: in his memra on Enoch and Elijah the phrase nasha shmayanaye occurs (ed. Frishman, I, 486), where the four-syllable form is required by (and perhaps was also occasioned by) the metre.

It also needs to be remembered that scribes may sometimes update earlier forms; a good example of this is Sirach 47:21, where 7a1 (and de Lagarde’s edition) has ḥanpayta, whereas the original reading was almost certainly ḥanpta (thus preserved in the Mosul edition). This form of updating may well also apply to the single apparent occurrence of ar‘anaya in Ephrem, H. de Ecclesia 49:2, where Beck’s text has ḥashe ar‘anaye; this, however, goes against the metre, which suggests that the original reading was ḥashe ar‘ane.

Approximately Datable Translations

Translators rarely get mentioned, and so the majority of translations are anonymous, and so not easy to date. Various criteria, however, are available, with the help of which an approximate date for a translation can be assigned. This is the case notably where:

(1) The rare cases where the date of the translation is given, or a firm terminus ante quem is available:

(a) Harklean and Syrohexapla.

(b) Severus, Against Julian of Halicarnassus. The trans-lation was made (or perhaps just commissioned) by Paul of Kallinikos in 528.

(c) PsDionysius; the corpus was translated by Sergius of Resh‘aina, who died in 536. Since only part of the corpus has been published, I have also made use of the single main witness, Sinai Syr. 52. The subsequent revision by Phokas of Edessa in the late seventh remains almost entirely unpublished, and it has only been used here for the section on the Myron in the Ecclesiastical Hierarchy, edited by Strothmann.

(d) Paul, bishop of Edessa, was very active as both translator and reviser of earlier translations; at least some of this work was undertaken in 623/4 while he was in Cyprus, whither he had taken refuge from the Persian occupation.

(e) The dates for some of Jacob of Edessa’s translations/revisions are known.

(f) Constantine of Edessa translated Sophronius’ Letter to Arcadius in 720/1, as is specifically mention in the colophon of BL Or. 8606, copied in Edessa two years later.

(2) The date of the manuscript containing the work provides a terminus ante quem:

(a) This applies above all to the texts contained in BL Add. 12150, dated 411 (Edessa), notably Eusebius’ Theophania and Titus of Bostra’s work Against the Manichaeans.

(b) Cyril of Alexandria’s Worship in Spirit, preserved in Add. 12,166, dated 553.4

(c) PsZacharias, Historia Ecclesiastica, is preserved in Add. 17,202, dated c.600.

(d) Cyril of Alexandria’s Thesaurus, in Add. 12,135, dated 611.

(3) Internal criteria. Philoxenus’ Commentary on the Prologue of John is witness to a concern for precision in rendering Greek terms having a bearing on Christology, in particular esarkothe and enanthropesen, and in the case of both terms a progression in time over usage has been charted; thus for esarkothe the earliest rendering, lbesh pagra, is attested in fourth-century authors and in the original form of the Creed at the Council of Seleucia-Ctesiphon (410); in the fifth century etgashsham comes into widespread use, and at the end of that century the neologism etbassar first appears, probably at the same time as another neologism, etbarnash, a calque on enathropese, for which the earlier rendering had been hwa barnasha.5 These two terms thus provide a valuable terminus post quem of late fifth century for any translation containing them.

(4) Both an original translation and a revision are known: this applies in particular to:

(a) Gregory of Nazianzus’ Discourses were originally translated probably c.500, and revised by Paul of Edessa in the first half of the 620s. The original translation is only preserved in part; to supplement the small amount of the original version so far published from it, use has been made of two manuscripts containing it.

(b) Severus’ Homilies were first translated in the first half of the sixth century, and then revised by Jacob of Edessa in the late seventh century and completed in 700/1. With only a few exceptions, only the revision has been published; since the original translation is obviously an important witness for new forms in –aya, resort has been made to some early manuscripts containing it.

(c) Severus’ Hymns were first translated by Paul of Edessa in the 620s, and then revised by Jacob of Edessa in 675; the published edition contains Jacob’s revision.

(d) Sergius of Resh‘aina’s translation of the Ps-Dionysian corpus, revised by Phokas, has already been mentioned under 1(c).

(e) The sixth-century translations of Porphyry’s Eisagoge and of Aristotle’s Categories were revised in the seventh century by (respectively), Athanasius of Balad in 645, and Jacob of Edessa.

(f) Basil’s Homilies, translated in a very expansive manner, probably early in the fifth century, required a completely new translation in the seventh century; some of this new translation is preserved in Cambridge Add. 3175.

For the purposes of the present contribution it is primarily (a) and (b) which have been used.

(5) On grounds of style of translation. Study of the development of translation technique for biblical and authoritative patristic texts has revealed a steady development from reader- to text-oriented translation style, the turning point being c. 500. A considerable number of translation features, and above all a verbum ad verbum approach and its consequences, can serve as excellent indicators of later sixth- and especially seventh-century translations.

Chronological Distribution

These adjectival forms can be distributed in chronological sequence as follows (underlined forms would seem to be particularly significant for the purpose of suggesting dates for undated texts):

Some preliminary observations

The ever increasing number of newly attested forms in each of the periods is striking: period V/VI in particular marks a turning point in the rate of increase which culminates in the periods VI/VII and VII. This is certainly not just due to the number of surviving texts available for the different periods, although this is a factor that always needs to be taken into consideration. The turn of the fifth/sixth century also happens to be the turning point in the history of translation practice,20 at least as far as authoritative texts are concerned, for it marks the fundamental shift from a reader- to a text-oriented approach to translation. One of the consequences of this will be a desire to reflect the grammatical categories of the Greek originals more closely, thus creating a need for new Syriac adjectival forms to correspond to the higher frequency of the presence of adjectives in Greek writings. Evidence of experimentation by translators can be seen from a small number of forms which never caught on, such as nuraya (in contrast to nuranaya).

A surprising feature which emerges from these Tables is the large number of new forms to be found in East Syriac authors of the VI/VII and VII periods, only a few of which are also (so far) attested in earlier translations from Greek. A noticeable presence amongst these new adjectives are forms in –tanaya; earlier forms are all extensions of feminine nouns, such as ‘idta > ‘idtanaya, but in the last two periods, not only are new examples attested, but an extension is made by analogy to include other suffixed forms in –tana, the most notable case being (ṭubana>) ṭubtana > ṭubtanaya.

One further notable feature is worth observing, namely the extension of forms in –aya to include biblical names and places: the earliest to be attested is adamaya, found quite a number of times in Narsai, but the extension to other names only comes with the sixth century and later. Similarly, adjectives which take their origin in biblical passages are first attested in the seventh century (see under ḥalanaya, ‘built on sand’, cf. Matt. 7:26), while their growth in frequency belongs to the eighth century onwards.

As mentioned at the outset, the prime reason for providing the present repertory was to provide a reasonably firm chronological framework for the expansion of these adjectival forms, certain ones of which can usefully serve as one of several different criteria available for dating works which are either undated or of uncertain authenticity. A few instances of cases where this criterion could be helpfully applied were given in the earlier article, ‘A criterion for dating undated Syriac texts’, the most dramatic of which was the tell-tale presence of five adjectival forms in –aya21 in one of Narsai’s liturgical memre (ed, Mingana, no. 17), none of which are to be found in any other memre by Narsai, and all of which are subsequent innovations. It would not be difficult to apply this criterion to other texts with informative results, but this must await another occasion – and for others to apply it. Looked at from a much wider perspective, the evidence presented here also serves as an illustration of one of the ways in which Syriac literary culture became more and more Hellenized over the course of the fourth to early eighth century.22

Works cited

• I.E. Barsaum, Sriṭotho d-dayro d-kurkmo/Deyrul-Zafaran Manuscripts, II (Damascus, 2008).
• S. P. Brock, A. M. Butts, G. A. Kiraz and L. van Rompay (eds), The Gorgias Encyclopedic Dictionary of the Syriac Heritage (Piscataway NJ, 2011).
• S.P. Brock, “Towards a history of Syriac translation technique’, in R. Lavenant (ed.), III Symposium Syriacum (OCA 221; Rome, 1983), 1-14.
• - , “Diachronic aspects of Syriac word formation: an aid for dating anonymous texts”, in R. Lavenant (ed.), V Symposium Syriacum (OCA 236; Rome, 1990), 321-30.
• - , “Some diachronic features of Classical Syriac”, in M.F.J. Baasten and W. Th. van Peursen (eds), Hamlet on a Hill. Semitic and Greek Studies presented to Professor T. Muraoka (Orientalia Lovaniensia Analecta 118; 2003), 95-111.
• - , “Secondary formations from Greek loanwords in Syriac”, in H. Juusola, J. Laulainen and H. Palva (eds), Verbum et Calamus. Semitic and Related Studies in Honour of Professor Tapani Harviainen (Studia Orientalia 99; Helsinki, 2004), 31-39.
• -, “A criterion for dating undated Syriac texts: the evidence from adjectival forms in –aya”, Parole de l’Orient 35 (2010), 111-24.
• - , “Charting the Hellenization of a literary culture: the case of Syriac”, Intellectual History of the Islamicate World 3 (2015), 98-124.
• A.M. Butts, K. S. Heal and S. P. Brock, Clavis to the Metrical Homilies of Narsai (CSCO Subsidia 142; Leuven, 2020).
• A.M. Butts, Language Change in the Wake of Empire. Syriac in its Greco-Roman Context (Winona Lake, 2016).
• - , “The use of Syriac derivational suffixes with Greek loanwords”, Orientalia 83 (2014), 206-37.
• A. de Halleux, “Le symbole des évêques perses au synode de Séleucie-Ctésiphon (410)”, in G.Wiessner (ed.), Erkenntnisse und Meinungen, II (Göttinger Orientforschungen, Reihe Syriaca 17; 1978), 161-90.
• - , “La philoxénienne du symbole”, in Symposium Syriacum 1976 (Orientalia Christiana Analecta 205; 1978), 295-315. .
• J. Gribomont, “Le symbole de foi de Séleucie-Ctésiphon”, in R.H. Fischer, A Tribute to Arthur Vööbus (Chicago, 1977), 283-94.
• H. Hugonnard-Roche, “Questions de logique au VIIe siècle: Les Épitres de Sévère Sebokht et leurs sources grecques”, Studia Graeco-Arabica 5 (2015), 53-104.
• D. King, The Syriac Versions of the Writings of Cyril of Alexandria. A Study in Translation Technique (CSCO Subsidia 123; Leuven, 2008). .
• L. van Rompay, “Some preliminary remarks on the origin of Classical Syriac as a standard language: the Syriac version of Eusebius of Caesarea’s Ecclesiastical History”, in G. Goldenberg and S. Raz (eds), Semitic and Cushitic Studies (Wiesbaden, 1994), 70-89. .