I. Content and Organisation

I.1. Structure of the Volume

I.1.A. Preliminary remarks

In the Introduction, we address several major issues, such as the Structure of the Corpus, Sources, and Historiography, as well as more specific questions concerning Geography, Chronology, Historical Context, Typology (material, function, formulae), as well as the current Location of the epigraphic monuments. With the exception of atypical formulae, Corpus entries will contain references to the discussion of corresponding formulae in the Introduction. At the same time, this Introduction will serve as a general survey of the main issues connected with the study of Byzantine inscriptions originating from the Northern Black Sea region.

I.1.B. Inscriptions

The inscriptions are grouped by region, and within each region they are further grouped by location, category of text, formula, name (in that order and from larger to smaller groupings).

I.1.B.a. Region

Inscriptions in the Corpus are organized into six large groups corresponding to six regions, each characterized by a geographic, cultural, and historical unity. We will refer to these regions by the following names: (1) Dniester Estuary, (2) Cherson and the Herakleian Peninsula (from the river Chernaya to Cape Feolent), (3) Mountainous Crimea (Southwest Crimea except Cherson as described above), (4) South and Southwest coast of Crimea, (5) Kerch, (6) Taman Peninsula and Kuban. Inscriptions of unkown provenance are discussed as a separate group. A special note on the use of the place name 'Cherson' is in order. In this publication, I use the name 'Cherson' to refer to the medieval city, which occupied the territory of the ancient Greek polis of Chersonesos. The name 'Chersonesos' (also 'Tauric Chersonesos') is used here in reference to that ancient geo-political entity and also to the archaeological site of the National Preserve of Tauric Chersonesos. Medieval, that is, Byzantine, Cherson is to be distinguished from the modern-day 'City of Kherson,' the capital of Chersonskaya county in Ukraine. The boundary between Mountainous Crimea and the South and Southwest coast of Crimea is formed by the crest of the south slope of the Main chain of the Crimean mountains, stretching from Balaklava to Theodosia (Pavlova 1964, 3–6, 57–89; Lebedinsky 1976, 3–6); accordingly, Inkerman falls under Mountainous Crimea (but also due to the nature of most epigraphic monuments from the area (except V 153); and due to the absence of Byzantine inscriptions from the left bank of the river Chernaya, which belongs to the Region of the Herakleian Peninsula). The eastern boundary of the region of Taman Peninsula and Kuban does not extend beyond Anapa (ancient Gorgippia): further to the northeast, prosopography reflects the presence of Adygean names and words (see, e.g., Yaylenko 1987, 169 (stanitsa Gostagayevskaya)), and hence a different socio-cultural sphere; for the same reason, we exclude Latyshev 1896, no. 102 from this collection.

I.1.B.b. Find place

Within the Regions of Dniester Estuary, Mountainous Crimea, South Crimea and Taman Peninsula, we use modern place names to designate areas where inscriptions were found, and arrange them in the order of the Russian alphabet. Old names (this primarily concerns the villages of Mountainous Crimea) are supplied in parentheses. In as much as we were able to ascertain, modern toponyms used in the Corpus are up to date.

I.1.B.c. Category of text

Within the Regions of Cherson and Kerch, as well as within specific modern population centers in those Regions, inscriptions are grouped according to category of text in the following order: building, dedicatory, demonstrative, apotropaic, invocations, image-related, funerary, graffiti, liturgical, and of uncertain function (including fragments).

I.1.B.d. Formula

When a particular category of text is represented by a large number of examples (this mainly concerns the monuments from Cherson and the tombstones of Kerch), we group them according to formulae rather than chronology. This is justified by the fact that in many cases the dates of inscriptions are rather approximate.

I.1.B.e. Name

Inscriptions of the same category are arranged by the first letter (in Greek alphabetical order) of the Name of a person mentioned in the text (if several names are mentioned, then according to the first name). Inscriptions with the same Formula and the same Name are arranged in chronological order (date of the editio princeps) or according to their position within a complex. Insciptions of the same category, without names, do not follow a particular order. An exception to the principle of arrangement according to Names are Building Inscriptions, arranged in chronological order. In addition, in the latter category, official inscriptions precede private ones.

I.1.C. Indices and Concordances

The volume is accompanied by the following Indices: Greek words and fragments of text in Greek; Personal Names; Attested persons; Geographical Terms, Rulers of Rome, Byzantium or Bosporan kingdoms; Divine, religious or mythic figures; Geographic names and Find Places. In addition, we include indices of Symbols, Numerals, and Months; a Concordance (comparatio numerorum), as well as Bibliography, and a list of Abbreviations.

Personal names mentioned in inscriptions are rendered in English according to the following principles: (a) if a name has a common English equivalent, e.g., Alexander, Helen, George, etc., that form of the name is used; (b) when a name has no modern English equivalent, an orthographic transliteration is used; (c) names of Roman origin, e.g., Felix, that appear in Hellenized form in the inscriptions are restored to their Latin transcription.

Names of medieval cities are given according to the usage of the time: e.g., Cherson, and not Chersonesos. In those cases when several names are known, we choose the older, or the etymologically more sound one (e.g., Tamatarkha, not Matrakha). In other cases, we use one or the other depending on the date of the inscription, e.g., Pantikapaion for IV-Vth centuries, and Bosporos for VI-XVth centuries. We distinguish in spelling between Bosporus, the region, and Bosporos, the medieval city.

I.2. Criteria of selection

I.2.A. Material and character of monuments

Included in this volume are lapidary inscriptions, i.e., all those made on free-standing stones and on rock surfaces, including inscriptions accompanying images, and graffiti. Inscriptions on frescoes are included, except those accompanying images. We exclude inscriptions on portable objects, such as icons, vessels, armour, tools, and weights. Monograms are also excluded.

I.2.B. Find Place

In geographic terms, this collection covers the entire length of the northern Black Sea coast, from the mouth of the Dniester to the eastern shore of the Taman peninsula (see I.1.B.a), including all of Crimea. As a separate group, we present the monuments, generally attributed to the northern Black Sea, but without exact provenance. Where some indications of possible provenance are present, the inscription is placed at the end of that regional grouping.

I.2.C. Chronological framework

We include in this collection all Greek inscriptions dated between the Vth century and 1475 C.E. From the IVth century, we include only the monuments of Christian nature (marked with a cross or christogram, or with Christian content). The lower limit of the collection, year 1475 C.E., is chosen for two reasons: firstly, it marks the fall of the Princedom of Theodoro, the last Greek-speaking state in the Northern Black Sea area; secondly, it marks the fall of the Genoese fortress of Sudak, where Greek epigraphy is also attested until then. We also include in this collection inscriptions without a secure chronological attribution, which might be Late Byzantine or post-Byzantine in date, but referring to activities of the Greek-speaking population in the region.

I.2.D. Sources

The information about the inscriptions included in this volume comes from several sources. First and foremost, it comes from the surviving epigraphic monuments, preserved mostly in the local museums of the regions, where they were originally found, i.e., in Crimea (Central Museum of Taurida, Alushta Branch of the Central Museum of Taurida, Bakhchisaray historical and cultural preserve, National Preserve of Tauric Chersonesos, Yalta State Museum of History and Literature, Theodosia Museum of Antiquities, Kerch Historical and Cultural Preserve) and in Taman peninsula (Historical and Ethnographic Museum of Temryuk), but also in other museums in Russia (The State Historical Museum, The State Hermitage, Museum of History of Religion), Ukraine (Odessa Archaeological Museum, Local History Museum of the City of Kherson), and France (Musée du Louvre), as well as those still in situ. Virtually all inscriptions with a known place of storage have been inspected by the author de visu. The inscriptions that have been lost since their discovery or first publication are presented on the basis of earlier editions or drawings (either published or preserved in archives).

II. History of Studies and Collections

This topic could serve as a subject of a special monograph, or several, and requires not only analysis of publications, but also of archival materials. Here we present only a brief survey of the matter in order to enable the reader to understand better how a particular inscription fits into the history of collecting, publishing, and research (full references to particular studies can be found under corresponding names in the Bibliography).

II.1. End of the XVIIIth – first half of the XIXth century – an era of travellers and antiquarians

Educated travellers of the late XVIIIth-early XIXth centuries were among the first to take note of the Byzantine inscriptions in the Northern Black Sea area: F.K. Marschall von Bieberstein in 1793-1794 and in 1807-1826 (V 219 and V 340), P.S. Pallas in 1798 (V 6 and V 13), P.I. Sumarokov in 1799 and 1803 (V 180, V 315, V 316, V 330), N.A. Lvov in 1803 (V 330). At the same time, scholars with antiquarian interests also began to take interest in Byzantine inscriptions, e.g., D.L. Oderico in 1792 (V 158) and L. Waxel in 1803 (V 149, V 315). In 1830s-1840s, a new generation of travellers continued the trend: P. Koeppen (V 200), F. Dubois de Montpéreux (V 208, V 235), N.N. Murzakevich (V 1, V 2, V 316, V 321, V 335, V 315) and archbishop Gavriil (V 316). The first controversy surrounding an inscription (V 6) also dates to the beginning of the XIXth century, with a whole series of scholars weighing in on the matter: P. S. Pallas, L. Waxel, E. D. Clarke, D. Raoul-Rochette, E.-M. Cousinéry, G. L. F. Tafel and others. The main achievement of this period was the first systematic publication of the Byzantine inscriptions of the Northern Black Sea in CIG IV, published in 1856–1859 (V 6, V 13, V 149, V 180, V 313, V 335, V 158, V 315). An interest in collecting Byzantine inscriptions also dates to this period: V 180 finds its way into Sably, the Crimean estate of the Borozdin family.

Overall, this period is characterized by rather unsystematic and often unprofessional explorations, an absence of excavations, and an haphazard nature of publications.

II.2. 1850-1870s – first excavations and museums

From the middle of the XIXth century onwards, the interest in collecting grows: in 1848, the museum of the Odessa Society of History and Antiquities acquires V 3 from Akkerman; in the early 1850s, the Hermitage receives V 330 from Taman; and at the same time, V 57 from Cherson turns up in the collection of the French general Robert and later makes it to the Louvre. The beginning of excavations in a number of places gives a new boost to discoveries and a new scope to collecting: excavations take place in the area of Taman in 1853 under the direction of K.R. Begichev (V 325) and in the same year in Cherson -- under A.S. Uvarov (V 20, V 50); in 1861 — during the construction of a cathedral (V 68); in 1877–1878 — on the initiative of the Odessa Society of History and Antiquities (V 21, V 22, V 89); in 1878 — under the direction of Odessa Society of History and Antiquities in Kerch (V 268) and under D.M. Strukov in 1871 at Ay-Vasil (V 232) and Partenit (V 241, V 242, V 244). As a result, such notable collections as the museum of Odessa Society of History and Antiquities (now the Оdessa Archaeological Museum), the Chersonessan Repository of Antiquities (now the National Preserve of Tauric Chersonesos) and the Museum of the Melek-Chesmen Mound (now part of the Kerch Historical and Cultural Reserve) were formed. The tradition of discoveries in the course of academic travels also continues: by F.A. Struve in Shabo in 1866 (V 4) and by D.M. Strukov in 1870s at Inkerman (V 151, V 159, V 152, V 154, V 156) and Sudak (V 247) – the findings of Strukov remained unpublished.

The main tendency of this period is for greater systematization of the search, collection, and publication of inscriptions, although short of their thorough study.

II.3. Late XIXth - Early XXth century – an era of systematic searches and publications

This period witnesses the blossoming of research in the field of Byzantine epigraphy of the Northern Black Sea region. Two main reasons underlie this development. Firstly, the systematic exacavations of К.К. Kostsyushko-Valyuzhinich, and later of R.Ch. Loeper in Cherson, surveys of A.L. Bertye-Delagard and M.I. Skubetov in Mountainous Crimea, as well as the efforts of K.E. Dumberg, V.V. Shkorpil and Yu.Yu. Marti, aimed at collecting the Early Byzantine tombstones discovered in Kerch at the turn of the century. Their activities led to the discovery of most Byzantine inscriptions from the Northern Black Sea known today. Also, during this period, Burachkov and Lyutsenko assemble their private collections, which later join the holdings of the State Historical Museum. Secondly, starting from 1882, all newly found monuments come under the regular inspection of V. V. Latyshev, who fully deserves to be called the Father of the Northern Black Sea Epigraphy. Following his first publications in the academic periodical "Materials on the Archaeology of Russia" (for the years 1892–1899), Latyshev produced (in 1896) the only existing compendium of Byzantine and post-Byzantine Greek inscriptions of the region – “Collection of Greek Inscriptions of the Christian Era, from the South of Russia,” which included about 120 lapidary inscriptions. This collection consisted of three parts: inscriptions published by Latyshev’s predecessors, Latyshev’s own publications from 1882 till 1895, and various additions made in the process of working on the Collection. Latyshev, it should be noted, was not entirely happy with this publication, as it had been put together in a rush to mark the anniversary of the Imperial Russian Archaeological Society. He periodically returned to his editions of texts and revised them, often stimulated by polemics with other scholars, primarily Yu.A. Kulakovsky and G. Millet. Subsequently, up to 1918, Latyshev published new finds of inscriptions in “Proceedings of the Imperial Archaeological Commission”, as well as in the “Notes of the Odessa Society of History and Antiquities” and in the “Proceedings of Taurida Learned Archival Commission.” Besides Latyshev, other scholars of the time engaged in the publication of epigraphic monuments: W. Jurgiewicz, V.F. Miller (V 234), as well as Loeper, Shkorpil, Marti and Kulakovsky, mentioned above, although their studies did not reach the same standard of academic quality as those of Latyshev. Among the main shortcomings of this period, we may list the lack of attention to the archaeological context of epigraphic finds, little use of comparanda from other parts of the Byzantine world, as well as the disappearance, without a trace, of numerous monuments, after their publication or transcription.

II.4. Soviet period – an era of unsystematic studies

Due to the atheistic ideology of the Soviet regime, the study of any subject related to Christianity was frowned upon in that period. It is not surprising then that the study of Byzantine inscriptions of the Northern Black Sea was limited to the publication of new finds, and not all of them at that. In the period before WWII, only N.V. Malitsky (V 175, V 176, V 177, V 180) and M.A. Shangin (V 108, V 42) applied themselves to the study of Byzantine inscriptions, but their work was riddled with substantial mistakes.

After WWII, some new works appear in print: N.P. Rozanova (V 334), M.A. Tikhanova (V 183; she studied V 331 already in 1926), yet they suffer from the same deficiencies as the earlier publications. Also at this time, V.D. Blavatsky published V 275. The only systematic work on Byzantine inscriptions is undertaken by E.Ch. Skrzinska and E.I. Solomonik, although their efforts were also far from exemplary (see commentaries). A selection of Byzantine inscriptions was published in CIRB, although the principle of selection remained unclear. There are references to many Byzantine inscriptions from Cherson in the works of A.L. Jakobson. Of the authors who published in the West, only a native of Russia, B. Nadel, addressed Byzantine inscriptions of the Northern Black Sea (V 265, V 330).

The museums fared somewhat better. Many new finds ended up at local museums (Yalta, Alushta, Temryuk, stanitsa Troitskaya); an important Museum of Cave Cities (now Bakhchisaray historical and cultural preserve) was founded. Two inscriptions from Cherson were transferred to the Museum of History of Religion. We should also note, however, that several inscriptions from Cherson and Bakhchisaray disappeared during WWII, as well as much of the Simferopol collection, which disappeared from the museum of the Taurida Learned Archival Commission.

II.5. 1980-2000 – revival of interest

From the mid-1980s, an interest in Byzantine inscriptions increases. Remarkably, after the unsystematic approach of the Soviet period, researchers immediately turn to producing Corpora. In 1987, V.P. Yaylenko announces a project (still unrealized) aimed at publishing Byzantine inscriptions from the territory of the USSR. For a start, he publishes four new inscriptions from the Northern Black Sea (V 52, V 183, V 338, V 312), not without mistakes. A “Corpus of Christian Inscriptions from Bosporus” cannot be properly considered a Corpus: at best, it is an inventory that contains numerous mistakes in Greek, only the most basic information about inscriptions, and offers no new critical editions. At the same time, as before, publications of new finds, as well as new editions of previously known texts, continue. Yu.G. Vinogradov publishes an interesting (but not uncontroversial) study of the Bosporan inscriptions of the later Vth century.

Of European scholars, D. Feissel discusses V 329, and K. Zuckermann - V 5 and V 226. А. Brzóstkowska publishes V 91 (for discussion of deficiencies, see commentary). Also of note are publications of inventories from the lapidary collections of the Bakhchisaray historical and cultural preserve and the State Historical Museum; as well as an Album of Illustations for CIRB.

III. Geography and Chronology

There are good reasons for treating the issues of geographic distribution of inscriptions within a region in conjunction with their chronological distribution, while at the same time taking into consideration their historical context. On the subject of regions, see above I.1.B.a. Chronologically, with a view to the history of the Byzantine Empire, the inscriptions of the Northern Black Sea can be divided into three main periods (on the chronological limits of this collection, see above, I.2.c): Early Byzantine (IV-VII), Middle Byzantine (VIII-XII), and Late Byzantine (XIII-XV centuries). Early Christian (prior to IVth century) inscriptions are not attested in the Northern Black Sea region.

Dniester Estuary

Due to the fact that only three Byzantine inscriptions come from this region, we can merely note that they attest building activities of Moldavian princes in the middle of the XVth century, in Belgorod-Dnestrovsky (V 1 and V 2). The origin and nature of V 4 are a bit of a riddle (see commentary).



In Cherson, Early Byzantine epigraphy is the most widely represented, both thematically and quantitatively. Several characteristics are notable. All building inscriptions concerning fortifications are set up on behalf and by representatives of the central Byzantine authority (see below, IV.3.a), which is not the case with dedications of basilicas (V 28 might be an exception), where the initiators are typically wealthy citizens of Cherson. Such correlation is in accord with the historical situation whereby the imperial authorities take care of the city’s military security, while the city itself is wealthy enough to provide for extensive church-building. At the same time, we still observe expressions of polis ideology in the Vth century (V 6, 478-488 C.E. is set up on behalf of the city and dated by the old Chersonesian era) and even possibly at the turn of the VI-VIIth centuries (V 24 can be dated by the same local era).

In the Middle Byzantine period, building inscriptions testifying to local initiative disappear. They give way to invocations, primarily in churches, although demonstrative and funerary inscriptions are represented in this period in equal numbers. The “Dark Ages” (2nd half of the VIIth – early IXth centuries) are marked by V 16, containing non-Greek names and a unique formula. The establishment of the Theme (ca. 840 C.E.) meant political stability, which led to an increase in the number of epigraphic monuments, although not of building inscriptions. Construction of fortifications remains a prerogative of the imperial and thematic authorities (see V 11, 1059 C.E.).

Finally, in the Late Byzantine period, we mostly have building inscriptions: one local (?), from cape Feolent and two Theodorite (V 13 and V 14), as well as one dedication from the Herakleian Peninsula (V 108).

Byzantine inscriptions found in Cherson fall into three distinct groups: building inscriptions – from city walls and towers; funerary – from necropoleis to the east of the city; and the rest (invocations, apotropaic, etc.) – both from the territory of the city and from suburban churches. A number of inscriptions were found within the ancient citadel, but in no relation to any particular monument, while V 108 comes from a settlement in the Berman's Gully on the Heracleian Peninsula.


The following inscriptions date to the Early Byzantine period: V 5 (construction of a wall, 392-393 C.E.) from the southern section of the wall, enclosures 25 and 26); V 6 from Tower XVII (renovation of walls, 487-488 C.E.); V 7 (about a construction, 565-574 C.E.) from the southern section of the wall, Curtain Walls 17-18, inner side; V 10 (V-VIth centuries) near the Hellenistic gates. For V 8 (restoration, IV-VIIth centuries) the findspot is unknown.

These inscriptions date to the Middle Byzantine period: V 11 (construction and renovation of gates, 1059 C.E.) from the wall in front of Tower XXI, outer side; V 12 (reconstruction, XIth century) from the wall near Tower XVII. We can assign to the Late Byzantine period V 13 (2nd-3rd quarter of the XVth century) and V 14 (renovation of the fortress and walls (?), 1432-1441 C.E.) from the southern section of the wall, enclosures 25 and 26. These two inscriptions raise the possibility of the presence of Theodorites in Cherson and about their involvement in the construction of fortifications.

V 15 dates to the VIIIth – first third of the IXth century. Its findspot is unknown, but it speaks of some construction, and is set up on behalf of the Chersakoi, that is, citizens of Cherson.


The greatest number of Byzantine inscriptions, except for funerary, is found in churches. Mostly, these are basilicas and martyria, and in addition (especially in the Middle Byzantine period) chapels. In most cases, inscriptions relate to the church where they were found. First and foremost, they are Early Byzantine building inscriptions and dedications of basilicas: V 9 and V 94 from Church 15; V 19, V 28, V 29 (?) from Church 20; V 23, V 30 from Church 23; V 21, V 22 from Church 36; V 44 from Church 28 (?). From the martyria (including suburban) come: V 104 from Church 19, and V 24 and V 31 from Extramural Cruciform Church. From the “Church with arcosolia” comes V 27, and from Church 17 — V 23, while V 18, V 25, V 32, V 49 are of unknown provenance.

The following inscriptions date to the Middle Byzantine period: invocations (V 47 from Church 13 (Little Church В), V 48 from Extramural Cruciform Church (?), V 50 from Church 23), liturgical formulae (V 97 from Church 19) and graffiti (V 89 from Church 36). Construction of churches is also mentioned in V 16 and V 17 (both found in Block IX, but not in situ). In this period, we begin to find burials inside churches: V 66 from “Church inside Church” on Maiden's barrow (Devichya Gorka), V 68 from Church 26 (?), V 70 from Church Е in Block I (?). The only example from the Early Byzantine period is V 72 from Church 19, and from the Late Byzantine period - V 107 from the “Chapel of 1896", on Cape Feolent.

Necropolis of Quarantine Bay

This is the conventional name applied to a cemetery that stretches from the citadel of Chersonesos to the coast of the Quarantine Bay. Included in this area is the so-called Monastery Animal Farm (excavations of 1905 and 1908). Here, the following Early Byzantine tombstones were found (a date is provided for later ones). In 1896: V 76 (Burial vault 785), V 56, a relief dated to the IV–Vth centuries (Burial vault 784, earth infill)— was moved here by accident, as rubbish from the eastern part of the site. In 1905: V 63 (Burial vault 1951), V 45, invocation, IX–Xth centuries (Burial vault 1662 ) — was moved here by accident, from a nearby citadel. In 1907: V 62 (Burial vault 2281), V 71 (Burial vault 2383), V 78 (Burial vault 2367), V 87 (mound; Х–XIth centuries). In 1909:V 71 (upper layers). Palaeographically close are V 61 and V 63, V 62 and V 78. It is possible that V 60 also belongs here.

Necropolis by the Extramural Cruciform Church

This cemetery, in its northern part, borders on the Necropolis of the Quarantine Bay. At its center lies the Extramural Cruciform Church, which was of considerable importance in antiquity. The following inscriptions were found at this site: in 1902 — V 85 (Burial vault 1431, wall graffiti), in 1906 — V 77 (soil heap), V 69 (Burial vault 2031; X–XII centuries C.E. — probably, a secondary burial), V 48 (Burial vault 2031, earth infill; invocation, 1202–1203 C.E. — perhaps moved here by chance (?) from the nearby Extramural Cruciform Church, in 1996 — V 79 (chance find near the church), in 2002 — V 78 (Burial vault of 2002 C.E., wall dipinto), in 2006 — V 82 (Burial vault 1/2006, wall graffiti).

In sum, almost all Early Byzantine epitaphs, with attested findspots in Cherson (eleven or twelve out of nineteen), come from two nearby cemeteries: the one by the Quarantine Bay (six or seven), and another – by the Extramural Criciform Church (five), where wall inscriptions in burial vaults are common. V 61 from the Western Necropolis (?) stands on its own; and so does V 72, found in Church 19, which differs from the local tradition in the layout of text – we may peculate that this tombstone marked the grave of a visiting official.

Curtain Wall 1, moat

Only one inscription comes from this area - V 52 (V–VI centuries C.E.). The slab was covering the entrance to the Burial vault 4.

Mountainous Crimea


In Mountainous Crimea, monuments of the Early Byzantine period are not numerous: V 159 from Inkerman (clearly connected to Cherson), V 171 from Mangup (possibly, a building inscription, concerning a wall) and V 227 of unknown provenance (a dedication, apparently from a basilica). The two latter ones are most certainly not of local provenance: erected on behalf of an emperor and of the wife of a comes. Such sparce representation is not surprising and corresponds to the historical situation as we know it: on the one hand, the Byzantine penetration into Mountainous Crimea was not particularly strong, and on the other, which is also more important, the level of hellenization among the local population was quite low. A good indication of this can be found in the Eastern grotto of Bakla Cliffs, where among hundreds of petroglyphs, which are mostly Christian and depicting a cross, there are none that can with certainty be identified as Greek inscriptions.

The Dark Ages (VIIth - early IXth centuries) are represented only by V 115 and V 119 from Bakla and V 226 of unknown origin (Mangup?). These inscriptions persuasively illustrate the period of Khazar penetration into Crimea: V 226 informs of the building of a church under a khagan and a tudun; V 115 displays side by side the Greek name of the deceased (apparently, that of a nobleman) and the non-Greek name of his wife; V 119 contains fragments of non-Greek words.

In the middle of the IXth and in the Xth century, the number of inscriptions grows, due to the foundation of the Theme of Cherson and the stabilization of political situation. Among these, we may first list the epigraphic monuments of Bakla (V 111, V 113, V 120 and possibly V 121) and Mangup (V 172, V 181–175, 183, and perhaps V 195), as well as V 122 from Basman (connected to Cherson).

The "darkest period" in the history of Mountainous Crimea, XI-XIIth centuries, is represented by V 134 from Gluboky Yar (1034 C.E.) and V 174 from Mangup (1178 C.E.). The existence of these two unfortified sites (Glyboky Yar and Mangup) testifies to the relative political stability at this time.

The majority of inscriptions in this region, however, belong to the Late Byzantine period. It is here (as well as on the South and Southeast Coast of Crimea) that the peninsula's center of Greek culture can be found - after 1204 C.E. and the decline of Cherson. This volatile era is characterized by a short life span of distinctive local epigraphic traditions.

The XIIIth century is marked by the construction of monasteries and churches in Inkerman (V 149; 1272–1273 C.E.), Mangup (V 174, V 196; 1220–1225 C.E.) and Eski-Kermen (V 219)

The political instability of the end of the XIIIth - beginning of the XIVth centuries is reflected in the epigraphic record of the period: V 175, V 176 and V 177 inform of enemy attacks, and of continuous renovations of the city and its walls. From the end of the XIIIth century, we begin to discern a distinctive epigraphic habit of rural settlements of Mountainous Crimea (see below): Golubinka, Kudrino, Goryanka and Vysokoye.

A new blossoming of the Theodorite epigraphy comes in the XVth century, with many stylistic features borrowed from the inscriptions of the Palaiologian period: ornamental ligature script, carving of letters in relief, rich decoration of framing elements. These traits, however, affect only the main centers, such as Mangup (V 178, V 179, V 189 and V 180 (?)), as well as coastal settlements such as Cherson (V 13 and V 14), Inkerman (V 148), Partenit (V 241) and Luchistoye (V 238). At the same time, many monuments display simple styling and decoration. In rural settlements, the number of inscriptions decreases, and examples are known only from Vysokoye (former Ashagy- and Yukhary-Kermenchik) and Goryanka (former Laki).

On the basis of their origin, we divide inscriptions of Mountainous Crimea into two groups: those from valley settlements and those from "cave cities."

Valley settlements

The first group (valley settlements with adjacent cemeteries housing small churches) is represented by V 167, V 168, V 169 from Kudrino (former Shury or Shuryu in Kachinskaya valley). In a side valley that branches off from Kachinskaya, we find V 140, V 141, V 142, V 143, V 144, V 145, V 147 at Goryanka, and the following inscriptions - V 123, V 125, V 126, V 127, V 128, V 129, V 130, V 131, V 132, V 133 - at Vysokoye. In Belbek valley, at Golubinka (former Fot-Sala), another group is attested: V 135, V 136, V 137, V 138, V 139. Outside of these valleys, additional epigraphic monuments were found: in the valley Maryam-Dere (former Mariampol) - V 197; at Krasny Mak settlement (former Biyuk-Karalez) - V 123 (possibly originating from Eski-Kermen); finally, V 110 comes from the area of Ay-Dimitry (the closest modern settlment is the village of Polyana).

We can illustrate a specific local pattern by analyzing the inscriptions from Vysokoye. Two inscriptions (V 127, V 128) were found at the same cemetery and date to the same period (1347-1348 C.E.). In addition, they also contain the same rare name Kalanitza. V 133 (also dating to 1347 C.E.) most likely also comes from this area. Two other inscriptions from the Vysokoye settlement contain a unique dating formula ἐπὶ ἔτους and are close in date (1361–1362 C.E.). V 125 also belongs to the same group, but there is an inscription of another type (V 129) found in the same church of Stt. Cosmas and Damian, and its date (1382–1383 C.E.) and formula are similar to V 131, which comes from a cemetery belonging to this church. At the same time, these two texts (V 129 and V 131) are palaeographically different, while the image and script of V 129 closely relate to V 132, also dated to 1381–1382 and to V 124 also possibly from Vysokoye, dated 1387. The already mentioned dating formula ἐπὶ ἔτους is repeated in V 123 (1440 C.E.). V 126 from the church of Stt. Forty Martyrs (much later in date - 1488) stands on its own. Thus, we see that in different funerary complexes of Vysokoye, we find the works of several letter-carvers belonging to consecutive generations: 1347-1348, 1361-1362, 1381-1382 (two masters), 1400 and 1448 C.E.

We find a similar situation in Goryanka. Here, two cemeteries are known: one belongs to the Church of Holy Trinity, and the second is in the forest. At each site, inscriptions of two periods are attested: 1301–1310 (V 141, V 142) and 1362–1364 C.E. (V 143, V 144). At the first one, there is also an inscription (V 140) dating to 1421, and a related one (V 135) dating to 1413 C.E. The inscriptions of the 1360s feature details of the life and death of the deceased. All early inscriptions in Goryanka are dated by month and year; in the inscriptions from the first cemetery, we find, as an additional feature, a rare form of dating, according to indiction.

By contrast, all epitaphs (except for the fragmented V 139), from Kilse-Bair near Golubinka, are more or less the same: the inscription is placed inside a niche, and the month is absent from the dating formula.

What we observe is that the majority of "rural" inscriptions is concentrated in neighbouring valleys: Kacha, Laki, and Maryam-Dere. They are all funerary, with the formula Ἐκοιμήθη, with the exception of graffito V 147 on the wall of the church in Laki, which might still be a short epitaph.

"Cave cities"

The conventional name, "cave cities", should not mislead us: some inscriptions indeed were carved on the walls of caves, but others were lapidary, and still others were inscribed on frescoes. Cave cities can be divided into three groups: (1) urban and quasi-urban centers, (2) fortresses, and (3) cave monasteries and churches. To the first group belong: V 111, V 112, V 113, V 114, V 115, V 116, V 117, V 118, V 119, V 120, V 121 from Bakla; V 160, V 161, V 162, V 163, V 164, V 165 from Kachi-Kalyon and from the adjacent plateaus Fytska and Tash-Air; V 171, V 172, V 173, V 174, V 175, V 176, V 177, V 178, V 179, V 181, V 183, V 184, V 185, V 187, V 188, V 189, V 190, V 191, V 193, V 194, V 195, V 196 from Mangup and its foothills; V 203, V 204, V 205, V 206, V 207, V 208 from Tepe-Kermen; V 217 from Chufut-Kale (two unintelligible graffiti should probably be attributed to the non-Greek population of the city) and V 219, V 220, V 221, V 222, V 224, V 225 from Eski-Kermen. To the second group belong: V 199 from the fortress Sandyk-Kaya and V 200, V 201, V 202 from Syuren fortress. Finally, from cave monasteries and churches originate the following: V 122 from Basman caves, V 134 from Gluboky Yar, V 170 — from Danilcha-Koba, V 209, V 210, V 211, V 212 from the Church of Donators in Cherkes-Kermen, V 213, V 214, V 215 from Chilter-Marmara and V 218 from Shuldan. Inkerman (V 149, V 150, V 151, V 152, V 153, V 159, V 154, V 155, V 156, V 157) stands apart: most of the inscriptions come from St. George Monastery (now St. Clement Monastery) and various churches, while V 153 comes from the burial ground near Sakharnaya Golovka. The origin of some inscriptions (V 226, V 180, V 227, V 124, V 146, V 229) is unknown.

The inscriptions of cave cities and fortresses are mainly located, just as cave cities themselves, along the border between the plains and mountainous parts of Southwest Crimea, following the line: Inkerman — Syuren fortress — Mangup — Shuldan — Chilter-Marmara — Eski-Kermen/Cherkes-Kermen — Kachi-Kalyon/Tepe-Kermen — Chufut-Kale — Gluboky Yar — Bakla. Here, while funerary inscriptions are attested, they do not predominate: there are many building, invocative and demonstrative inscriptions. In the interior of Mountainous Crimea, isolated inscriptions were found inside cave churches of Basman and Danilcha-Koba, as well as in small churches located not far from Danilcha-Koba, in Ay-Dimitry and fortress Sandyk-Kaya.

South and Southeast Coast of Crimea


This region, as well as the Dniester estuary, lack Early Byzantine inscriptions altogether, although Early Byzantine fortifications were present both in Alushta and Gurzuf.

The Middle Byzantine period is represented by V 243 from Partenit (906 C.E.) and V 240 from the neighboring Panair, V 258, V 250, V 253 and V 241 from Sudak, and an isolated V 237 from Livadia. Such distribution corresponds to two contemporary Byzantine centers in the region: political - in Sudak (cf. V 11) and monastic (with a possible bishop's residence) - in Partenit.

The blossoming of regional epigraphy begins in XIII-XIVth centuries. In the west, we see Theodorite monuments of the XVth century: V 238 from Luchistoye, V 239 from Massandra and V 241 from Partenit, while Sudak is found to be under Genoese influence up until the appearance of dating formulae "Since the Birth of Christ" (V 258). Still, even here we find dipinti on frescoes made in the ligature style and closely related to the Theodorite examples (V 245, V 246). The majority of inscriptions from Foros, Livadia, Ay-Vasil, Yalta, Partenit, Alushta, and Theodosia do not betray chracteristics of any distinct tradition, apart from V 234 perhaps, which somewhat resembles V 318 from Kerch.

Geographically, this region consists of a narrow strip of coastal land, stretching from Laspi in the west to Theodosia in the east: inscriptions from the cities and monasteries are located here. Yet several monuments (V 251, V 256, V 262 and V 263) were also found in the mountainous areas bordering this coastal strip in the north. Due to its geography, the region does not constitute an organic whole, but rather consists of several isolated centers.


The earliest to take shape are the epigraphic traditions of Partenit and Sudak. In the former, it is connected to the monastery of the Holy Apostles, founded at the end of the VIIIth century by Saint John of Gotthia and rededicated, after a period of decline, to Stt. Peter and Paul. This chronology is mirrored in the epigraphic record: V 243 (906 C.E.) and V 241, V 242, V 244 (XVth century). V 240 (X century) was found at the nearby monastery at Panair.


In Sudak (Sougdaia, Soldaia), Byzantine inscriptions date to the Middle (V 249, V 250, V 253 and V 254) and especially Late periods. During the latter, Sudak experienced a degree of Genoese influence (see, e.g., V 258, V 252). Here (and only here), inscriptions of different categories are represented quite evenly. We may also attribute to Sudak V 262, which was found at Kordon-Oba near Shchebetovka.

Laspi and Foros

The inscriptions of these two centers are so heterogeneous that it is impossible to speak of any epigraphic habit.

Yalta region

This region is represented by singular examples that have little in common with one another and come from a variety of places: Simeiz (X–XIIth centuries), the mountain range of Iograph; Ay-Vasil, and Upper Massandra (the last two - Late Byzantine).


V 238, found here, in fact belongs with the Theodorite epigraphy of the XVth century (see above).


Both inscriptions from this city date back to the XIVth -early XVth century, and slightly resemble one another in paleographic features, however, it is not yet possible to assert the presence of a local epigraphic habit.


Surprisingly, this settlement, which after a decline in the Early and Middle Byzantine periods, had grown into a large center by the XIV-XVth centuries, has yielded only one Greek inscription - a funerary monument of 1378 C.E., with non-Greek names (V 260). Contrary to common opinion, V 315 (dated 819 C.E., see commentary) is not from here.



Almost all Byzantine inscriptions of Kerch (ancient Pantikapaion, Byzantine Bosporos) are tombstones of the IV-VIth centuries. The dating of Early Byzantine tombstones is a difficult and poorly researched subject. Only three monuments are dated: V 295 (437 C.E.), V 304 (497–498 C.E.), V 305 (491–492 C.E.); V 306 dates to the late Vth century. All dated funerary inscriptions belong to the late period of the Bosporan Kingdom.

Two related groups of epitaphs can be discerned at Glinishche: V 267, V 280, V 281 and V 270, V 278, V 285, V 287. Inscriptions in the first group were found together in 1897, near the Jewish tombstone of Simon (Shkorpil 1898, 210, No. 20), a factor which speaks in favour of an early date, since at that time, in the religiously tolerant environment of Bosporus, Christians and Jews were able to share the same burial ground, that is, the date might be the IVth, or at the latest, Vth century. In any case, after the annexation of the region to the Byzantine empire in the first quarter of the VIth century, such sharing would have been impossible. Among the inscriptions of the second group, we find a funerary monument of Arsakis, a "Christian" (V 270) - such specification of one's religious affiliation is not attested after the IVth century (see commentary). We may speculate, therefore, that the two largest concentrations of tombstones in Glinishche, the garden of Woerle and the yard of Bondarenko, indicate the site of a Christian cemetery. The observations outlined above prompt a question: do we know of any Bosporan tombstones dated in the VI-VIIth centuries? The exceptions are V 314 (VIth cent.) and V 307 (691-692 C.E.), which differ in typology and palaeography from the examples described above. The virtual disappearance of tombstones might be related to the end of the Bosporan epigraphic habit (cf. strikingly alien examples of the VI-VIIth centuries: V 314, V 307, V 329, V 330). What is unclear is whether we should envision a sharp termination or a gradual decline of this tradition: it is hard to imagine that the Bosporans would have suddenly stopped carving epitaphs for their dead, and yet even this possibility cannot be ruled out with a view to the dramatic events and the renewal of population in the Bosporus at the start of the VIth century (see further details below).

All Middle Byzantine (V 297, V 305 and V 316) and one (V 318) of the two Late Byzantine inscriptions were found in the area of Predtechenskaya Square (see below). The findspots reflect the development of Kerch in the period after the VIth century (Ponomarev 1999). We should note that one of peculiarities of the Middle Byzantine epigraphy in Kerch is the use of the dating formula "since Adam" (see IV.4.C).

We may distinguish several zones with detectable concentrations of Christian inscriptions in the territory of Kerch (see in more detail below): 1) Predtechenskaya Square, which shows signs of public use in the Early Byzantine period, and of additional use as burial ground, in the Middle and Late Byzantines periods ; 2) The North slope of Mithridates Hill, where we find burials of the Bosporan nobility; and outside of the city: 3) Glinishche - possibly, the most important Christian neсropolis, and 4) the end of Karantinnaya Street - with isolated burials. Isolated finds of inscriptions were made on the South slope of Mithridates Hill, in the Tsarsky tumulus, and on the Theodosia Road. A funerary monument from the top of Mithridates Hill is most likely connected with the necropolis (see III.1.E.c). What is notable is that almost all Byzantine inscriptions are concentrated in the center of Kerch, and only a few - in the immediate environs, while they are virtually absent (perhaps with the exception of V 323) from the territory of the European Bosporus (Late Byzantine Theodosia is not part of it).

Although it is difficult to draw reliable conclusions about the ethnic composition of the Early Byzantine burials, we should note that on the North slope of Mithridates Hill and at the end of Karantinnaya Street, the number of Greek and non-Greek names is about the same, that is, 5 and 3, and 1 and 1, while in Glinishche, Greek names predominate: 22 out of 24, and it must be significant that two individuals with non-Greek names, Toudrougos and Tiginagas (V 293, V 300) were interred away from the main group of burials. Regretfully, we also do not know of the exact provenience of the epitaphs that specify the status of the deceased: an infant Nikianos (?) (V 279) and deacon Eusebios (V 295; 437 C.E.), and the same is true of V 313, V 271, V 288, V 319, V 297, V 299, V 307 (692–693 C.E.), V 311, V 304 (497-498 C.E.). The furrier Philoxenos was burried in Glinishche. Of the four Bosporans whose patronymics are recorded (V 267, V 283, V 286, V 302), as was traditional for Bosporus in antiquity, three were buried in Glinishche, but the presence of a patronymic in the epitaph of the furrier Philoxenos indicates that its use was not an indicator of high social status (as we shall see below, IV.2.A.b., all inscriptions with partronymics date to the IVth century). Finally, three out of seven funerary inscriptions on the North slope of Mithridates Hill were made on the walls of family crypts, some of which were richly decorated. One of those buried here bore the good old Bosporan name Pappos. Another inscription appears on a panel in front of the entrance to a crypt. Rich decoration is also present on an isolated funerary monument bearing a characteristic Bosporan name - Phannes. By contrast, the funerary monuments in Glinishche are simple tombstones. On the basis of these observations, we can suggest that the North Slope of Mithridates Hill was reserved for elite burials, including the descendants of the old Iranian-speaking nobility, while Glinishche was used for the burials of Greeks (and hellenized barbarians), as well as craftsmen.

Predtechenskaya (St. John the Precursor) Square

In Predtechenskaya (now Lenin) Square, in the city center, the following Early Byzantine inscriptions were found: V 265 (479–492 C.E.) - on the corner of Vorontsov (now Lenin) Street, apparently not far from the site of the tower mentioned in the text; V 314 - on the site of stone-built shops not far from the Basilica of the V-VIth centuries C.E., on a site not far from the present-day church of St. John the Precursor, from where it probably originates. In addition, here (on the former site of a small park), two inscriptions, V 304 (VIII–IXth cent. C.E.) and V 318 (XIII–XVth cent. C.E.) were found, indicating a burial plot of the Late Byzantine period. Two inscriptions, V 316 (767 C.E.) and 289, come from the church of St. John the Precursor, but they are carved on the church columns, which apparently originate from the Early Byzantine church - possibly, a nearby basilica (if so, then the basilica existed at least before 767 C.E.). Therefore, the dates of inscriptions found in Predtechenskaya Square (Vth, VIII–IXth, XIII–XVth cent. C.E.) testify to the continuity of development in this area over the Byzantine period: in the Early Byzantine period, they indiciate building activity (tower, church), and in later periods, use as a burial site.

Mithridates Hill

Only one inscription ( V 277) comes from Mithridates Hill proper. It was found at the top of the hill, near the monument of Stempkovsky. This is an Early Byzantine epitaph, and probably originates from an unexcavated Early Byzantine cemetery in the area.

North slope of Mithridates Hill

On the north slope of Mithridates Hill, we find a scatter of several Early Byzantine tombstones: V 268 — in the area of the Pugachevsky Lanes, V 305 (491–492 C.E.), V 306 (end of V cent. C.E.) - nearby, on Gospitalnaya Street, V 310 - in the same location, but slightly uphill. V 298 was found a bit further away, on 2nd Nagorny Lane. Also, up the slope - what appears to be a child's tombstone - V 275. Likewise up the slope and eastwards, we find V 272 (323–324 C.E.) on First Esplanadnaya Street. On the property of Zavarzin we find a chamber tomb with inscription V 266. Altogether, the evidence allows us to hypothesize a Christian cemetery of the IV(?)-Vth century C.E. in the area of Gospitalnaya Street, where distinguished Bosporans were buried, among them - comes Sauagos.

South slope of Mithridates Hill

Only one tombstone (V 320 (1375 C.E.), and that one Late Byzantine in date, originates from this area, specifically from Bosforskaya (now Sverdlov) Street. It is possible that the person named in the inscription, someone Kyra, was buried near a church located in the vicinity.


In this part of the city we find the highest concentration of Early Byzantine tombstones. Here, outside the walls of Pantikapaion/Bosporos, in a Christian cemetery located in the area of former Bratskaya (now Frunze) Street, the following inscriptions were found: V 293 — in the yard of house No. 13 (then the property of F. Collangelo), V 270 (III-IVth cent. C.E.), V 278, V 285 and V 287, all found together in 1903 in the yard of house No. 17 (then the property of the widow А. Bondarenko), V 284, V 290 and V 302 (the last two found together) — in the garden of house No. 19 (then the property of I. G. Chernyavsky). The greatest number of tombstones were found nearby, in the garden of K. Woerle: they are V 267, V 280, V 281 (found together in 1897, next to the Jewish tombstone of Simon), V 273, V 291 (found together in 1898), and V 269. V 294 was found in the vicinity, and V 274 also nearby, in the garden of Francesca. V 286 and V 292 were discovered in the garden of Poltavsky, while V 289 was found on the property of Iokhannyuk. V 276 and V 282 also orginate in Glinishche, but their precise findspots are unknown. V 300 was found at some distance from the main concentration of tombstones, near the Aleksandrovsky College (in the area of Petr Korolev Street).

End of Karantinnaya Street

The far end of Karantinnaya (now Kirov) Street, that is, its extension, house No. 11 (the former property of Serganidi), near the former almshouse of Zolotarev (now Komsomolsky Park) is the findspot of V 283; while V 296 was found behind the Arestantskie barracks (now the grounds of the canning factory Proliv (41 Kirov Street)), on the former property of I. Demidov. These two inscriptions are totally different in character: the first is the lengthiest Christian epitaph from Bosporus, while the second bears only the name of a deceased woman. Nonetheless, it appears that here, and likewise at Glinishche, burials took place outside the city walls.

Tsarsky (Royal) tumulus

A Christian graffito with the name Kosmas (V 309) was scratched on the vault of the dromos of the Tsarsky tumulus, which might have served as a church.

Theodosia Road

Inside a chamber tomb on the former property of Grigoriev, located on the Theodosia Road, we find graffito V 308 with the names of Phannes and Eutyches (the third name did not survive). It is not clear whether they belonged to those who were buried in the tomb.

Taman Peninsula

This region is represented by inscriptions of two periods: those belonging to the Late Bosporan kingdom, such as V 331, V 342 (both dated in the 2nd half of the Vth cent. C.E.) and V 332; and those from the time of the direct Byzantine control of the area, that is, V 329 и 309 (both VIth cent. C.E.)

V 336, V 339, V 343 are dated in the Middle Byzantine period, prior to 1000 C.E.; and V 340 dates a little later (1078 C.E.). The inscriptions most likely testify to the functioning of the Tamatarkhan eparchy on the peninsula.

The same number of inscriptions belongs to the Late Byzantine period (in contrast to Kerch where we find only one such inscription): these are V 324, V 334, V 335, V 338.

Geographically, most inscriptions originate from the main ecclestiastical, administrative, and cultural center of this region - Early Byzantine Hermonassa - Medieval Tamatarcha (modern Taman). Two inscriptions come from Phanagoreia (modern Sennoye), and also from nearby - a dubium, V 328. One inscription originates in Anapa (that is, more precisely, in Late Byzantine Napa rather than Late Bosporan Gorgippia). Single inscriptions come from the Utash settlement and the vicinity of Temryuk.

IV. Classification of Inscriptions

IV.1. Material

Unfortunately, the author did not have the opportunity to benefit from the expert advice of a professional geologist in the matter of identifying the lapidary material. For this reason the survey of material is only cursory (although admittedly a more precise identification would have been very helpful with respect to determining the origin of stone from specific locations). Rather our attention will be focused on the reasons for the choice of this or that material in each case.

The Northern Coast of the Black Sea is not rich in marble; here, limestone predominates. The number of inscriptions on marble is therefore significantly less (73 in all) than on other types of stone. In the Early Byzantine period, marble was mainly imported from Prokonnesos in the shape of architectural members, which were later reused for inscriptions (17 cases identified in Biernacki 2009). In the later periods, reuse of ancient and Early Byzantine spolia, including reliefs (e.g., V 50, V 101) is common. Spolia were used either without any alteration, or after the erasure of the original inscription. There is only one known case where a later inscription is tagged onto the original one (V 339). Thin marble revetment was readily used as material for inscriptions in the Middle Byzantine period (e.g., V 47). Cherson was the main centre of inscriptions on marble. In the Late Byzantine period marble is very rare, and such inscriptions are both exceptional and signs of the highest prestige (see, e.g., V 177, V 178).

Due to its ready and wide-spread availability in the area, limestone of every type is a material of choice in the region of the Northern Black Sea Coast in the Byzantine period. It was used for all categories of inscriptions.

By contrast with limestone, sandstone is very rare. Only 6 inscriptions are attested on this material: 3 from Cherson, 2 from Sudak and 1 from Alushta. Just as limestone, sandstone typically comes from local quarries.

A special characteristic of the Northern Black Sea Coast, and more specifically, of Mountainous Crimea, is a large number of rupestral inscriptions (28 out of 107): mainly they appear on the walls of limestone caves, and primarily, of cave churches. These inscriptions are mostly funerary, but there are also some building (V 160, V 171, V 203), demonstrative (V 111, V 221), invocative (V 213, V 214), and commemorative (V 165, V 196), as well as whole clusters of inscriptions such as, e.g., V 134. By and large, these inscriptions are graffiti.

Graffiti are known from all regions of the Northern Black Sea Coast, except for Dniester Estuary. Naturally, these inscriptions for the most part are of private character: invocative and commemorative.

Less common are painted inscriptions in red ochre on plaster. These are known from four regions: chamber tombs of Cherson (V 65) and Kerch ((V 305 and V 306); wall paintings of Late Byzantine Mountainous Crimea (these are building inscriptions V 209 and V 219, dedicatory (V 149), commemorative-demonstrative (V 200) and tombstones of ktetors (V 209); the latter may have been more numerous, just like frescoes of the Mountainous Crimea in general), and on the frescoes from the South and Southeast Coast of Crimea (V 239 from Massandra, and V 245 and V 246 from Sudak). Dipinti on stone are virtually unknown (with the exception V 188 and the rupestral V 215), however, Early Byzantine inscriptions of Cherson and Bosporus were often traced in red ochre.

IV.2. Palaeography

The subject of palaeography is one of the most difficult and least studied in Byzantine epigraphy, in comparison with ancient Greek and Roman fields: it was only recently that the first comprehensive studies began to appear, although such studies tended to focus on specific regions of the Christian world (Morss 2003). The situation with respect to the Northern Black Sea Coast is even more problematic: here, inscriptions were dated either visually, relying on one's intuition (as Latyshev did), or on the basis of randomly selected analogies (as Solomonik did). It is our view that the time has come to identify certain principles of palaeography for the region, relying first and foremost on securely dated inscriptions.

We should mention at the outset that for two reasons we do not address the palaeography of graffiti. Firstly, starting from the Middle Byzantine period (earlier graffiti are virtually unattested in the Northern Black Sea), they are heavily influenced by the uncial cursive, and minuscule, in particular, the characteristics that tell little about the local epigraphic habit. Secondly, graffiti often betray a letter-carver's inability to work with lapidary scripts, so that instead he writes "as well as he can."

IV.2.A. Early Byzantine Palaeography

In the Early Byzantine period we cannot identify a single "epigraphic school" in the Northern Black Sea region, rather two local traditions can be distinguished: that of Cherson and that of Bosporus.

IV.2.A.a. Cherson

One of the main problems for the study of Early Byzantine palaeography of Cherson is a small number of securely dated inscriptions: V 5, V 6, V 7, V 80, although we may add V 171 from Mangup. In addition, these inscriptions predominantly refer to the building activities undertaken by the initiative of central authorities, and therefore cannot be always expected to reflect the local tradition (although V 6 was erected by a body of Chersonites). Still, we have no other basis for the study of local palaeography. It is also worth observing that only V 80 can be seen as an extension of the Late Antique epigraphic tradition, while the next (in date) inscription, V 5, is completely alien to it.

The main change in the Chersonian palaeography of the IV-VIIth centuries has to do with the transformation of the crossbar in alpha: in 350-355 C.E., it is broken, in 392-488, it is horizontal and slightly leaning to the left, while in 533-574, it is broken again. Alpha with a broken crossbar dominates in the Justinian period (exceptions are rare, see IdC 10, dated to 536 C.E.). It is much more difficult to determine whether a straight/slanting crossbar really replaces the broken crossbar at the end of the IVth century. Among the Early Byzantine inscriptions of Cherson there is not a single one (with the exception of V 20, which is explained by the mosaic nature of the inscription) where the two forms co-exist. This circumstance supports the notion of two separate epigraphic habits. All inscriptions on plaster in the chamber tombs of the Extramural Cruciform Church, which date to the IV-Vth centuries (V 65, V 82, V 85), use alpha with a slanting crossbar, or as its cursive variant, alpha with a loop. The same forms of alpha are found in V 23, which belongs to Church 36 built at the end of the IVth century C.E. In addition, inscriptions with this form of alpha virtually never use rectangular epsilon and sigma (V 63 is an exception). Therefore, it is quite possible that the change in the shape of alpha indeed occurred at the end of the IVth century.

If we adopt this observation as a dating principle, we should be able to date many previously undated Early Byzantine inscriptions of Cherson. The group of inscriptions with slanting crossbar (a straight one is probably an exception: it is attested only in V 6, V 10) will then include, in addition to those mentioned earlier, the following texts: V 26, V 52, V 61, V 64, V 63, V 71, V 81 and V 101. Notably, tombstones predominate in this group.

The other side of the issue is how to determine whether the alpha's broken crossbar indicates that the date is the IVth or VI-VIIth centuries. On the basis of archaeological and architectural criteria, we can suggest that V 56, V 57, V 58, V 76, as well as V 44, and on the basis of palaeography (Late Roman beta) also V 18, belong to the first group. The second chronological group, which is also chracterized by the use of serifs, is much more numerous: here we find dedications V 20, V 24, V 28, V 32, V 33, and most likely V 23 and V 25, tombstones V 60, V 62, V 72 (although probably not part of the Chersonian tradition: see commentaries), V 71, V 74, V 75, V 78, V 79, V 77, as well as V 40, V 43, V 104 and possibly V 105. We should probably also place in this group V 10, V 27, as well as V 227, originating, to judge by all signs, from Mountainous Crimea and found in secondary use in Bakhchisarai. Noteworthy is the fact that alpha with broken crossbar survives in Cherson even through the Dark Ages and still occurs at the beginning of the Xth century (V 79).

V 49 presents rectangular letterforms, not typical for Cherson, and probably indicating a Bosporan influence, especially if we take into consideration the typically Bosporan name of the dedicant - Trophimos (see commentary).

IV.2.A.b. Bosporus

At Bosporus, the situation is similar to that in Cherson. Very few inscriptions are dated: V 272 (323–324 C.E.), V 295 (C.E.), V 265 (479–493 C.E.), V 305 (491–492 C.E.), V 304 (497–498 C.E.); V 306 is dated to the end of the Vth century C.E. There is a similar division between inscriptions where alpha has a broken crossbar, and those with a slanted crossbar (the only exception, where both forms are present is V 290). The first category is not as numerous as at Bosporus, however: V 267, V 272, and V 289. Characteristically, in all three the letters that are typically rounded are rectangular. V 272 probably dates to 323-324 C.E.; V 267, judging by the archaeological context, dates to the end of the IVth century; in V 289, alpha with broken crossbar is a later addition. All of these observations lead us to conclude that, just as in Cherson, the form of alpha with slanting crossbar is the earlier of the two. The appearance of alpha with a loop alongside that with broken crossbar, in V 305 and V 306, must be due to the cursive style of this inscription, made at the time when generally alpha with broken crossbar predominates.

The number of inscriptions where alpha with broken crossbar is used is much greater, consisting of 39 examples. All such inscriptions date to the Vth century C.E. As we have discussed above (III.1.E), in Kerch, we can distinguish two groups of funerary monuments dating to the IVth century C.E.: V 267, V 280, V 281 (garden of Woerle) and V 270, V 278, V 285, V 287 (estate of Bondarenko), including, in the first group, one inscription where alpha has a slanting crossbar. There is also a difference between the two groups: in the second one we find a mix of rounded and rectangular letter shapes, just as in V 284, V 290 and V 302 from the nearby garden of Chernyavsky. At the same time, two inscriptions, V 273 and V 291, although found in the same place as the first group, in contrast to the latter, display almost exclusively rounded shapes (except for a single rectangular epsilon in each), similarly to the second group. V 269, however, shows only rectangular shapes. Thus, two more types of inscriptions are distinguishable: those with exclusively rectangular lettershapes, and those with a mix of rounded and rectangular shapes. In addition to already mentioned V 269, V 271, V 280, V 281, the first type is represented by V 268, V 271, V 274, V 275, V 276, V 282, V 292, V 294, V 298, V 300, and what is particularly important, V 295, which is dated to 437 C.E. Within this type, we may point out a sub-type characterized by a delta with projecting right hasta: V 269, V 271, V 274, V 275 and V 292. Within the mixed type the use of such a delta is found in V 270 and V 283. The second (mixed) type is more localized: in addition to the already mentioned V 270, V 278, V 283, V 284, V 285, V 286, V 287, V 290 and V 302, we may place here the neighboring V 291, as well as V 273 (IVth cent.), V 279 and V 297 of unknown provenance - we can surmise that all these inscriptions date to the IVth century. An intermediate type represented by diamond-shaped theta and omicron, already known in Bosporus by the Roman period, is here seen in V 277 and V 283.

The development of Bosporan palaeography in the IV-Vth centuries can be summarized as follows. At the start, in the IVth century, we witness a co-existence of two writing styles, each of which uses the epigraphic script of official inscriptions in its own way: one uses alpha with slanting crossbar and completely rectangular lettershapes, while the other employs rounded lettershapes and alpha with broken crossbar. The second stage of development, some time in the IVth century (as can be seen in a group of inscriptions frome the garden of Woerle - see above), witnesses the birth of a new epigraphic koine, which is a combination of two earlier styles: rectangular lettershapes and alpha with a broken crossbar. Concomitantly with this hybrid style, in the second half of the Vth century, a cursive style is attested (V 305 and V 306), as well as the epigraphic habit of official inscriptions preserving rounded forms (V 265 and V 304, both on marble). V 331 (shortly before 491 C.E.) and V 342 (478-479 C.E.) are related to this habit.

As has been stated above (III.1.E), we have no proof for the existence of any VIth-century funerary monuments in Kerch, except for a single example (V 314). At the same time, we cannot exclude the possibility that the style of the previous century continued in the next (cf. V 275).

By contrast, the VIth century is well represented on the other - Asian - side of Bosporus. There we have building inscriptions V 215 (548? C.E.) and V 330 (589-590 C.E.), as well as a tombstone V 339.

IV.2.B. Middle Byzantine Palaeography

Despite the differences, the Early Byzantine palaeography, being a derivative of the Late Antique, is much more homogenous than the Middle Byzantine in the Northern Black Sea region. Nevertheless, it is possible to identify in the latter several styles that are evident across regional borders.

While at the end of the VIIth century, both in Cherson and at Bosporus, a monumental Early Byzantine script was apparently still in use, as can be seen in V 24 (674–675? C.E.) and V 307 (692–693 C.E.), the inscriptions of the "Dark Age" demonstrate something quite different. Serifs disappear, letters become elongated, some lettershapes disappear (e.g., rectangular epsilon, Y-shaped upsilon), while the vertical extension of diagonal hastae develops, and the old alpha with broken crossbar continues in use. There are only a few examples from this period, but such ones as V 16 (and possibly V 15) from Cherson, V 115 and V 119 from Bakla, V 226 from Mountainous Crimea, and V 254 from Sudak demonstrate a clear influence of barbarization, not only in onomastics, but also in the manner of inscribing. The only dated ("non-barbarian," except for the deceased's grandfather's name) example is V 316 (767 C.E.). V 321 is close to it in date.

V 315 (819 C.E.), which originates from the church of Holy Apostles in Kerch, already shows clear signs of a new epigraphic style (borrowed apparently from the south, that is, Asia Minor): letters become narrower and elongated, with pointy instead of rounded loops, special shapes develop for some letters (primarily sigma), and in addition, we witness the return of alpha with slanting crossbar and with a loop. This particular style subsequently develops in Bosporus and is then exported from there to V 322 (884 C.E.) of unknown provenance, V 243 (905 C.E.) from Partenit, and V 331 from Kuban. In its simplified form it can be seen in V 240 from Partenit.

In Cherson, at this time (IX-Xth cent.), a new style appears, possibly under the influence of Bosporus. This style is distinguished by the use of the same lettershapes: first and foremost, beta (and ksi) with bottom underline, but with less pointed loops: V 36, V 46, V 47, V 50, V 66 (the last occurrence of an alpha with a broken crossbar), V 91, V 98 (975–976 C.E.). Beyond Cherson, the same style is attested in Mountainous Crimea: V 115 from Bakla, V 134 from Gluboky Yar and V 184 from Mangup.

The same style also exists in a simplified form, devoid of particularly characteristic lettershapes, but marked by a general preference for rounded letters and an abundance of ligatures and abbreviations. Examples of this style are V 30, V 45, V 87, V 93, V 97, V 104 from Cherson, V 172 and V 195 from Mangup, V 237 from Livadia, V 343 from Taman Peninsula. The lettering of this style can be quite ornate, as can be seen in the Chersonian funerary inscription V 67 (with a rare mixing of the two forms of alpha, cf. V 15), as well as in V 89, and, to some extent, in V 122 from Basman.

A new style, common to Cherson and Bosporus, develops in the XIth century, under the influence of Constantinople, and is first attested in V 11 (1059 C.E.). Its main charectiristics are extremely narrow and highly ornate letters of different heights, an abundance of ligatures and abbreviations, as well as embelished ends of hastae, which replaced former serifs. Expresssed to a different degree, this style can be seen in V 12, V 17, V 109 and V 68 (1183 C.E.), V 248 from Sudak, V 319 (1065 C.E.) from Kerch, and V 340 (1078 C.E.) from Taman Peninsula.

Incursions of the minuscule script (found in manuscripts) into the epigraphic habit (as in V 55 and V 194) is a separate phenomenon, not related to regional styles.

IV.2.C. Late Byzantine Palaeography

The palaeography of the XIII-XVth centuries is even less homogenous, represented by several local short-lived schools.

IV.2.C.a. Coast

The coast of Crimea from Laspi to Kerch is a good exampe of the observation made above: we would be hard pressed to find here even two inscriptions that are alike. Even two inscriptions from Alushta (V 233 and V 234) have little in common. If we add to them V 260 from Theodosia and V 320 from Kerch, we would find some similar features among the four inscriptions (cf. the shapes of mu and alpha), but no common style shared by any two of them.

The only identifiable style is found on frescoes: image-related inscriptions with elongated and slighly ornate letters, represented by V 245 and V 246 from Sudak, while V 239 from Massandra, and possibly V 232 from Ay-Vasil, probably belong here as well.

V 238 from Luchistoye, as well as V 241 and V 244 from Partenit, are akin to the Theodorite tradition, while V 235 is closer to that of Mountainous Crimea (see below).

IV.2.C.b. Theodoro

The most distinctive epigraphic style of the Late Byzantine period is undoubtedly that which developed in the kingdom of Theodoro in the XVth century. This style developed under the influence of the Palaiologian epigraphy and is chracterized by thin elongated and highly ornate lattershapes cut in low relief, and also by rich ornamentation. Examples of this style are: V 178 (1403 C.E.), V 179 (1425 C.E.), V 189 (1456 C.E.) and V 187 from Mangup-Theodoro, V 180 of unknown provenance, V 241 (1427 C.E.; although this one might not be of Theodorite origin, see commentary to the inscription) and V 230 from Partenit, V 13 and probably V 14 from Cherson, V 148 and V 238 (1459 C.E.) from Phouna. This style existed from the start of the XVth century down to the fall of Theodoro in 1475. Within it we can distinguish an earlier and a later version: the former represented by many examples, the latter by a few (V 189, V 238), wherein lettershapes become even more elongated. A somewhat simplified expression of this style is V 190 from Mangup.

Predecessors of this style can also be found in Crimea: for instance, V 207 from Cherkes-Kermen (2nd half of the XIVth cent.). In turn, the precursor of the latter is the image-related style found on frescoes of the XIIIth century, represented by V 149 (1272-1273 C.E.) from Inkerman and V 219 from Eski-Kermen (middle of the XIIIth cent.). At Mangup itself, building inscriptions of the XIII-XIVth centuries (V 174, V 175, V 176, V 177) show no unity of palaeography, which could perhaps be explained by the frequent change of rulers in Theodoro.

IV.2.C.c. Palaeography of Mountainous Crimea outside Theodoro

The lack of palaeographic unity characterizes inscriptions of the rest of Mountainous Crimea. Here it might be more appropriate to distinguish styles of individual villages: Golubinka (all inscriptions except V 136, but including V 121 from Bakla); Goryanka (except V 141, which is closer to V 136 from the neighboring Golubinka), Vysokoye (all inscriptions whose images survive date to within two decades). Even less palaeographic unity is found among the inscriptions of cave-cities, although we can identify some individual styles: the minuscule style, which is represented by V 161 from Kachi-Kalyon, V 175 (1988-1289 C.E.) from Mangup, V 204 and V 207 from Tepe-Kermen and V 213 from Chilter-Marmara; another style can be seen in V 160 and V 162 from Kachi-Kalyon. Some sort of an epigraphic koine can be discerned by the end of the XIVth century in V 200 from Syuren fortress, V 209, V 210, V 211, V 212 from Cherkes-Kermen, and partly in V 177 from Mangup.

IV.3. Types of Inscriptions and Epigraphic Formulae

Overall, among Byzantine inscriptions, categories of text (corresponding to their functions) differ significantly from those known in classical antiquity; at the same time, there is no established typology in Byzantine epigraphy. Here we propose a typology that reflects the specifics of the Northern Black Sea region. Among such specific traits are: a complete absence of state inscriptions of any kind (either polis-level or imperial) and an increase in the number of invocations (starting from the Middle Byzantine period) and graffiti. In addition, new types of inscriptions develop: primarily, we have in mind inscriptions that employ short fixed phrases. Some of these phrases are apotropaic (e.g., "Flee, envy: Christ chases you away"), while others are liturgical formulae otherwise unattested in epigraphy and their function in inscriptions is unknown (except V 194); finally, some liturgical and other formulae ("The Light of Christ shines for all," "Light, Life," "Alpha, omega," "Jesus Christ is victorious") sometimes accompany inscriptions of other types, but are also attested independently. Their function we designate as demonstrative, that is, aimed at demonstrating a certain formula, the purpose of which demonstration remains unclear (but not liturgical, in any case). As a special type we distinguish inscriptions of non-votive nature accompanying reliefs, and as yet another type - inscriptions designating ownership. Specific formulae of these last two types, as well as the liturgical formulae, are discussed in commentaries to relevant insccriptions. Under Varia we list inscriptions whose function either does not fall under any of the categories described below, or is unclear. Under Incerta we list those badly damaged texts whose function is it impossible to determine. Graffiti of uncertain nature, and those attested in complexes with inscriptions of other types (functions), are treated separately.

IV.3.A. Building Inscriptions

In spite of a relatively large number of building inscriptions in the Northern Black Sea region, the formulae used in them almost never recur, and for this reason they are treated in commentaries to specific inscriptions, alongside other unique formulae. We find only a few cases of similarity: Ἐπὶ + name in the genitive as an opening formula in the monuments of Early Byzantine Cherson and Bosporus (V 5, V 313, and possibly, V 9); consistent formulae in inscriptions of Theodoro's ruler Alexios (V 179 and V 180); and finally, popularity of the formula Ἐκτίσθη in Mangup (V 172, V 176, V 177, V 179).

We should mention that some building inscriptions provide no date or no year. Inscriptions without any date are V 15 and V 16 from Cherson of the "Dark Ages" and Late Byzantine V 203 from Tepe-Kermen. Inscriptions without a year date are V 17 from Cherson and V 226 from Mountainous Crimea. In the Early Byzantine period, inscriptions set up by the state were neved dated by year (e.g., V 5), while local inscriptions could be dated by a local era (V 6, and probably, V 265). In the VIth century, V 329 and V 330 (Hermonassa) were dated by indiction, not by year, while V 171 (Mangup) was dated both by indiction and by year of emperor's rule.

IV.3.B. Dedicatory Inscriptions

Dedicatory inscriptions are mainly attested in the Early Byzantine period; in the Middle Byzantine period we have V 237 in Livadia, V 248 and V 249 in Sudak, V 338 from Taman peninsula; and in the Late Byzantine period, dedications are attested in Inkerman, Alushta, and Taman peninsula (V 149, V 233 and V 334).

IV.3.B.a. Δέησις τοῦ δούλου τοῦ θεοῦ

This dedicatory formula is attested in the Northern Black Sea region four times and only in the Late Byzantine period: V 2 from Belgorod-Dnestrovsky (1431-1432 C.E.), V 149 from Inkerman (1271-1273 С.E.), V 233 from Alushta (1403-1404 C.E.), and V 242.2 from Partenit (1471 C.E.).

Elsewhere, the formula occurs quite often (more than 25 times, according to PHI7 Database) in the Early, Middle, and especially in the Late Byzantine period: geographically, the closest example is Beševliev 1964, No. 31, 238 (XIIIth century) and Pomyalovsky 1881, No. 93.

IV.3.B.b. Ὑπὲρ εὐχῆς

Ὑπὲρ εὐχῆς - is the most common dedicatory formula in the Northern Black Sea region, and equally in the Christian epigraphy. It is typical, first and foremost, of Early Byzantium and is represented mainly by inscriptions from Cherson (V 18, V 19, V 20, V 21, V 22, V 23, V 24, V 25), and by a single inscription from Mountainous Crimea (V 227). All inscriptions with this formula appear on architectural members (cornices, panels of the altar partition, base of a half-colonnete, mosaic). Analogous examples are known from all over the Mediterranean (e.g., Wessel 1989, No. 134; Feissel 1983, Nos. 4, 56, 98, 102B, 111, 257, 258, 261, 275, 276; Grégoire 1929, Nos. 2, 15, 40, 91–93, 131 (bis), 135, 138 (bis), 140; I.Eph 3286, 4139–4141; Milet VI 2 965; Hagel 1998, Anm 34, Aph 3; IGLS 1045, 1689, 2001, 2030, 2041).

IV.3.B.c. Ὑπὲρ σωτηρίας

This formula, which is related to the previous one, is attested with certainty only in V 28 from Cherson (regarding its possible continuation see commentary). It may have been also present in V 27 also from Cherson. In spite of the generally frequent occurence of this formula in Christian epigraphy (97 examples altogether, according to PHI7 Database), at the start of inscriptions it occurs more rarely. In our region, this formula most often introduces the dedicant's name (15 times), and more rarely - a continuation of the formula: ὑπὲρ σωτηρίας καὶ ἀντιλήμψεως (IGLS add. 175; SEG VIII, 232), ὑπὲρ σωτηρίας καὶ προσφορᾶς (IGLS add. 96a) or ὑπὲρ σωτηρίας καὶ μνήμης (IGLS 680). With the continuation καὶ ἀφέσεως τῶν ἁμαρτιῶν it is attested in V 249 from Sudak.

IV.3.B.d. Ὑπὲρ ψυχικῆς σωτηρίας

This formula is an expanded version of the previous one and is attested in the Northern Black Sea region in V 338 and V 219, where it is further expanded by an addition of καὶ ἀφέσεως τῶν ἁμαρτιῶν. It is notable that we have not been able to find any exact analogies. The formula ἕνεκα ψυχικῆς σωτηρίας is attested in Millet 1899, no. 1.12 (Mystras); the expression ψυχικὴ σωτηρία is listed in Δεμιτσᾶς 1896, no. 353 (Epirus) and Plassart 1923, 176, 178 (Attica).

IV.3.B.e. Ὑπὲρ εὐχῆς καὶ σωτηρίας

This combined formula, widely attested in Christian epigraphy (42 times, according to PHI7 Database), occurs in the Northern Black Sea region apparently only in V 26 from Cherson (end of the IVth-early Vth cent.).

IV.3.B.f. Ὑπὲρ ἀφέσεως τῶν ἁμαρτιῶν

In its pure form the formula occurs only once, in V 29 from Cherson. In Christian epigraphy, it is also generally rare as an opening formula: we could find only a single parallel from the Early Byzantine period - IGLS 2043 (Epiphaneia, VIIth cent.?).

IV.3.B.g. Ὑπὲρ εὐχῆς καὶ σωτηρίας καὶ ἀφέσεως τῶν ἁμαρτιῶν

This particular formula is a combination of Ὑπὲρ εὐχῆς, Ὑπὲρ σωτηρίας and Ὑπὲρ ἀφέσεως τῶν ἁμαρτιῶν, or of its components (see IV.3.B.e). It is attested in the Northern Black Sea region once, in a dedication (?) from the Herakleian peninsula - V 108 (XIVth cent.). We find analogies only in Middle Byzantine inscriptions: Thierry 1963, No. add. 2 В 1 (Kızıl Çukur in Cappadocia) and SEG XIV, 694 (in Geberkilise, Caria, twice)

IV.3.B.h. Ὑπὲρ ὑγιείας

The addition "for salvation and forgiveness of sins" can also be attached to the formula Ὑπὲρ ὑγιείας (see above, IV.3.B.g.). We find an example in Livadia - V 237 (X-XIIth cent.) where this formula is repeated twice. The formula "For health and salvation" is known in Christian epigraphy (see МАМА VIII, 426 (IVth cent.), and is also listed in Grégoire 1929, № 285 (Olympos in Lycia, Late Byzantine period?), only with a changed word order. The addition of words "and for the forgiveness of sins" is not known anywhere else.

IV.3.B.i. Oὗ ὁ θεὸς οἶδεν τὸ ὄνομα, ἐποίησεν

The formula is reconstructed in one inscription from Cherson, on a panel with relief (V 31). It is attested in several Christian inscriptions elsewhere (e.g., SEG XXXVII 464 d8; SEG XXIII 653 (ἐψήφωσεν), and it is undoubtedly related to the formula Ὑπὲρ τῆς εὐχῆς, οὗ ὁ θεὸς εἶδεν τὸ ὄνομα (I.Ilion 156; SEG XV 141; Σωτηρίου 1952, 198; Feissel 1983, No. 102A, 104, 109) and to some other similar formulae (e.g., IdC 73).

IV.3.C. Demonstrative

For the definition, see above (IV.3). Inscriptions of this type are known, primarily in the Middle Byzantine period, in Cherson, Bakla, Eski-Kermen, as well as Sudak.

IV.3.C.a. Φ(ῶς) Χ(ριστοῦ) φ(αίνει) π(ᾶσιν)

The abbreviation ΦΧΦΠ stands for Φῶς Χριστοῦ φαίνει πᾶσιν, not for Φῶς Χριστοῦ φωτίζει πάντας (as was suggested by Shangin; see commentary to V 42), which is confirmed by its full spelling in several cases (SEG VIII 56; SEG VIII 216; Guarducci, EG IV 453). It typically occurs as a supplementary symbol on four tips of the arms of a cross, or in the corners between the arms of a cross. The formula is attested in Christian epigraphy more than 13 times, according to the PHI7 Database, starting from the Middle Byzantine period. It typically occurs in abbreviated form (for exceptions see above), and in the Northern Black Sea region it always occurs as an abbreviation (V 35 is an unusual form of abbreviation) and with a cross (V 35, V 42, V 89 from Cherson, V 200 from Syuren fortress (all examples are of the Middle and Late Byzantine periods)). See in more detail: DACL I 2, 2325; VII 1, 685; VIII 1, 1108–1111.

IV.3.C.b. Φῶς, ζωή

The origin of this widely attested in Christian epigraphy formula (more than 30 times, according to PHI7 Database) is Job 1.4 (see Guarducci, EG IV 310, 439–440); in expanded form it occurs in IdC 54: "I am Light, I am Life." The formula is usually represented in one of two possible ways: with words completely spelled out (placed one over the other) and as a crossword with omega in the center. The first variation can be seen in V 36 and V 50 from Cherson (the latter is combined with an invocation) and V 220 from Eski-Kermen; the second variation is found in V 37 from Cherson, V 250 from Sudak and V 329 from Taman peninsula.

Both variations are attested in Christian epigraphy outside the area of our interest: the first in Popescu 1976, No. 49, 50; Grégoire 1929, No. 265 (this one is similar to our V 36 with respect to the position of letters vis-a-vis the cross; IGLS 1726; SEG XXXV 1119); the second — in Popescu 1976, No. 91; IGLS 1682, 1701, 1862, 1869 etc.

IV.3.C.c. Φῶς

A variant with just one word "Light" is attested in two inscriptions from Cherson: V 39 and V 40. It could be viewed as an incomplete form of the previous formula (see above), but the single-word usage is attested in Christian epigraphy outside our area: in the form of two intersecting words (SEG XXXV, 1562; Palestine, VItb century (?)); or in combination with alpha and omega (SEG XXXI, 674 (19); in Bulgaria, IV-VIth centuries), as in the case of V 39.

IV.3.C.d. Α Ω

Concerning this formula, which occurs in V 38, V 40, V 62, V 14.1.2, V 265, V 296, V 305 (all Early Byzantine), see in detail in RAC I, 1–4.

IV.3.C.e. Ἰ(ησοῦ)ς Χ(ριστὸ)ς νικᾷ

This formula, widely attested in Christian iconography, is always found in conjunction with the image of a cross. See in more detail Dölger 1910, 186ff. and CCAGV II, 245ff. In the Northern Black Sea region it is known from the end of the VIIIth - beginning of the IXth century (V 226) and is attested 20 times.

IV.3.C.f. Κύριος φωτισμός μου καὶ σωτήρ μου

In the Northern Black Sea region, this formula occurs only once, and in a very fragmented state, in V 97 from Cherson. It derives from Psalm 26.1, in our case used without the final phrase τίνα φοβηθήσομαι (as in Thomsen 1921, 107). Psalm 26.1 is often used in liturgy, e.g., as a prokeimenon. Although analogous inscriptions from Romania, Asia Minor and Syria appear on quite different objects: marble vessels (Popescu 1976, No. 60, 118), funerary monuments(?) (MAMA VI 385), and cornices (IGLS 1669, 1679), none of them are executed in the shape of a cross.

IV.3.D. Apotropaic formulae

Apotropaic formulae can be used in Christian epigraphy either independently, in special apotropaic texts, or also in invocations, funerary monuments and other inscriptions as supplementary symbols. In the Northern Black Sea region, however, only the first variation is attested (only 4 examples).

IV.3.D.a. Φεῦγε ζῆλος, Χριστός σε διώκει

This rare formula occurs only twice, on nearly contemporary inscriptions from Cherson (V 42 and V 109). V 42, in addition, bears another formula Φῶς Χριστοῦ φαίνει πᾶσιν on the arms of the cross. Both V 42 and V 109 carry images of a foliate cross with arms' tips of elaborate shape (the so-called, teardrops, see V 67 and V 89).

Similar examples from outside the region confirm the apotropaic function of this formula. The closest analogy is on a silver amulet from Smyrna (?) (Grégoire 1929, No. 90bis): Φθόνος. Φεῦγε μεμισμένη· Σολομῶν σε διώκει, which parallels other such amulets with the "seal of Solomon". Spells with the formula φεῦγε are widely attested in lapidary epigraphy: IGLS 1443 (Apamene, cornice, Χριστοῦ τὸ νῖκος. Φεῦγε Σατανᾶ); Gatier 1985, I, 160bis (Θεὸς βοηθός. Βάσκανε, φεῦγε). Apotraic formulae with the verb διώκειν are also known: IGLS 1599 (Idjâz): ἡ Τριάς, ὁ θεός, πόρρω διώκοι τὸν φθόνον. Shangin cites as a parallel to our formula a text from a manuscript at the Russian National Library (St. Petersburg) gr. 116 (CCAG XII 169): ...καὶ ὁ ἕτερος ἄγγελος τὸν στρόφον ἀπολύει· Φεῦγε, στρόφε· καὶ ὁ ἕτερος· Ὁ Χριστός σε διώκει.

IV.3.D.b. Σταυροῦ προκειμένου, ὁ φθόνος ἀπέστω

The formula is attested in the Northern Black Sea region only once: in V 43 from Cherson (VI-VIIth cent.). Literal analogies are not known, but there are many close parallels: [τοῦ στ]αυροῦ πα[ρ]όντος, ἔκθρος οὐ κ[ατισ]χύσι (IGLS 494; Qinnesrin, 550 C.E.), [σ]ταυροῦ προκιμένου, οὐδὲν ἰσχύει ὁ φθόνος (IGLS 1910; El-Bardouneh, VIth cent.), σ[τα]υροῦ [παρόντος], οὐδὲν ἠσ[χύ]ει φθόνος (Grégoire 1929, V 244 (3); Bargylia in Caria). The opening phrase, σταυροῦ προκειμένου, is attested in IGLS 1676 and 1696. The formula Φθόνος ἀπέστω is also familiar from SEG XXXVII 1271 (Anemurion in Cilicia, end of the Vth cent.). As is the case of the inscription from Bargylia, we might also be dealing with a syllabo-tonic poem: two lines with two amphibrachia in each, forming together one twelve-syllable verse.

IV.3.E. Invocations

The role of invocations increases in the Middle Byzantine period: before then, only V 49 from Cherson and V 266 from Kerch can be dated with any certainty.

IV.3.E.a. Κύριε (Χριστέ), βοήθει

This invocative formula is, without a doubt, the most widely attested in Christian epigraphy; it is known in lapidary inscriptions, as well as in graffiti and dipinti. Its origin is related to that of another no less popular formula - κύριε ἐλέησον (see below). In the Northern Black Sea region, it is attested multiple times, but only in non-lapidary form.

It is well known that in the Byzantine period the verb βοηθέω governs all oblique cases, but in the Northern Black Sea region, its normal usage with dative is attested only once (V 213), which is incidentally also rare in Christian epigraphy. Genitive was also rare with βοήθει (cf. Wessel 1989, No. 517; Feissel 1983, No. 262; Bandy 1970, No. 73A; Grégoire 1929, No. 96, 148, 226; I.Eph 185.14, 1285.15, 4312b, 4312c): in our region, it occurs in V 47 from Cherson, and even there the author slips into accusative when expressing the epithet (τὸν ἁμαρτωλὸν καὶ ἀνάξιον δοῦλον). In fact, most of the time βοήθει takes accusative. Notably, in Christian epigraphy, accusative in this formula occurs more rarely than dative or genitive (Beševliev 1964, No. 254; Grégoire 1929, No. 40bis; Studia Pontica 250; I.Eph 4148, 4284).

If βοήθει takes as object the name of a person, the latter almost invariably (with the exception of V 150 and V 251) is followed by an attributive phrase "your servant" (τοῦ δούλου σου/τὸν δοῦλόν σου).

As a direct object of the verb βοήθει we also find "this house" (τὸν οἶκον τοῦτον), in V 50, on an ancient funerary relief from Cherson (and in combination with the formula φῶς, ζωή); we find a parallel to such usage in IGLS 1451, 2635.

A related variation is Χριστέ, βοήθει, which is attested in the Northern Black Sea region once (V 48) followed by a direct object. For parallels see Ὀρλάνδος, Βρανούσσης 1973, No. 132; IG XII,5 712; Studia Pontica 278d; I.Eph 1285; I.Apameia und Pylai 65; SEG VII 954.

IV.3.E.b. Κύριε, ἐλέησον

This formula, a relative of the preceding one, plays an important role in liturgy, serving as a conclusion for most Byzantine litanies. In our region, it is probably attested in a very fragmented V 185. Elsewhere, parallels are numerous: e.g., SEG VIII 167, SEG XXXVI 1268 (bis); Bull. ép. 52, 151). In V 266, this formula is followed by "servant of God" (instead of "Your servant" - for which we find only two parallels: CIG 8915b and SEG VIII 745.

IV.3.E.c. Κύριε/Σωτήρ, σῶσον

Variations of this formula, directly addressing either Lord or Saviour, are attested once each: V 53 (also, possibly, V 345) and V 336. In V 53, we also observe a horrific mixture of three cases: τῷ δοῦλον σου ...σακίου (although it is possible that the article is in the accusative, from which the final nu has been dropped as is typical for Modern Greek, and the name appears in the genitive under the influence of σου). For parallels to this formula, when it is addressed to the Lord (Κύριε), see Beševliev 1964, No. 136 (XIII–XIVth cent.), Negev 1981, No. 82, 83, 85 (Early Byzantine); and when it is addressed to the Saviour (Σωτήρ): Grégoire 1929, No. 450 (Ankyra, IX–Xth cent.).

IV.3.E.d. Θεοτόκος, ...

Invocations of the Mother of God are numerous in Christian epigraphy. There are cases where nominative is used instead of vocative (e.g., SEG XXX 1701; Tel Feyran in Palestine, Vth cent.). In the Northern Black Sea region, there is only one case of this invocation: V 2 from Belgorod-Dnestrovsky.

IV.3.F. Funerary

This category of inscriptions is the most numerous in the Northern Black Sea areain all its sub-regions, and in all time periods.

IV.3.F.a. Ὑπὲρ (μνήμης καὶ) ἀναπαύσεως

In Byzantine epigraphy, this formula is used both in dedicatory and funerary inscriptions. In the Northern Black Sea region, however, it is attested only on funerary monuments. The formula is known in two variations: short (Ὑπὲρ ἀναπαύσεως) and expanded (Ὑπὲρ μνήμης καὶ ἀναπαύσεως).

The first variation occurs in two inscriptions, from Cherson (V 60) and Tepe-Kermen (V 206), for which there are many parallels elsewhere: e.g., Hagel 1998, No Mer 1 (Meryamlik); Grégoire 1929, No. 160, 244; IGLS 695, 9283. It is also important to note that a version of this formula occurs in a Jewish epitaph - CIRB 736.

The second variation is more frequent both in our region (V 61 from Cherson, V 115 from Bakla, V 253 from Sudak and V 339 from Taman) and in Christian epigraphy in general (cf. Hagel 1998, No Kan 7b; Negev 1981, No. 74; Lefebvre 1907, No. 410, 535, 619, 620, 627, 648). The funerary inscription from Taman is the closest to the Egyptian examples, as it also contains in its second half the formula Ἐτελειώθη... The addition of words τοῦ δούλου τοῦ θεοῦ before the name of a deceased in monuments from Bakla and Taman possibly finds an analogy in one reconstructed example (Popescu 1976, No. 12).

The exact formulation of V 314 from Kerch (Ὑπὲρ αἰωνίας ...) remains unclear, but if we were to judge by a single known parallel (SEG VI 442), the next word might have been μνήμης.

A special variation of the formula is Ὑπὲρ μακαρίας μνήμης καὶ ἀναπαύσεως on a funerary monument from Sudak (V 253). This variation could be a product of contamination with the formula μακαρίας τῆς μνήμης (see IV.3.F.e); alternatively, it might be a derivative of the formula Ὑπὲρ ἀναπαύσεως, κοιμήσεως καὶ μακαρίας μνήμης (BCH 25, 192b, Heracleia Pontica; MAMA I, 260, Dedeler in Lycaonia, Middle or Late Byzantine).

IV.3.F.b. Κύριε, ἀνάπαυσον

In the Northern Black Sea region we find several inscriptions with this formula. In three of them (V 62, V 64 and V 63 from Cherson) the direct object of the action is expressed as "the souls here laid up/down": τὰς ψυχὰς τὰς ἐνθάδε ἀνακειμένας (κατακειμένας). The closest analogy is IGLS 355B (Antiocheia, 521 C.E.), but its triple attestation specifically in Cherson indicates a particular local variation. A shortened version of the direct object (τὴν ψυχήν) is attested once: V 153 from the vicinity of Inkerman. For parallels, see Lefebvre 1907, № 97, 99, 513; Ricci 1903, № 11. Another variation "grant rest... along with the righteous" (V 64) is known from Lefebvre 1907, № 650 (Nubia). A related version is "your soul with the righteous" in V 65: cf. ICUR 4433; in the rendition "May God rest your soul along with the righteous," it is known from Smyrna in an inscription dated to 541 or 543 C.E. (Grégoire 1929, № 71).

In another inscription from Cherson (V 32), the object of ἀνάπαυσον is the name of a deceased, and the inscription appears to be dedicatory in function. Finally, V 254 from Sudak survives in fragments, but in it we find a combination of the formula with the word Amen (cf. Moab 320). Notably, according to PHI7 Database, the formula κύριε, ἀνάπαυσον is attested in Asia Minor four times (once with "rest the soul"), once in Greece, once in the Latin North Africa (questionable), while at the same time in Syro-Palestinian region it occurs 23 times, and in Egypt and Nubia, 70 times. Most likely, the formula came to Cherson from the East.

IV.3.F.c. Κεκοίμηται

This formula occurs only in V 66 (915 C.E.) and V 68 (1183 C.E.) from Cherson. Its peculiarity consists in the fact that it begins with a verb, and adds "who is in blessed memory and a servant of God" (which connects it to the formula which begins with Ἐκοιμήθη: see below). A standard version in Asia Minor is "Here fell asleep a servant of God": there are 12 examples, according to PHI7 Database, and all of them from Ankyra, and the only dated inscription (Grégoire 1929, № 5а) is from 569 C.E. The additional phrase ὁ τῆς εὐλαβοῦς μνήμης instead of "servant of God" is in Jerphanion 1928, 287, № 60, also from Ankyra. Outside of Ankyra, the formula is attested once in Rome (ICUR 19826; catacombs of Kyriake), where the name of the deceased is the first word, and "in peace" is added. Thus, in Cheron, we are either dealing with a borrowing from Ankyra, or with an independent tradition, unattested elsewhere in the Northern Black Sea region: conversely, other formulae popular in the region in the same period - the Middle Byzantine - are not known in Cherson. These are: Ἐτελεύτησεν (region of Bakla) and Ἐτελειώθη (Bosporus and its zone of influence; see below).

IV.3.F.d. Ἐνθάδε κατάκειται

The formula Ἐνθάδε κατάκειται (variant with Ἔνθα is more rare and occurs only in V 268) originates in Late Antiquity and occurs on pagan tombstones, but it becomes widespread in Christian epigraphy in the Early and Middle Byzantine periods. It is one of the most frequent in the Northern Black Sea region. In the Early Byzantine period it is attested in Cherson (V 72; in V 73, V 74, V 75, its exact form is unknown), in European (V 267, V 268, V 269, V 270, V 271, V 272, V 273, V 274, V 275, V 276, V 277, V 278, V 279, V 280, V 281, V 282, V 283, V 284, V 285, V 286) and Asian (V 332) Bosporus. Afterwards, it completely disappears. It is noteworthy that the formula is consistently rendered in this form, without such variants as Ἔνθα κεῖται (see, e.g., Guarducci, EG IV, 411, 415, 445, 452, 512, 525).

A variant of the formula, which was no less popular in the Northern Black Sea region, is the expression Ἐνθάδε κατάκειται ὁ δοῦλος (ἡ δούλη) τοῦ θεοῦ (a variation with Ἔνθα is known only from V 72). In the Early Byzantine period, it is attested in Cherson (V 71) and Pantikapaion (V 287, V 288, V 289; V 290, V 291, V 292, V 293, V 294), and in the Middle Byzantine period, in Bosporus (V 316 dated 767 C.E.; V 315 dated 819 C.E.) and in Kuban (V 336 dated 912 C.E.). At some point in the Xth century, both variants (Ἐνθάδε κατάκειται and Ἐνθάδε κατάκειται ὁ δοῦλος (ἡ δούλη) τοῦ θεοῦ) are mostly replaced by a new formula Ἐκοιμήθη ὁ δοῦλος (ἡ δούλη) τοῦ θεοῦ, but do not disappear completely: they are still attested in V 234 (Alushta, 1291-1392 C.E.), V 335 (Taman), and possibly V 242 (Partenit, 1471 C.E.).

IV.3.F.e. Ἐκοιμήθη

This formula appears already in Early Christian inscriptions (e.g., ICUR 3978, 10612, 11711; Wessel 1989, № 883; IGLS 733; Ricci 1903, № 2, 4) and continues to be used in the Middle Byzantine period (e.g., Malay 1994, 511 (IX–Xth cent.). In the Northern Black Sea region, it is known in this short form only in V 69 from Cherson (X-XIIth centuries).

A variant of the formula (the only exception might be V 215) which is much more prevalent in the region is Ἐκοιμήθη ὁ δοῦλος (ἡ δούλη) τοῦ θεοῦ, which is known in the Early Byzantine period elsewhere (e.g., I.Iznik 574), but spreads widely from the end of the Xth century (e.g., Guillou 1996, № 146 ( 1075 C.E.); Grégoire 1929, № 327 (Thyatira, 1006 C.E.); IdC 95 (X–XIth cent.), 96 (1052 C.E.); IGLS 810 (Antioch, 1063 C.E.), 814 (Antioch, 1042 C.E.)). In the Northern Black Sea region, this formula occurs in inscriptions from two different chronological periods: X-XIth centuries (V 67 from Cherson, V 122 from Basman, V 319 from Bosporus, and probably, V 70 from Cherson and V 322 of unknown origin), and middle-end of the XIVth century (V 232 from Ay-Vasil, V 260 from Theodosia, V 320 from Bosporus, and V 340 from Taman peninsula, and probably, V 235, V 256, V 318).

Finally, the variant Ἐκοιμήθη ὁ δοῦλος (ἡ δούλη) τοῦ θεοῦ becomes dominant in Mountainous Crimea in the XIII-XVth centuries. It is attested in Vysokoye (V 123, V 125, V 126, V 127, V 128, V 129, V 130, V 131), Golubinka (V 135, V 136, V 137, V 138), Krasny Mak (V 165), Kudrino (V 167, V 169), Laki (V 140, V 141, V 142, V 143, V 144, V 145), Mariampol (V 197), as well as on Mangup (V 189), Kachi-Kalyon (V 162, V 163, V 164), Syuren fortress (V 200), Tepe-Kermen (V 205), Cherkes-Kermen (V 210, V 211, V 212) and Eski-Kermen (V 224), in Inkerman (V 151, V 159), as well as in inscriptions of unknown origin (V 124, V 146, V 229, V 228), and in Shchebetovka (V 262) in Southeast Crimea.

Sometimes before the words "servant of God" (male/female) we find an epithet "who is in blessed memory" (ὁ ἐν μακαρίᾳ τῇ μνήμῃ). In origin, this epithet must be a variant of another widely known attribute "of blessed memory" (μακαρίας τῆς μνήμης or even μακαρίαν τὴν μνήμην (Sironen 1997)). In the Northern Black Sea region, it is attested only in V 67, while exact parallels are not on record.

In V 321, we find a unique formulation, which to some extent conveys a wrong idea: Ἔνθα ἐκοιμήθη. This expression could be a compression of the formula Ἔνθα(δε) (κατα)κεῖται ὁ δεῖνα. Ἐκοιμήθη (cf., e.g., IG XIV 245).

IV.3.F.f. Ἐτελειώθη

As an independent, that is, introductory formula (see, e.g., IGLS 2897, I.Eph 4143), Ἐτελειώθη in the Northern Black Sea region is attested in two inscriptions of the IX-Xth centuries from South and East Crimea: V 240 (Partenit) and V 315 (Bosporus). In both cases it is accompanied by the attribute "servant of God." This formula also sometimes introduces the date of burial (V 240, V 315, V 316 (Bosporus, 767 C.E.), V 336 (Tamatarkha, 912 C.E.), or is used with the metathesis of the name (V 188, Mangup, XIV-XVth cent.) - — Ἰω(άννης) ἐτελειώθη). This inscription cannot be identified as funerary proper, as it seems to be rather a note about the death (perhaps of a ktetor), painted on a colonnette of an altar partition. All other examples belong to the VIII-IXth centuries and originate from the zone of Bosporus' cultural influence (on the Bosporan influence in Partenit, see commentary to V 243).

IV.3.F.g. Ἐτελεύτησεν

In Early Byzantine funerary inscriptions of the Northern Black Sea region, this formula typically serves to introduce a date: V 71 (Cherson, VIth century), V 288 (Bosporos, IVth cent.). In the Middle Byzantine period, however, it appears as an opening phrase of epitaphs, but the usage is limited to the area of Bakla (V 117, Bakla, X-XIIIth cent.), V 134.3, V 134.8 (Gluboky Yar, 1034 C.E.). In Christian epigraphy, according to PHI7 Database, the formula is widely attested (175 times), and in both positions mentioned above.

IV.3.F.h. Μνημεῖον / Μνῆμα

The description of tombstones as "monuments" is attested in the Northern Black Sea 4 times, for which the term μνημεῖον is used 3 times. These are inscriptions from Cherson (V 76, V 78 and possibly, V 77) and one from Bosporus (V 295 dated 437 C.E.). There are over a hundred known analogies: (e.g., Wessel 1989, no. 624 (Catania), 826 (Syracuse); Feissel, Spieser 1979, № 25 (IV–Vth centuries); I.Iznik 551; Lefebvre 1907, no. 515, 581).

The term μνῆμα occurs twice, both times in V 79 from Cherson. It has many parallels (about 130), e.g., Wessel 1989, nos. 828, 829; Hagel 1998 (36 times), Gatier 1986 (10 times).

IV.4. Epigraphic dating and eras

In comparison with antiquity, the number of given dates in Byzantine inscriptions of the Northern Black Sea region is much lower. They occur almost exclusively in building and funerary inscriptions, and in the latter category become compulsory, in contrast to earlier times.

Most commonly a date is introduced by the word ἔτους, more rarely by the word ἔτει (it is attested only in Goryanka, XIVth century: V 142 and V 143) and also by a prepositional phrase ἐπὶ ἔτους (XIV-XVth cent., except V 134.3: building inscription V 2 (Belgorod-Dnestrovsky), V 175, V 176, V 177 (Mangup), V 209 (Cherkes-Kermen), funerary V 123, V 125, V 130 (Vysokoye), V 124 (Vysokoye?) and V 167 (Kudrino) and commemorative V 215.3 (Chilter-Marmara, 1402-1403 C.E.). The formula ἐν τῷ ἔτει in V 305 (Pantikapaion, 491 C.E.) in this form is otherwise unattested in Byzantium, although its relative, ἐν ἔτει, was very popular in Early Byzantine Syria and Palestine (about 50 examples, according to PHI7 Database) — it originates, however, from the Bosporan epigraphic tradition of classical antiquity (CIRB 1315 and 13 other examples with the year date after the article). In the Northern Black Sea region of the Xth century, we also twice find another formula - ἔτος, well known in Byzantium (62 times, according to PHI7 Database): V 66 (Cherson, 915 C.E.) and V 172 (Mangup, 994-995 C.E.).

IV.4.A. Local eras

In the Early Byzantine period, we find in the Northern Black Sea region the practice of dating by two local eras: that of Cherson and that of Bosporus.

Until now the only source of evidence for the local Chersonesian era has been the inscription of emperor Zenon (V 6, 487-488 C.E.). It is dated to the year 512 C.E., from which we may conclude that the start of the Chersonesian year-count began 25 years before Annus Domini. We may hypothesize that dating according to this local era at the end of the Vth century, that is, after a long interruption (the last ancient inscription dated in this way is IOSPE I2, 439) could have been due to a surge of Chersonesian patriotism, which had also found expression in the creation, in the Vth-VIth centuries, of a local chronicle, relating the events of the IIIrd-IVth centuries and used by Constantine VII Porphyrogennetos in De administrando imperio 53. In my opinion, we could also add V 24 (674-675 C.E.?) to the list. In that case we would find that dating by the local era continued in use until the end of the VIIth century, which is supported by the information provided by Byzantine historians who wrote about the semi-independent status of the city up to the beginning of the IXth century.

The majority of Early Byzantine monuments dated by the local Bosporan era belong to the last third of the Vth century (V 305, V 304, V 331, V 342). At the same time, V 272 (323-324 C.E.?) and V 294 (437 C.E.; see also commentary to V 273) show that the tradition of dating according to the Bosporan era (beginning from 296 B.C.E.) continued in the IV-Vth centuries. The last inscription dated in this way is V 304 (497-498 C.E.), see also commentary to V 275 (507-508 C.E.?): the end of this tradition should probably be attributed to the incorporation of the Bosporan kingdom into Byzantium in 517-527 C.E.

IV.4.B. Dating by rulers

As we have remarked earlier, the growing political role of Byzantium in the Northern Black Sea region led to the dwindling of local epigraphic traditions, including the principle of dating: in building inscriptions, a date is always given by the rule of Byzantine emperors. In general, dating by kings and high state officials also existed in the Bosporan Kingdom (even if duplicated by the local era); however, in the period of Byzantine control, the principle of dating by emperors (typically introduced by the formula ἐπὶ τοῦ δεῖνος) became dominant, accompanied also by the specification of indiction, as a sign of pan-imperial chronology. In the Early Byzantine period, the examples are V 5, V 7, V 171 (the only one to mention the year of rule), V 265, V 329. These inscriptions also mention other officials, particularly those involved in building projects. As a local variant, the formula "at the time of bishop" seems to have been in use (V 9). The tradition of dating by emperors' rule continued in the Middle Byzantine period (V 11, 1059 C.E.; V 12, with the formula επὶ τοῦ δεῖνος βασιλέως), although in V 17 only the topoteretes (deputy/commander) of the theme (military unit or district) is named, while in V 226 the named officials are Khazar: khagan and tudun. In the Late Byzantine period, inscriptions that mention emperors are unknown. Instead the principle of dating by local rulers emerges: V 2, V 148, V 173, V 174, V 175, V 176, V 177, V 178, V 179, V 180, and in most cases they are accompanied by a date "since the creation of the world." In Mangup, starting in the Xth century (V 172, V 179) and in particular from the rise of Theodoro (V 14, V 148, V 179, V 180), we find a characteristic and unique formula ὑπὸ ἡμερῶν. This formula was similarly applied to non-Greek rulers (Temir-Kutluk in V 144, V 1414 C.E.) who also imitated Byzantine emperors (cf. МАМА IV, 94; VI, 340; Feissel, Avramea 1987, № 25) in using the formula ἐπὶ τῆς βασιλείας (V 226 and possibly V 177). Finally, we should note one Late Byzantine case where the dating is given by the rule of a mitropolites.

IV.4.C. The Era "since Adam"

The dating "since Adam" is attested in 5 inscriptions that form a group: V 307 (Kerch, 691-692 C.E.), V 316 (Kerch, 767 C.E.), V 315 (Theodosia, originates in Kerch, 819 C.E.), V 317 (of unknown origin, 884 C.E.), and V 243 (Partenit, 906 C.E.). All of them, except the last one whose origin is unknown, are connected to Bosporos (the author of V 243 was a monk Nikolaos from Bosporos). This fact allows us to suggest that the principle of dating "since Adam" was in use in Bosporos in the period from the late VIIth to the early Xth centuries. In Byzantine epigraphy, such dating is very rare and nowhere occurs before the end of the XIIth century (Stylianou 1960, 98).

In V 336 (912 C.E., originating from the eastern shore of the straits of Kerch), which is palaeographically close to V 243 and 333, we already find the reference "since the creation of the world" (see below). This reference was usually dropped, but could have been included here specifically to contrast it with the date "since Adam." The chronological limits of usage for the dating formula "since Adam" in the area of Bosporus are 691 C.E. to 906 C.E., coinciding with the period of Khazar control over Bosporus. The disappearance of the formula should be associated with the establishment of the Byzantine imperial power in the region (see Mayko 2009). The typical Byzantine dating principle by year, without era, begins to dominate in the Bosporus from the middle of the XIth century (V 319, V 340).

In V 243 and V 316, we also find a reference to indiction. In V 307, such reference might not have survived.

In many inscriptions, a year "since Adam"/"since Creation" is juxtaposed with an indiction, which does not however automatically signify its start in September rather than March. Still, three inscriptions strongly indicate the start of the year in September: in V 243 and V 315 only the dates of a Septembrian year would correspond to the mentioned days of the week, while V 11 (6567 "since Creation" could have been set up only in 1059 C.E., because in 1060 C.E., Isaak Komnenos was no longer emperor. We may conclude on the basis of these observations that starting from the Xth century, a year "since Creation" must have been understood as a Septembrian year, and in two epigraphic traditions at once: Chersonian (V 11) and Bosporan (V 243 and V 315).

IV.4.D. The Era "since Creation (of the world)" (Anno Mundi)

The only dating by the year "since Creation" is in V 336, mentioned above: here, it could have been used in contradistinction from the era "since Adam" (see IV.4.C). According to PHI7 Database, the greatest popularity of this form of dating is attested in the IX-Xth centuries.

In the majority of cases, there is no designation of an era, and a date is simply introduced by a reference "in the year ..." - all together there are 61 examples, and they are attested in all sub-regions of the Northern Black Sea coast. The earliest example dates to the year 915 (V 66), which corresponds to the date of the first appearance of the era "since Creation" in the Northern Black Sea region. Among the building inscriptions from Theodoro we find two cases where (apparently under the influence of the Palaiologian epigraphy), the year date is spelled out: either fully (V 241, V 1487 C.E.) or partially (V 178, V 1453 C.E.; cf. also V 258, see below).

IV.4.E. The Era "since (the Birth of) Christ"

The dating "since Christ" occurs in the Northern Black Sea region only once - in V 258 (Sudak, 1412 C.E.) - where it can be attributed to the Genoese indfluence (see also commentary to V 172). According to PHI7 Database, in the entire Byzantine epigraphic corpus, the formula ἀπὸ Χριστοῦ is attested 4 times, and in each case it is listed as a supplementary dating formula.

IV.4.F. Alternative Count of Days of the Week

In two inscriptions from Mountainous Crimea, dated to the end of the XIVth century ((V 167, V 1442 C.E., Kudrino) and V 124, V 1437 C.E., Vysokoye (?)) we find an alternative count of days of the week, beginning with Monday. At the same time, there is just one example of a traditional count in the Northern Black Sea region in the Late Byzantine period: V 242.1 (1472, Partenit). Such traditional count is already attested in Partenit in the Xth century (V 243) and thus could have been a stable local tradition. We cannot therefore conclude with certainty that the alternative count was typical for the Gotthian eparchy.

IV.5. The Language of Inscriptions

In general, the Greek of the Byzantine inscriptions in the Northern Black Sea region falls in with what we know about the development of the language in the IVth-XVth centuries. The main question that arises in conjunction with our geographic zone of interest is whether any attested aberrations from the norm should be seen as legitimate linguistic phenomena or as a result of "barbarization" of Greek in the contact zone of the Northern Black Sea coast. This is a difficult question to answer. In addition, we should address the question of whether some forms might be dialectal.

Two main criteria allow us to evaluate the language of inscriptions: phonetics, whose special character finds reflection in orthography, and grammar, primarily, morphology. The analysis of syntax is mostly precluded by the formulaic nature of inscriptions. We also exclude from our analysis all questionable and restored readings.

IV.5.A. Orthography and Phonetics

Some phonetic features in the Byzantine inscriptions of the Northern Black Sea region derive from Late Antiquity. One of them, for example, is the use of beta to convey Latin sound "b" (V 5, 392-393 C.E.), the name Σαμβάτιος instead of Σαββάτιος (V 17, V 282, see commentary).

During the Byzantine period, we can observe three stages of phonetic changes in our region. The first stage, which can be studied using examples from Cherson and Pantikapaion, marks the disappearance of long vowels and transformation of the diphthong "ai" into "e," while eta is used to convey both the sounds "e" and "i": πρησβυτέρον and Χρηστοφόρου (V 61), ἐνθάδη (V 282); ει in one case (V 282) can be alternatively used to represent the sound "ei" or "i": κατάκειτη and Σωτήρεις. Juding by the dates of these examples, the first stage of phonetic changes should be placed in the IV-Vth centuries, rather than in the VI-VIIth, as was the opinion of Latyshev (Latyshev 1896, 24).

The second stage of phonetic changes, attested in inscriptions, reflects all the typical features of itacism, except for the transformation of upsilon into "i." The first dated examples that attest the change are V 6 (487-488 C.E.) and V 305 (491-492 C.E.). The change of "e" and the diphthong "ei" is also found in an earlier V 5 (392-393 C.E.), but this one probably does not belong to the Northern Black Sea tradition. The fate of the diphthong "oi" (which by the Roman period conveys the same sound as upsilon) is less clear: the first example of its transformation into "i" dates to 906 C.E. (V 243, Partenit).

The final stage of phonetic changes, which starts in the IXth century and is attested in all Byzantine territories, is the transformation of upsilon into the sound "i." The earliest examples in our region are V 50.2 (Cherson, IX-Xth cent.), V 243 (Partenit, 906 C.E.) and V 172 (Mangup, 994-995 C.E.).

The interchange of nominative and accusative cases is typical for Late Byzantium in general and might be related to the fact that the final nu in the accusative was no longer sounded; for other examples, see ἐκτήσθην instead of ἐκτήσθη in V 241 (Partenit, 1427 C.E.).

One expression of dialectism that should be noted is the dropping of nu from some verbal forms such as part. praes. and aor. act. –οντ- and –αντ- (V 52; Cherson, V–VIth cent.; V 108, twice; Cherson, XIVth cent.).

Another phenomenon attested in the Northern Black Sea region of the Late Byzantine period is the interchange of theta and tau: τοῦθο (V 108; Cherson, XIVth cent.), ἐτελιότη (V 188; Mangup, XIV-XVth cent.); and the interchange of chi and kappa: the name Τηλεχλῖς instead of Τηλεκλῆς (?) in V 82.3 (Cherson, 2nd half of the IVth cent.); the reverse interchange is in the name Σωτήρικος instead of Σωτήριχος, in Late Byzantine Mountainous Crimea (V 121, V 208; although this case might be a distinct name).

We should also note a phonetic peculiarity attested in Belgorod in the XVth century: the forms βιηθείᾳ (V 1) and αὐφέντος, ἀφεντίας (V 2.3), in which upsilon drops out, reflecting the change in pronunciation. For this we have a parallel on a tombstone dated 1635 C.E. from Moscow (Avdeyev, Vinogradov 2012).

I give little weight to such ordinary orthographic mistakes known among Byzantine scribes as haplography (e.g., V 5, V 134, V 243) and dittography (e.g., V 128), because they do not always reflect the ways of pronunciation. Still some spellings are noteworthy: ἀννάπαυσον in Cherson (V 64, V 159), as well as κατάκιτε in Bosporus (V 268, V 269, V 274, V 275, V 285, V 286, V 287, V 289.1, V 291, V 294), which does, however, alternate with κατάκιται). A similar type of mistake is the confusion of letters that visually resemble one another, such as lunate epsilon and sigma in V 18 and V 19, and later, incorrect spelling of diphthongs, e.g., ωυ in V 123, βοΰθη in V 262.

IV.5.B. Grammar

The main change in the category of nouns, albeit typical of Late Antiquity in general, is the contraction of inflections in personal names ending in –ιος (see Tokhtasiev 2007). There are many examples from Bosporus in the IVth-Vth centuries (V 270, V 273, V 274, V 282, V 297), to which we should add Θεοδωράκις (V 71; Cherson, VIth cent.), Βασίλης (V 4; V 126; Vysokoye, 1448 C.E.), Δημέτρις (V 82.1; Cherson, 2nd half IVth century), Μιχαήλης (V 108, V 91.1, V 91.3; Cherson, X-XIth cent.; V 138, Golubinka, 1271 C.E.?), Μαυρίκις and Συμεόνις (V 124; Mountainous Crimea, 1387 C.E.), as well as a noun μημόριν (V 80; Cherson, middle of the IVth cent.). Regarding the form Κυρακός instead of Κυριακός, see commentary to V 278.

We find that instead of the pronoun οὗτος other derivative colloquial forms are used in inscriptions: τοῦτος (V 16.2, Cherson, VIII-IXth cent.; V 176, Mangup, XIVth cent.) and τοῦτοι (V 219, Eski-Kermen, XIIIth cent.); του instead of αὐτοῦ (V 197.2, Mariampol, XIV-XVth cent.; in V 166 we find αὐτοῦ του in this position). In V 42 (Cherson, XIIth cent.), a non-classical form διώχνω is attested instead of διώκω (see commentary).

As far as verbs are concerned, there is an example of an unaugmented aorist (V 203). The main change, however, affects the governance of cases. We have remarked on the verb βοηθέω in IV.3.E.a. The verb σῴζω takes the genitive (V 2.3), and the verb εὔχομαι the accusative, in the sense "to pray for someone" (V 243). A more difficult case is the word διαφέρων: acting as a participle, it takes the dative (V 22, cf. I.Iasos 638, IGLS 2028A), but used substantively (more rare), it takes the genitive (V 20, V 21, V 24).

We also witness a lapse in the correct usage of cases and prepositions: e.g., preposition σύν in V 149 (Inkerman, 1272–1273 C.E.) takes the genitive. With nouns, irregularities in cases are rare, but do occur, e.g., accusative instead of genitive (typical for Byzantium) in V 243.

A very particular mistake is the use of the word υἱός instead of θυγάτηρ in V 130 (see commentary).