6. ITALO-GREEK SCRIPTS (TENTH – TWELFTH CENTURIES)

In the preceding pages, we have classified a number of Greek minuscule scripts in terms of style or chronology, but not in terms of geography; and this may surprise you, especially if you have some experience with Latin paleography, where scripts are routinely attributed to countries, geographical areas or even individual scriptoria with great confidence. This is only rarely possible with Byzantine scripts. This may be be in part a reflection of the political reality that the Byzantine Empire was very centralized, unitary state, at least after the Arab conquests, which subtracted from its territory cities like Alexandria, Antioch or Caesarea, which might have continued competing with the capital for cultural, if not for political primacy; a unitary culture of writing would be a natural reflection of this circumstance (thus Perria in her manual, p. 97). Another explanation was suggested by H. Hunger in the introduction to his 1954 paper about the Perlschrift (p. 23), and it is the fact that a number of medieval monastic libraries in the West survive to the present day, either in their original home or after having been moved collectively to a public library, whereas Byzantine libraries have nearly all suffered dispersion, with a few exceptions (Mt. Athos, Meteora, Patmos) which, however, consist mostly not of books produced on site but rather of volumes brought from various other locations by owners seeking refuge for themselves and their libraries. This would suggest that geographical differentiation may in fact have existed, but that the key to interpreting it has not (yet) been discovered; in that sense, these remarks constituted a fitting introduction to Hunger's identification of a specific writing "style" or "canon" (Hunger used the words more or less indifferently, much to the later consternation of Cavallo and his followers). It must be seen as a disappointment, then, that, in terms of geography, neither Hunger nor anyone else since then has been able to say anything very specific about the geographical distribution of the Perlschrift, beyond noting that about 4 in 10 of the Perlschrift exemplars with a known geographical origin were copied in Constantinople. Hunger himself suggested as a "working hypothesis" that the Perlschrift should be considered a Constantinopolitan script; but in fact a proportion of 40% would also be perfectly compatible with the hypothesis that this script was equally prominent throughout the Byzantine Empire, which in fact brings us back to Perria's view that geographical differentiation of Byzantine minuscule scripts is not really possible.

The earliest dated manuscript in minuscule which was certainly produced in southern Italy is now kept at the Monastery of St. John the Theologian on Patmos (Patm. 33), and was copied in Reggio di Calabria in 941 CE by two scribes named Nikolaos and Daniel, one of which (Lake 1,15, plate 33) writes an "Old Angular Minuscule", while the other (ibid., plates 31 & 32) writes a script which is generally taken to be an "Anastasius Type" (which we have already seen, in part D of this page). This has sometimes been taken to show that the "Anastasius Type" was localized in southern Italy; more likely it demonstrates that south Italian Greek manuscripts did not always exhibit a distinctive, regional script but rather participated in fashions attested also elsewhere; it is anyway very unusual among Italo-Greek manuscripts in that it is a very large, luxuriously decorated volume (see most recently I. Hutter, "Patmos 33 im Kontext", RSBN 46 [2009], pp. 73-126; Hutter also argues for the intriguing hypothesis that the script seen in Patm. 33 is not an "Anastasius Type" at all, but instead the rather awkward result of a scribe attempting to modify his normal script, which according to her hypothesis is the Angular Minuscule we see in Par. gr. 515, to the unusually large dimensions and line-spacing of the Patm. 33). The very fact that this manuscript ended up on Patmos, just off the coast of Asia Minor, shows that south Italy was not culturally isolated: books travelled, and there is no doubt that scribes did also; indeed there is reason to believe that many of the Greek-speaking inhabitants of southern Italy in those troubled times were refugees from the eastern parts of the Byzantine Empire which had recently been conquered by the Arabs. Obviously such refugees would have been likely to bring books, as well as scribal traditions, with them. At any rate, other tenth-century Greek manuscripts copied in southern Italy also present relatively unremarkable angular scripts, e.g. Vat. gr. 1591 (964 CE); Reg. gr. 75 (ca. 982 CE; Lake 9,335).

Nonetheless, there are several Byzantine scripts which do indeed seem to be distinctively south Italian.

The most distinctive of all is the so-called "Ace of Spades Minuscule", a name which was coined by Devreesse (in his already-mentioned book, p. 34) and refers to the shape of an epsilon-rho ligature which it frequently uses (but which is very well-attested in other scripts as well and whose presence on its own should definitely not be taken as a sign of Italo-Greek provenance!). The most salient characteristics of this script, which are all illustrated in the sample below, are:

None of the manuscripts written in the Ace of Spades Style has a subscription explicitly mentioning its date or place of execution; nonetheless, there is no doubt that this is in fact an Italo-Greek script, and it can be quite securely dated to a period running from the middle of the ninth to the middle of the tenth century. This is because of a number of manuscripts in which it is combined with other scripts which can be confidently dated and localized (namely, the ones we will see below; see P. Canart's article in Atti del 4 Congresso Storico Calabrese, pp. 55-69:60-61 [repr. in Id., Études…, pp. 215-229:220-221]), sometimes with both scripts evidently being the work of a single scribe.

For practice reading and transcribing the "Ace of Spades" minuscule, please see Vat. gr. 1553.

Hunger's counterfactual intuition that Greek minuscule scripts would be more easily classifiable if more Byzantine monastic libraries had remained intact is perhaps best vindicated in the case of the script associated with St Nilus the Younger and his "school", which, while well documented and easily identifiable thanks to the large number of exemplars which have remained in the library of the abbey founded by St Nilus at Grottaferrata near Rome, is yet perhaps not so distinctive that it would have been singled out by modern paleographers as a style in its own right if these exemplars had been dispersed far and wide. A native of Rossano (today a small town in the province of Cosenza in northern Calabria but in the ninth century an important Byzantine administrative center), St. Nilus left his home town sometime before 980 CE with a group of followers, sojourning in various parts of Calabria and Campania before founding the abbey at Grottaferrata in 1004 and dying shortly afterwards. These details and many others are related in a biography (see n. 1370 in F. Halkin's B[ibliotheca] H[agiographica] G[raeca]) which is preserved in a manuscript written at Grottaferrata by one of his disciples and still kept in the library there (Crypt[oferratensis] B. β. 2; see the edition by G. Giovanelli). This βίος also informs us that Nilus copied a great many manuscripts, despite not owning an ink-pot (he instead shaped one out of wax on his desk: Giovan. § 18, p. 65); that he used a "delicate and compact handwriting" (λεπτῷ καὶ πυκνῷ χρώμενος ἰδιοχείρῳ: § 15, p. 63); and that he taught the art of calligraphy to his disciples (Giovan. § 20, p. 67).

Samples of Nilus' own script may be seen in the manuscripts Crypt. B. α. 19, B. α. 20 and B. β. 1, the first of which is shown in the viewer below.

You may also gain a general impression of this type of script from the detail below, which is taken from Ott. gr. 251, attributed to a member of Nilus' "school" named Arsenius (RGK 3,49; see E. Follieri's article "Due codici greci già cassanesi…" in the Festschrift for Giulio Battelli, pp. 159-221, repr. in Ead., Byzantina et Italograeca, pp. 273-336).

The best survey of this script is still the fundamental article published by Santo Lucà in the proceedings of a conference held in 1988 (pp. 319-387), who described it in detail and distinguished between several sub-types, many of which area associated with specific scribes who may be thought of as members of Nilus' "school", which constituted, in Lucà's memorable phrase, an "itinerant scriptorium" until it came to its definitive location in Grottaferrata, where many of the relevant manuscripts are still preserved (however, most of these were written in Calabria or Campania in the second half of the ninth century, during the "itinerant" phase of the school before the abbey of Grottaferrata was founded). Lucà assures us (p. 137) that the scripts of this school are "easily recognizable", echoing Follieri's view (in paper at the Paris conference, p. 149; repr. in Byz. et Italogr., p. 220) that this is one of the "tipizzazioni grafiche veramente caratteristiche" of Byzantine Italy (thus also Perria in her manual, p. 105); but in fact none of these scholars is able to give a description of this script which would allow one to recognize it. Follieri (p. 150) describes it as "a minuscule of small module, round, vertical or slightly right-leaning, with low proportions of minuscule letter-shapes (especially lambda, kappa, pi), often rich in abbreviations, some of which are characteristic". Perria's description (p. 107) is similarly generic: "rather cursive, a script with a fluid appearance, small module, whose letters have a shape intermediate between round and square, often rich in abbreviations". Lucà's lengthy description, on the other hand (pp. 323-5), is so expansive that it could be taken to apply in one way or another to almost any imaginable minuscule script.

To some extent, as noted by Lucà himself (p. 326), this paradoxical situation may be explained by the fact that the scribes of Nilus' school seem to have sought above all to produce a harmonious overall effect on the page, avoiding any excess in execution which would disturb the eye. The result is a script which does not provide any salient characteristic which could be used to describe or distinguish it (in this specific sense it is similar to the Perlschrift). And yet, this script really is recognizable, once your eye has gotten used to it. One thing that makes it so is certainly a general impression of "flatness", which, however, is produced not so much by the shapes of the individual letters (you can see in the sample above that many letters, in particular epsilon, omikron sigma and upsilon, in fact have tall rather than wide "aspect-ratio") but rather by the spacing of the letters and, above all, by the width of the strokes, which seem to come from a round and fairly large nib. This "flatness" is a characteristic also of other Italo-Greek scripts; here it is combined with a very regular module: one has the impression that the scribe is making a sustained effort to create letter-bodies of uniform size. Said module is indeed "small", as is always noted; however, the lines of writing are also very close together and the ascenders and descenders are short, so that the overall impression is not so much of a "small" as of a "dense" script, as was indeed noted by St. Nilus' biographer (πυκνῷ […] ἰδιοχείρῳ).

For practice in reading and transcribing this script, we have chosen Vat. gr. 2138, which was written in Capua (Campania) in 991 CE, i.e. about a decade and a half before the founding of the abbey of Grottaferrata, by a scribe named Κυριακός (RGK 3,358). It was kept at Grottaferrata until the eighteenth century, when it was transferred to the Vatican Library (see S. Lilla, I manoscritti Vaticani greci, p. 82). It is considered one of the finest examples of the script of the "school" of St. Nilus, but was not chosen for the sample above because it is untypical of that script in one specific way, namely the noticeable stroke contrast which indicates that the scribe used a flat nib held at an oblique angle.

There is no doubt that Greek books continued to be copied in Rossano after St. Nilus left his home town; and in 1095 CE, a little over a century after his departure, a new monastery was founded there by Bartholomew of Simeri. It was dedicated to the Θεοτόκος Ὁδηγήτρια (the Mother of God who Shows the Way) but is commonly referred to as the Monastery of "Santa Maria del Patìr", the Πατήρ in question being the founder St. Bartholomew of Simeri. The monastery had a scriptorium and a considerable library, which remained in the abbey until the late seventeenth century, when, faced with the general degradation of the state of the Greek monasteries of Calabria, the General of the Basilian monasteries of south Italy, Pietro Menniti, brought it, together with a number of other Calabrian monastic libraries, to the Collegio di San Basilio in Rome. These "Basilian" collections were then purchased by the Vatican Library in 1786; today they are the Vaticani greci 1963-2123.

In twenty-two of these manuscripts (namely, Vat. gr. 1989, 1991, 1992, 1998, 1999, 2000, 2003, 2009, 2016, 2021, 2022, 2042, 2048, 2050, 2060, 2064, 2089, 2091, 2115, 2119, 2121, 2123), together with five further Vatican manuscripts which were not part of the purchase of 1786 (namely, Barb. gr. 482, 501; Vat. gr. 1270, 1495, 1642), as well as about fifty manuscripts now kept in other libraries, S. Lucà (RSBN 22-23 [1985-86], pp. 93-170) has identified a "Rossano style" (for an updated account and recent bibliography on this style, see D. Bianconi's article "Lo stile di Rossano. Culture grafiche tra metropoli e periferia", Scripta 10 [2017], pp. 9-31:9-10). Many of these manuscripts are explicitly localized at the Patìr monastery by their colophons. The dated exemplars attributed to the Rossano Style run from 1086 CE (certain pages of Sinait. gr. 401) to 1125/6 CE (Vat. gr. 2048, ff. 141-220), with the greatest concentration in the first decade of the twelfth century (however, Bianconi does not consider the Sinait. gr. 401 to be a representative of this style).

Truth to tell, the "Rossano Style", as described by Lucà, is more a style of manuscript than a style of script. Lucà himself notes (p. 99) that "the Rossano style (…) does not present particular graphic characteristics which allow it to be immediately recognized; similar individual letters, ligatures and pseudo-ligatures are found in other contemporary scripts and especially in the Perlschrift." He then goes on to list non-graphic characteristics which are typical of this style (of manuscript), namely pale ink; the use of certain patterns of page ruling, sometimes reinforced in pencil (which is unusual); quires of eight leaves with a flesh side on the outside and quire numberings (or "signatures") on the first recto and on the last verso (this is typical of Byzantine manuscripts); and a simple "reverse silhouette" decoration (often referred to with the French term dessin en réserve) in carmine red which recalls a style known as Blütenblattstil, well-known from manuscripts produced elsewhere in the Byzantine world from the mid-tenth century onwards (the term was coined by K. Weitzmann, Die byzantinische Buchmalerei…, pp. 22-32) but which elsewhere is usually polychrome (see the samples below of "classic Blütenblattstil and of "Rossano Style" decoration).

As for the script itself, here again Lucà's description is rather generic (pp. 97-98): "a small, fluid script, hanging from the ruling line, with regular and uniform letter-bodies, with vertical or slightly right-leaning axis, marked by balance beween height and width and by ascenders and descenders of moderate length (…)"; he also noted a fairly high proportion of majuscule letter-shapes. Of course this description could apply to a great many scripts, including notably many exemplars of Perlschrift; and indeed some scholars have suggested that the "Rossano style" is merely a local variant (or "allotrope", in M. B. Foti's memorable expression, Il monastero..., p. 31) of the Perlschrift. Recently, D. Bianconi (in his already mentioned article in Scripta 10 [2017], pp. 9-31), has suggested that a specific precursor of the "Rossano style" might be found not in any Italo-Greek script but rather in a type of Perlschrift which is found in several manuscripts which were copied at the Constantinopolitan monastery of the Θεοτόκος Εὐεργέτις, such as Marc. gr. Z. 101, Bodl. Auct. T.2.2, Paris. Coisl. 248, or Paris. gr. 581. At any rate, in addition to the generic resemblance with the Perlschrift, Lucà also noted, in the "Rossano style", certain characteristic letter-shapes, including majuscule delta with a curved top; psi with a curved baseline stroke; zeta shaped like a 2 or a 3; rho open to the left; slanted majuscule nu. None of these are singular to this script (the last two are shared in particular by the "Ace of Spades" minuscule), and moreover none are used systematically in the "Rossano style", but only sporadically. One feature of this script which is truly distinctive is the habit of placing the acute accent on an initial vowel indifferently before or after the breathing which accompanies it (Byzantine scribes normally follow the same habit we do, putting the breathing always first). Most of these features are seen in the sample below from Vat. gr. 2050, f. 37r, copied at the Patìr monastery in Rossano in 1105 CE by a monk named Bartholomew (RGK 3,59; his subscription is on f. 117r; he was certainly not Bartholomew of Simeri, the founder of the monastery, since the latter is referred to in the subscription itself as "our father Bartholomew"). Note in particular the word ἄλλως in the first line of the second column, with the acute accent preceding the smooth breathing.

For practice reading and transcribing this script, please see Vat. gr. 2050.

There is one more style of Italo-Greek script which is associated with a specific location, namely the one which R. Devreesse identified in his already-mentioned book (p. 40) and which he proposed to call "écriture de Reggio"; it was studied in greater detail in an article by P. Canart & J. Leroy ("Les manuscrits en style de Reggio…", in the 1974 Paris conference proceedings, pp. 241-261, repr. in P. Canart, Études…, pp. 319-339). Despite the name given to this script by Devreesse, it is in fact localized on either side of the Strait of Messina, i.e. in Reggio on the mainland side and in Messina on the Sicilian side, with a large number of witnesses being found among the manuscripts which were previously kept in local monasteries (esp. the Archimandritate "del Santissimo Salvatore in lingua phari" in Messina) and which are now in the University Library in Messina. Unlike the previous scripts, this one has fairly specific characteristics which allow it to be recognized, especially the "modular contrast" between wide and narrow letters, as illustrated in the image below. It is notable that minuscule beta and mu are wide, while nu, which is morphologically similar, is consistently narrow, as are majuscule epsilon and eta. This modular contrast be considered an exasperation of a tendency which was already present in the script associated with the "school" of St. Nilus, with which the "Reggio script" indeed shares an overall "flattened" look as a result.

The earliest explicitly dated exemplar of this script is Vat. gr. 1646, copied by one Nikolaos (RGK 3,524) in 1118 CE (his rather illegible subscription is on f. 278v); most of the others are also from the twelfth century, though Canart & Leroy included even some late thirteenth-century and early fourteenth-century volumes in their list of "Reggio script" manuscripts.

For practice reading and transcribing "Reggio Style" manuscripts, please see Vat. gr. 2290.

Despite the sporadic presence of cursive elements (e.g. the alpha-gamma in line 5 of the "Reggio script" sample above), Italo-Greek bookhands are generally very formal. However, in twelfth-century Italo-Greek manuscripts there are a number of examples of more cursive bookhands, often incorporating elements which can be observed also in chancery scripts and notarial documents from the same period (we will have more to say about this in the next page; regarding the specific case of Italo-Greek documentary hands, see A. Bravo García's article in the proceedings of the conference held in 1988 in Erice, vol. 2, pp. 417-445). Some of these hands make an extremely generous use of abbreviations, including some abbreviations which are not usual outside of the Italo-Greek context, such as the sign for ἀπο which you can see in the sample below from Vat. gr. 1611 (fourth line, in the heavily abbreviated expression τὸ καλὸν ἀπὸ τοῦ κακοῦ); this manuscript was written, probably in Rossano (see S. Lucà's already mentioned article in RSBN 22-23 [1985-86], pp. 93-170:107) in 1116/17 CE by an anonymous scribe in an otherwise unknown "School of St. Peter" (it should be noted that the Italo-Greek origin of this manuscript is highly likely but not objectively attested). Others appear to be simply more cursive variants of typical Italo-Greek scripts, like the hand of one Bartholomew of Bordonaro near Messina (RGK 3,60), who subscribed the Messan. gr. 32 in 1150/51 CE (Lake 9,354) and whose hand has also been identified by S. Lucà ("Il Vaticano greco 1926 e altri codici della Biblioteca dell'Archimandritato di Messina", Schede medievali 8 [1985], pp. 51-79:54) in the Vat. gr. 1635 (also shown below). This Bartholomew uses a script which recalls the main features of the "Reggio Style" but incorporates a large number of cursive elements. Though "informal", this script is very elegant; the repartition of letters between the "narrow" and "wide" categories is not always the same as in the usual "Reggio Script", e.g. epsilon and sigma here sometimes join the ranks of the "wide" letters, as do delta (in its majuscule shape) and the cursive rho connected to the left.

For practice reading the cursive variants of Italo-Greek scripts, please see Vat. gr. 1611 and Vat. gr. 1635.

In 1978, Nigel Wilson demonstrated (in an article in Scrittura e civiltà 2, pp. 209-219) that a remarkable, illuminated manuscript of the Byzantine historian John Skylitzes (late 11th c.) which is now kept in Madrid (Matrit. Vitr. 26.2) and whose origin and date had long puzzled paleographers, was an Italo-Greek product of the twelfth century. There has been some controversy since then about whether it was produced in Palermo (this was Wilson's own proposal), in Messina (thus S. Lucà in ASCL 60 [1993], pp. 1-91:86) or perhaps even in Constantinople by scribes sent especially for the purpose from Sicily (thus B. Fonkič in Erytheia 28 [2007], pp. 67-89, who also provides a good summary of other opinions); but no one has questioned Wilson's basic conclusions, which were based on the close similarity between the script of the Madrid Skylitzes (especially in ff. 88-95 and 187-194) and that which is found in ff. 211-230v of the medical manuscript Vat. gr. 300, which can itself be shown to be an Italo-Greek product by a number of indications, including the presence of Reggio script on ff. 262-273, ruling lines in pencil, and some marginalia which we will not describe here. Wilson thought that the script in Vat. gr. 300 and in the Madrid manuscript were so similar that they might be the work of the same (anonymous) scribe, perhaps after an interval of a few years (you may judge for yourself from the two samples below). Since 1978, about a dozen more manuscripts have been noted which are written in this style (mostly by S. Lucà and by Fonkič; they are listed in the already-mentioned article by Fonkič, pp. 70-71), which is known as the "Skylitzes Type" in honor of the Madrid manuscript. Among its characteristics, most of which you may see in the samples below, are:

The last characteristic in the list above is perhaps the most unusual one; it may be seen in the sample from the Madrid Skylitzes, in the penultimate line in the expression εἴσω τειχῶν, and twice in the sample from Vat. gr. 300 (lines 5 & 6), in the expressions τῶν ὀφθαλμῶν and ὅρασις ἰνδαλμάτων.

For practice reading and transcribing this script, please see Vat. gr. 300.

(On to Informal and Scholarly Hands)