Читать книгу The Medieval Salento - Linda Safran - Страница 10

CHAPTER 2

Languages

One of the most important lessons to be learned from examining linguistic choices is that language, like names, is not a secure indicator of cultural or ethnic background. Speaking, reading, writing, and commissioning texts are learned behaviors whose use is socially determined. As numerous sociolinguistic studies have shown, different languages might be appropriate in different situations, and a person might have many reasons to commission or execute a text that was not in his or her ancestral tongue. In the medieval Salento, the relevant languages of inscription were Greek, Latin, and Hebrew; Aramaic, Old French, and pseudo-Kufic script also make an appearance. Griko, the local dialect of Greek, and Romanzo, the Romance vernacular that became modern Italian, were spoken languages that, before the end of the fourteenth century, very rarely intrude into written texts. The contemporaneous use, juxtaposition, and combination of these languages are among the features that gave (and continue to give) the Salento a unique regional character. While Jews and Christians could look back to a golden age of linguistic unity before the Tower of Babel was built (Gen. 11:1), postbiblical languages had long since become an index of diversity, a criterion for belonging to or being excluded from certain groups. In what follows, I argue that texts in the public domain were visible social statements that contributed in meaningful ways to the construction and communication of individual and communal identity.

Information about languages comes from a variety of public texts, some carefully planned and formally carved or painted, others informally incised or painted ad hoc. For Hebrew, this material evidence is supplemented by manuscript data. I discuss each language group in turn and consider linguistic patterns found in different types of texts. By “types of text” I refer not to medium but to the primary function of the inscription or graffito: hortatory, dedicatory, didactic, devotional, or funerary. Certainly there is overlap between these categories, but if a devotional text also asks readers to pray for the author, I consider it hortatory. Dedicatory and devotional texts can be very similar; in assigning an inscription to the former group, I look for verbs that stake a personal claim to such notable action as building or rebuilding a church or paving an entire floor. I include fieri fecit texts (so-and-so “had made”) in this group only if they demonstrate extensive patronage and include additional information. Simpler claims of patronage, artist’s signatures, deictic texts that designate or identify something, labels or captions that instruct, inform, or merely indicate the writer’s presence are all considered under the rubric of didactic texts. Devotional texts, or supplications, either request divine help or ask that the person named be remembered by God or, less commonly, by the Virgin or a saint. These are usually termed “votive” texts, but I want to remove the implication that a vow has been promised or fulfilled because of a complete lack of evidence that this was the case: “vow” is nowhere used in public texts in the Salento in any language.¹

Language Distribution

Although the types of medieval public texts are more or less the same everywhere, the linguistic map of the Salento is very different from that of the rest of Italy (it most closely resembles Calabria). The majority of the peninsula did not have ancient Greek colonies, two centuries of Byzantine rule, and an influx of medieval Greek speakers. Other regions of Italy do not contain a microregion of towns in which a Greek dialect is still spoken.² No other Italian province can claim so many medieval public texts in Greek.

The Salento also has a significant number of texts in Hebrew, and this does have parallels elsewhere in Italy. Nevertheless, even expanding the geographical and temporal parameters to offset the paucity of Jewish remains yields less than a handful of synagogue dedications and epitaphs from only five sites. From other sources we know of many more Jewish communities in the Terra d’Otranto, especially in the late medieval and early modern periods.³ There is no doubt that these communities maintained Hebrew as their sacred tongue and continued to produce religious and scientific manuscripts until the sixteenth century, but they are both geographically and materially underrepresented here.

The southern Salento (the province of Lecce) preserves more than twice as many Greek texts as Latin ones, whereas in the northern Salento (most of the provinces of Brindisi and Taranto) Latin inscriptions outnumber Greek ones by a ratio of approximately three to one. (The south also had two-thirds of the sites known to have a Jewish presence.) If my inscriptional and graffiti evidence is an accurate indication of a larger truth about spoken Greek and Latin, we would conclude that there were relatively more Greek speakers in the southern part of the Salento and relatively more Latin in the north. Indeed, some scholars have made more sweeping claims about the precise borders of a Greek-speaking south and a Latin-speaking north.⁴ But inscriptional language is not the same as speech and the surviving material evidence may not be representative of larger communities of speakers. Even a modest painted inscription or a poorly carved tombstone denotes a certain level of social or financial means and depends on the availability of skilled craftsmen and other factors. When language is a product of sociolinguistic choice, a patron might choose a less common tongue to reach a particular subset of his potential audience.⁵

Dominant and alloglot (minority) languages probably could be found in most sizable communities in the Salento, even if it is not possible to document this archaeologically. Considering only the precisely dated material evidence, we see an increase in the use of Latin for formal texts beginning in the twelfth century, particularly after midcentury, and Latin texts continue to outnumber Greek ones in the thirteenth century; in the fourteenth and early fifteenth centuries, however, the two languages are evenly distributed over the extant dated texts. In other words, while we might have expected the proportion of Latin texts to rise on pace with its increasing value as a language of power and administration—an increase that is apparent from documentary sources—dated Greek texts remained just as abundant. By contrast, the limited evidence for Jewish sites indicates that these were razed or reused by the tenth century (when Hebrew epitaphs disappear) or the thirteenth century (when synagogues are converted to churches at Trani) or later (late fifteenth century, when the synagogue of Lecce is destroyed). Such actions resulted in the loss of most Hebrew inscriptions even though manuscript and documentary evidence for continuous Jewish cultural activity survives.⁶

The proportions in which the various types of texts are preserved in Greek and Latin reveal some surprising differences. There are more Latin dedicatory inscriptions, perhaps because the latter tend to be epigraphs carved in stone and often still in situ in Roman-rite churches. On the other hand, there are three times as many devotional texts in Greek as in Latin, which may reveal something about differing expressions of piety—not that users of Greek were more pious, but perhaps that their public piety was more likely to be recorded in written form while Latin (actually Romance) speakers may have offered objects or images to their churches instead. Perhaps later Greek speech communities, such as that at Vaste in 1379/80 [157], had a greater need for visible public prayers than their neighbors because of the ever-declining numbers of Orthodox clergy and the infiltration of non-Orthodox church practices (see Chapter 6). If the early bilingual Hebrew-Latin epitaphs are excluded, there are eight times as many funerary texts in Greek as in Latin. Such a great disparity surely is due to multiple factors that probably include Orthodox funerary customs and local habits of display and imitation. Given the size of the sampling and absence of corroborating evidence, it is not possible to make claims about greater Greek literacy or financial clout. In the end, we cannot even be sure that differences in linguistic proportion are meaningful, as both absolute and relative numbers of texts are products of what has survived and what has been published or made accessible.

In addition to the texts in Hebrew, Latin, and Greek that figure in the Database and are discussed below, other languages were occasionally used for public texts. The former pavement of the Norman cathedral at Brindisi illustrated the story of Roland and labeled those scenes in French (Rollant, l’arcevesque Torpin) [21.sc].⁷ Kufic or pseudo-Kufic script is used to evoke Arabic at San Pietro at Otranto, Santa Maria di Cerrate near Squinzano, and elsewhere, but the comprehensibility of these texts was probably nil and their meanings nonverbal and abstract.⁸ There was an Armenian community at Ceglie, near Bari, beginning in 990,⁹ and two funerary stelae in the Salento record Armenian names, but the language of both is Greek [111, 159]. The use of these other tongues was very restricted, and we can seriously discuss the local population’s languages and literacy only in terms of Hebrew, Latin, Greek, and their associated vernaculars, alone and in combination.

The official imposition of a Tuscan variety of Italian since the nineteenth century has not erased traces of earlier languages and dialects in the Salento. The region’s linguistic picture seems always to have been complicated by the interaction of alloglot tongues with one or more majority languages. In general, the dialects called “Salentine” by linguists have many analogies with those in Sicily and southern Calabria, although they display greater lexical archaism and a significant admixture of Greek is apparent in semantics, phonology, and syntax. In the north, such sites as Taranto and Massafra have an Apulian dialect, with pronunciations different from those farther south and a linguistic system more akin to that of Naples.¹⁰ Except for this northern fringe, the linguistic map corresponds well to the entity I defined in the Introduction as “the Salento.”

Hebrew

Medieval Jews, or at least the Jewish intellectuals who led them, believed that the twenty-two letters of the Hebrew alphabet were identical with the Torah and with the name of God himself: God had used these letters, rotated and in combination, to create the entire world.¹¹ For Jewish mystics, the Hebrew language “was not regarded as a means or a tool, but as the subject and ultimate purpose of speculation.”¹² As noted in Chapter 1, Hebrew names were mandatory for Jews because Hebrew was the language spoken by God and his angels; even some Christian sources agreed.¹³ As was the case with keeping their names, maintaining the Hebrew language was one of the criteria for the Jews’ liberation from Egypt and its identity as a chosen people.¹⁴

Jewish prayers were almost all in the holy tongue, the lashon ha-kodesh, although a few prayers persisted in Greek for many centuries.¹⁵ A medieval Italian source permitted certain intercessory prayers to be recited in Aramaic, particularly those directed to the angels “appointed [as supervisors] over the gates of prayer” although not to those who serve as one’s “guardian angels” or constant spiritual companions.¹⁶ The same text indicates that it is a mitzvah—a positive commandment—to translate Torah readings into the vernacular, although whether this translation is to be done ad hoc or from a prepared translation is not clear. Although this vernacular proviso was already noted in the Talmud, its relevance to thirteenth-century Italy is evident from the author’s personal statement: “My opinion is that of my brother, Rabbi Judah, who says the principle behind the translation is to comment on the words of Torah for women and the ignorant who do not understand the holy language.”¹⁷ This halakhic (legal) text, Shibolei ha-Leqet (“Gleaned Ears”), was written in Rome in the mid-thirteenth century by Zidkiyahu ben Abraham ha-Rofe (“the doctor”), a member of the learned Anav family, which traced its origins in Italy to forced exile from Israel under the Roman rulers Pompey and Titus. The work is an invaluable source of information on many aspects of medieval Italian Jewish life and is cited often in this book.¹⁸

While Jews lived, worshipped, and died in southern Italy until the sixteenth century, material evidence for their presence consists mainly of funerary inscriptions that do not postdate the tenth century. Late antique epitaphs in the region consistently combined a Hebrew text with one in Greek,¹⁹ but in the seventh and eighth centuries Latin became a more frequent adjunct [81, 122–127, 131–133, 135]. All but one of these bilingual examples come from Taranto, where the overall epigraphic record is dominated by Latin even though Greek-speaking Jews and Christians also are attested there.²⁰ The latest funerary texts, which are usually assigned to the ninth or perhaps tenth century, use only Hebrew [10–14, 16–18, 121, 128–130, 134, 149–150],²¹ and in one case Aramaic [136]. It is risky to draw broad conclusions from such a small set of data, but I agree with Cesare Colafemmina that communication solely in Hebrew reflects an emerging sense of Jewish identity vested in language and manifested simultaneously in a flourishing of Hebrew literature that included the first historiographic work (Sefer Josippon), the first family chronicle (Megillat Ahima‘az), and astrological, philosophical, mystical, and medical treatises in addition to liturgical poetry (piyyutim).²² Perhaps this literary production had a trickle-down effect, resulting in a wider Jewish (male) literacy and more Hebrew epitaphs. The religious texts and poetry produced in Apulia were known outside the region: several piyyutim became part of the Ashkenazic prayer book, and a scholar in twelfth-century France remarked that “from Bari comes forth the Law, and the word of God from Otranto.”²³ The preponderance of Hebrew-only tombstones may also indicate that Jewish graves became less visible to non-Hebrew readers as the centuries progressed. The Christian and Jewish cemeteries at Taranto were contiguous,²⁴ but later Jewish burials may simply have been farther from the neighboring Christian ones and so had a more limited and homogeneous audience. It seems likely that the bulk of Jewish tombstones at Taranto were used to rebuild the city walls after the tenth-century Arab raids [139],²⁵ but the fate of later Jewish epitaphs there and elsewhere is unknown.²⁶ Were Salentine Jews not permitted such public displays? Did they lose the epigraphic habit for other reasons? Or were all of their tombstones after the ninth or tenth century simply repurposed by Christians for either practical or ideological reasons? Unfortunately, we lack answers to all these questions.

Among the formulas used to commemorate the Jewish dead are the Hebrew po schichvat (here lies), po yanuach (here rests), and mishkav (tomb of). In the Latin parts of the Jewish epitaphs, Hic requiescit (here rests) predominates.²⁷ The deceased is sometimes remembered positively as bene memorius, in Hebrew b’zikaron tov.²⁸ An invocation for shalom al minuchato, “peace upon the resting place,” echoing Isaiah 57:2, is very common; it is equivalent to the Latin Sit pax in requie eius or, in one case, Sit pax super dormitorium eorum [123.B]. “Amen” often concludes a short Hebrew funerary text. Often the deceased is noted as a righteous man whose memory merits a blessing, zecher tzaddik livrachah, from Proverbs 10:7 [12.D, 14, 124.B, 126.B, 128.B]. This is the most common biblical citation or paraphrase, although a few epitaphs quote from Psalms, Job, or Isaiah [18, 124.B].²⁹ Overreliance on phrases drawn from the ancient funerary ritual of the land of Israel led the composer or the carver of Leah’s epitaph in Brindisi [16] to conclude that text with a line from Song of Songs in which the male gender of the original has not been amended for a female commemoration.³⁰ Similar errors of gender are not uncommon, as in a funerary inscription at Taranto [121], where the verb “rests” is masculine even though the commemoration is for an unnamed wife.³¹

All-Hebrew epitaphs tend to multiply scriptural and liturgical references, and two examples in Trani abbreviate the common prayer “May his soul be bound in the bond of life” (1 Sam. 25:29) [149, 150], echoing the earlier abbreviation in Brindisi of “The holy one, blessed be he” [16]. Such complex textual referencing and abbreviations are indications of Hebrew literacy that serve to advertise and solidify Jewish identity through language. Some epitaphs are notable for their relationship to contemporary Hebrew liturgical poetry [18.A], and there are rhyming lines as well as acrostics that yield the name of the text’s author.³² This is the case at Oria [81], where “Samuel” may have composed the commemoration for his mother, Hannah, if indeed this eighth-century pair deliberately echoes the well-known biblical mother and son of 1 Samuel 1–2.³³ Not only was the Oria inscription composed by and for a Jewish patron, but it seems to have been carved by a native Hebrew speaker as well. He mistakenly began the Latin text at the right rather than the left and had to correct himself; the letter N is consistently rendered with a backward diagonal;³⁴ and ES at the beginning of line 3 is a meaningless repetition of the letters immediately above [81.B].³⁵ In most of the Hebrew inscriptions the quality of carving is quite good, and in the bilingual epitaphs the Hebrew text tends to be better than the Latin, carefully incised and with more regular letters. Were two different carvers employed, or does a Jewish craftsman betray here his greater familiarity with Hebrew? In a seventh–eighth-century stela from Taranto, the Latin epitaph omits a syllable and reads benemorius rather than benememorius, although the correction is inserted above the line [133.C]; in another, requiescit is misspelled.³⁶ Yet the only Latin error in a four-line seventh- or eighth-century epitaph from Taranto [123.B] is suum for suo; the unusual preceding word, barbane (uncle), is a third-declension ablative from barbas, here rendered properly.³⁷ When both languages are executed well, it is impossible to link the carver’s competency to his religious affiliation.

Another important body of information about the use of Hebrew in medieval southern Apulia comes from inscriptions originally placed inside synagogues. The texts from Gravina [50] and Bari [9] were introduced previously because they provide onomastic data. Two other synagogue inscriptions, devoid of names, survive in Trani [147] and Lecce [56]. Despite its brevity, the latter is especially valuable as the only physical testimony for a synagogue building in the Salento proper. A Jewish community in Lecce is attested by 524, and because this particular edifice was not transformed into a church (dedicated to Santa Maria Annunziata) until 1495,³⁸ it testifies to the endurance of Salentine Jewry.

These synagogue inscriptions differ from the majority of Jewish epigraphs because three of them bear specific dates (1184/85, 1246/47, 1313/14), given in years since the creation of the world. The epitaphs communicate the age of the deceased and not the date of death. Unlike Christian inscriptions, Jewish ones pay no attention to such specifics of dating as day, month, ruler, or indiction. The three dated synagogue dedications specify the extent of the construction or renovation for which the donor was responsible. Trani’s synagogue may have been built entirely by one unnamed individual, but the patron commemorated at Bari gave only a window and his counterpart at Gravina provided the pavement and seats. Such parceling out of donations implies communal effort and has its roots in the Jewish synagogues of late antiquity, where the mosaic decoration was often credited to many individuals.

The synagogue texts occasionally include both non-Hebrew loan words and nonstandard orthography. For instance, the word for benches or seats, iztabaot, used at Gravina [50] and Trani [147] comes from the Greek stibadion, here spelled two different ways .³⁹ In the Bari inscription [9] the date is communicated by a gematria: adding the numerical values of the Hebrew letters in the word “Hadasah” gives the year. That this word should be read numerically and not literally is indicated graphically by small “v” signs over each letter (the symbolic implication of Hadasah/Esther, who saved the Jews of Persia, was doubtless present as well). This use of letters for numbers is common in Hebrew manuscripts and is found on a tombstone from Trani, where it is used to indicate the year of death [150]. This is the reverse of the Greek practice of isopsephy, in which numbers are used to indicate letters: in the Salento, three or four eleventh-century Greek devotional texts conclude with the number 99, whose isopsephic value is “Amen” [32.E, 32.F, 33.A, 80.A].⁴⁰

Even formal texts seem to betray the influence of spoken language. Two such elements are present in the Latin on the Oria epitaph [81.B], where the Italianate G replaces the Latin I at the beginning of the deceased woman’s father’s name, Julius, which ends in the genitive -u.⁴¹ In one Taranto inscription [123.A], Silanus ends with a vav (hence Silano or Silanu), probably reflecting that the final -s was disappearing in pronunciation at this time, a feature typical of southern Italian dialects.⁴² Iotacisms that reflect current speech are common in all of the languages used for Jewish as well as Christian texts.

In a manuscript of the Mishnah produced in Otranto circa 1072 and now in Parma, glosses written in the vernacular language but with Hebrew characters clarify which plants cited in Mishnaic Hebrew could not be grown alongside others.⁴³ This vernacular would not reappear in local texts until the late fourteenth century. Shabbetai Donnolo’s pharmaceutical terms provide tenth-century evidence for a distinctive Salento dialect, but because Donnolo’s scientific terms are very similar in Greek, Latin, and proto-Italian it is difficult to know which language they represent (the absence of many final consonants supports the vernacular).⁴⁴ A sampling from the Otranto Mishnah reveals the unambiguous Greek sources of many terms, including klivanidt, from Greek κλιβανίτης, a kind of bread, and savani, from Greek σάβανον, a linen cloth.⁴⁵ Robert Bonfil’s investigation of two South Italian Hebrew chronicles, Megillat Ahima‘az and Sefer Josippon (written outside the Salento), concluded that their authors were actually thinking in Latin or the Romance vernacular even when writing in Hebrew; because they were used to speaking those other tongues, it infiltrated their texts.⁴⁶

By the medieval period, the use of Greek in Jewish liturgy had largely been replaced by Hebrew.⁴⁷ That Hebrew was used for praying and writing facilitated interaction with faraway Jewish communities and was an important marker of Jewish difference from Christianity. Yet as a sacred language, and one known only to men, it could not be used generally for speaking. Romaniote (Byzantine) Jews, including those in the Salento, spoke a hybridized Greek, not Hebrew.⁴⁸ In the ninth century, the Jews of Venosa (in Basilicata) needed a translator when a scholar visiting from Baghdad delivered a Sabbath sermon in an unfamiliar language that was probably Hebrew.⁴⁹ However, in the public disputation circa 1220 between the Jews of Otranto and Nicholas-Nektarios, abbot of the Orthodox monastery at Casole, the abbot records that the Jews conferred among themselves in Hebrew.⁵⁰ Nicholas-Nektarios knew the language—on occasion he even wrote Hebrew in Greek characters⁵¹—so he probably was correct about what he heard. Perhaps the simplest explanation is that the most learned local Jews could converse in Hebrew, or at least quote from written sources, but this was not a widespread phenomenon; medieval Hebrew was for writing and worship, not for speech. In later chapters I suggest that the most learned Jews also dressed differently from their coreligionists and engaged in certain ritual practices that “regular” Jews did not. They were, in effect, equivalent to such Christian role models as abbots and bishops, whose standards of behavior and learning were different from those of most Christians.

Latin

Latin was used in the Salento for a large number of dedicatory inscriptions, a few epitaphs, and two kinds of public expression unattested in Hebrew: painted or incised devotional inscriptions and hortatory texts.⁵² Except for the Latin faces of ten bilingual Hebrew epitaphs from Taranto and early dedicatory or didactic inscriptions from Oria [83, 84] and Brindisi [20, 25], the Latin texts can be dated between the twelfth and fourteenth centuries. As with Hebrew, Latin public texts are far outnumbered by surviving examples in Greek.

The dedicatory inscriptions vary widely in scale and include commemoration of whole churches or monasteries [1, 21.A–B, 57, 58, 144], roof beams [78.B], an altar [38], pavements [86.A–G, 140.C–E], and individual wall paintings [78.C], in addition to partial renovations of or additions to existing buildings [35]. A Byzantine claim to have rebuilt the entire city of Brindisi from its foundations stands out for its hubris [20]. Over 60 percent of these inscriptions give the year of the dedication with Anno Domini or Anno Dominice, year of the Lord, or Anno ab Incarnatione, year since the Incarnation. The date is usually supplemented by additional elements and highlighted by its placement in the first or last line. In two-thirds of the dated inscriptions there is a reference to the current king and/or the local lord, complete with titles; in several cases the month or day is noted, and even more often the indiction [1, 28.W, 38, 79.A, 86.E, 117.A, 144]. This imprecise but widespread system of dating according to fifteen-year cycles was much more common in Greek public texts. Its use in Latin probably indicates awareness of the Byzantine indiction, which began on September 1, but may also indicate familiarity with other kinds of cyclical reckoning, with different starting dates, used irregularly by the papacy, notaries, and others. This method of dating was falling rapidly out of fashion in Europe by the fourteenth century, but it endured much later in the Salento with its strong Byzantine ties [79.A].⁵³

Patrons inscribed dedications in Latin for many reasons. They rarely cite as motives the remission of their sins [1.A] or eternal life [58.A]; more often the dedication is simply to honor Christ, the Virgin, and various saints [1, 38, 57, 67.B, 78.C, 144]. The phrase ad (or in) honore Dei [144] is an echo of the liturgy.⁵⁴ Sometimes no reason is given; perhaps generic piety was motive enough. On the column base at Brindisi, the dedication seems to be in honor of the “magnificent and benign emperors”; unfortunately, the text is incomplete [20]. Pious donations of goods (bonorum, donis) and serfs (colonis) were publicly recorded at Castro [35] and Lecce [58.A]. Of the Latin dedicatory texts that preserve information about their patrons, a clear majority were high-ranking ecclesiastics or priests, alone or in combination with lay donors, followed by feudal lords, then architects or artists.⁵⁵ Two of these texts were in Leonine hexameters, the metrical form underscoring the episcopal patronage [57, 78.C].

In contrast with the dedications, devotional and hortatory texts in Latin rarely preserve a date of any kind. An exception is at Statte in 1416, where Nicholas Bertini, a traveler from northern Italy, left his name and the date of his visit to nearby Taranto [117.B].⁵⁶ This kind of specificity fits the late date of this graffito, incised at a time when ideas about individualism were taking hold in northern Italy. In the Salento, too, dates are included in many postmedieval graffiti not included in my Database. Medieval visitors, by contrast, wanted to transcend the specifics of their recorded presence and solicit eternal favor or remembrance.

With the notable exception of incised graffiti, devotional invocations almost invariably begin with an abbreviation of the vocative Memento Domine, “Remember, Lord.” This is followed by “your servant,” either famuli tui, the correct genitive case, or the dative famulo tuo, which by the Middle Ages is a much rarer form. The dative is used disproportionately in Salentine Latin texts, however, either because of a contamination from the Greek or an attraction to the case of the patron’s name in the vernacular. In any case, the meaning of the inscription would have been, and remains, clear. The priest Sarulus who dedicated images of Saint Nicholas and Saint Margaret in two different crypt churches at Mottola must have been unperturbed when artists wrote famulo in one and famuli in the other [75.A, 76.C].

Latin devotional graffiti occur in large numbers at only two sites, San Marco in Massafra and Santa Lucia in Palagianello [66, 94]. These differ from more formal texts in several ways: they are incised irregularly, rather than painted or carefully carved; they are often much shorter and may be superimposed; some are in minuscule, or in a combination of capital, uncial, and minuscule letters; and they very rarely use the formal Memento Domine formula. Instead, most Latin graffiti begin with Ego, “I am,” followed by a name or names. These two sites, along with a third [116.A] that is also in the province of Taranto, contain the only Latin hortatory injunctions in the Salento. The overt request that readers of these texts pray for the person named therein was directed to Latin readers such as the monks, priests, and other clergy who identify themselves by profession in didactic Latin graffiti at the same sites.

In Santa Lucia (the former Trinità) at Brindisi, a painted text combines a devotional introduction, Memento Domine, with “rest in peace,” ending with “Amen” [27.A]. Given its location low on the left wall of the church, there is little doubt that the inscription marked a family burial place. At San Paolo, in the same city, a painted “Here lies” (hic iacet) records the tomb of an elite named man originally from Florence [26.A], while a carved marble plaque marks the burial of an elite unnamed woman [26.C]. The metrical inscription of the latter adds that an altar and a joint tomb have been prepared nearby by her devoted husband, Nicholas Castaldo. A now-lost sarcophagus lid for a magister states “Here lies the body” (hic jacet corpus) and warns of excommunication by the archbishop for anyone disturbing the tomb [31]. Like the Brindisi epitaph the lid gave a specific date, and this seems typical of elite commissions.

If we consider not only epitaphs and dedicatory and devotional inscriptions but also the didactic explanatory labels that accompany wall paintings, we find some evidence for the Latin literacy of a few painters. At Acquarica del Capo [1.A], where a long dedicatory inscription in Latin is signed by two artists in Greek, the caption for Saint Hippolytus on horseback is labeled both in Greek, “Ο ΑΓ(ΙΟC) ΙΠΠΟΑΥΤ(OC),” and in Latin, “S(AN)C(TV)S VPOLIT[VS].” The initial U (as V) in the Latin legend betrays the hand of the Greek painters.⁵⁷ In the Saint Catherine cycle in the vault at Casaranello, which is entirely Latin in its captioning, the evil emperor is mislabeled once “MANSENCIUS” (for Maxentius), and “AGVSSTA” (Augusta) is also misspelled.⁵⁸ This is sloppy Latin and hardly indicative of the artist’s better Greek if he was trained in France, as scholarly consensus has it.⁵⁹ Vernacular pronunciation probably influenced the spellings of “CATERINA” (T for TH), “MASENCIUS” (S for X), “PORFILIUS” (L for R), and “IMPERATRICS” (CS for X).⁶⁰ Similarly, at the Candelora crypt in Massafra, Saint Stephen is identified as “STEFANUS” (F for PH) and Nicholas the Pilgrim as “PELLEGRINUS” (LL for R) [63.B].⁶¹

Greek

The impressive quantity of Greek texts produced in the medieval Salento gives the lie to Nikephoros Gregoras’s lament that by his time, the first half of the fourteenth century, nothing remained of Greek poetry or spoken language in the ancient Magna Graecia: καὶ οὐδὲν ἔτι ἴχνος ἐλλέλειπται μὴ ὅτι γε μούσης ἀλλ’ οὐδὲ διαλέκτου .⁶² Greek public texts of a funerary and devotional nature greatly outnumber Latin and Hebrew ones, but there are fewer dedicatory and didactic inscriptions in Greek than in Latin and only four hortatory texts, [43.A, 48, 114.B, 156.A], which are both dedicatory and funerary. The Greek dedicatory inscriptions commemorate a village [48], walls [139], a tower [30], hospitals [4, 46], a ciborium [114.C], and several churches [36, 43.A, 108.A, 109.A, 113.A, 146.A, 154.A]. A church (ναός) is invariably all-holy (πάνσεπτος), except for one humble δόμος (house) [154.A].⁶³ The usual Greek term for constructing, ἀνοικοδομέω, is preferred, albeit often misspelled, but κατασκευάζω and ἀνηγέρθη are other options. Decorating church walls with frescoes is communicated by ἐζωγραφήθη or ἀνιστορήθη, which might be done at the “efforts and expense” (κόπου και ἐξόδου), “cooperation and effort” (συνδρομὴς και κόπου), “labor and travail” (πόνου και μόχθου), or simply the “expense” (δαπάνης) of the patron. Only three of the dedicatory inscriptions in Greek include the name and titles of the secular ruler, far fewer than in the Latin texts and good evidence that most patrons who chose to be remembered in Greek were disinclined to use a method of dating that highlighted a European ruler after the Byzantines were evicted circa 1071. The date is regularly included at the beginning or end; the month is noted three times, the day and hour once each.⁶⁴ Few of the Greek dedications share the concern with precise details of dating seen in Latin dedicatory texts; most are content with the year since the creation of Adam (like the Hebrew, and unlike the Latin post-Incarnation dates) and the indiction. Reference to the indiction was much more widespread in Greek and used by a larger group of patrons from different social classes.

As was also the case in Latin, almost all of the Greek devotional texts begin with the injunction “Remember,” Μνήσθητι, directed to the Lord (Κύριε). Only in a single late monument is the addressee Christ [157.G], the Virgin (δέσποινα) [157.A], or an unspecified saint, presumably the one depicted adjacent to the relevant text [157.I, K].⁶⁵ With very few exceptions, the supplicants identify themselves as “servants” (δούλοι) of the Lord, just as they did in Latin. The only other devotional formula is “Lord, help your servant” (Κύριε βοήθει) [53.C, 64.A, 66.G, 71.A(?), 94.D, 143.C–E, 154.A, 155.B (only 154.A is a dedicatory text)].⁶⁶ The overwhelming popularity of Μνήσθητι Κύριε is not paralleled in other regions where abundant Greek inscriptions are found, including Cappadocia, Greece, and the Balkans.⁶⁷ There the usual invocation is either Κύριε βοήθει or δέησις τοῦ δούλου τοῦ Θεοῦ, “petition of the servant of God,” and the latter is nonexistent in the Salento. The verb μιμνήσκωμαι rarely appears in public texts outside of Italy, where its popularity probably derives from its use in the liturgy. Salentine diptychs of the dead endlessly repeat the formula “Remember, Lord, your servant(s),” and from the mid-fourteenth century onward the local euchologia (service books) were supplemented by an amplification of the anaphora unknown in other regions that also begins with Μνήσθητι Κύριε.⁶⁸ There may also be a connection with the identical Latin phrase “Memento Domine,” used in the Roman-rite canon of the mass.

The majority of funerary texts in Greek also use a repetitive formulation, Ἐκοιμήθη ὁ δούλος (ἠ δούλη) θεοῦ, literally, “The servant of God has fallen asleep” but usually translated as “The servant of God died.” In the devotional texts variety was introduced only in the fourteenth century, but in the epitaphs variety preceded standardization: in the tenth through early twelfth centuries there were still such phrases as ‘Υπὲρ κοιμήσεως καὶ ἀναπαύσεως (“for the sleep and repose”) [159], ἀπόθανε (“died”) [33.K],⁶⁹ Ἐνθαδε (“here lies”) [111], and Ἔνθα τέθαπται (“here is buried”) [32.J]; μακαριώτατος (deceased) is also used once [33.E].⁷⁰ These recall the semantic range found in contemporary and earlier Hebrew epitaphs.

Unlike the liturgical diptychs, which often commend the deceased to the bosom of Abraham, Isaac, and Jacob,⁷¹ very few epitaphs in Greek add such injunctions as “help your servant” or “pray for him.” They do, however, supplement the declarative statement of death with multiple details about the date. In this they resemble Greek and Latin dedicatory inscriptions more than devotional texts.⁷² The year of death is always given according to the Orthodox calculation from the beginning of the world. Only a few epitaphs cite the indiction, but more give the month and date and some even the day of the week or the precise hour of death [37, 99.A, 114.A, 137.A, 156.A].

The Greek used in all types of inscriptions and graffiti ranges from excellent to crude. Every possible iotacism is found: η–ι, η–ε, η–ει, ει–ι, ι–οι, and others (for example, μη for μοι [143.D]), and much exchanging of other vowels (ω–ο, ω–ου) and consonants (θ–τ) as well. The large number of undeclined names in public texts—Βενεδίτους at San Vito dei Normanni in the late twelfth century [109.A], Μαργαρίτα at Vaste in the fourteenth [157.I]—suggests vernacular influence.⁷³ Some of the damned in the Last Judgment at Santo Stefano in Soleto are identified in Greek (Ο ΠΛΟΥΟΙΟ, NECTOPHO) but consistently omit the final S [113.B].⁷⁴ In the late fourteenth-century hospital dedication from Andrano, where ξενόνας ίτη σπητάλη is a vulgarization of the classical ξενών (hostel), it was necessary to signal its equivalence (ἤτοι, “that is” or “in other words”) to the Italian spitali (from Latin hospitale) [4.A], which must have been the more familiar term.⁷⁵ At the same time, the Greek epigraph over the right door at Santa Caterina in Galatina identifies the church, or part of it, as a kappella, Italian for “chapel” but with the more Greek-looking orthography of k instead of c [47.A].

In a devotional text from Vaste, the priest George is identified as son of Lawrence, and either he or his father is an ὁβφέρτος, probably an oblate,⁷⁶ of Saint Stephen, to whom the rock-cut church was dedicated [157.G]. In the elegant inscription by George of Gallipoli for a liturgical candelabrum [49], the Latin “patron” (πάτρωνος) is used, probably for metrical reasons, in lieu of available Greek terms.⁷⁷ Perhaps the best example of vernacular impact in a Greek linguistic context is the sundial outside Santa Maria della Strada at Taurisano [145]. It has a Greek caption, Αί ὥραι τής ἡμέρας, but these “hours” are the Latin liturgical ones—Prime, Terce, and so on—identified here by their first letter rendered in Greek.⁷⁸ The language of this sundial is bilingual, both literally and metaphorically.

Bilingualism

When Hebrew and Latin texts are juxtaposed in a single Jewish tombstone, the information in each text may differ minimally or substantially. In one stela at Taranto [123.A–B] the two epitaphs are very similar; they share the desire to highlight the important names, ensuring that they appear at the beginning or end of a line. Yet in another example from the same city, only the Latin side communicates the dead man’s age and his father’s name, while the Hebrew side combines excerpts from a psalm and a proverb [124]. The stela from Oria is similar [81]: the Hebrew face quotes from the ancient funeral liturgy, praises Hannah as a wise woman, and affects literary pretensions with its rhyming lines and initial acrostic that reveals the author’s name, Samuel. Only in the Latin text on top is the deceased woman given a title (“Lady”) and her father named with an abbreviated honorific (“R,” for “Rabbi”). It seems clear that this information was directed toward different audiences, not only in terms of literacy but also in content: titles in Latin, liturgy and poetry in Hebrew.

Hebrew bilingualism seems to end before the ninth century, but different types of Greek-Latin bilingualism are much more numerous and of longer duration. The absence of visual bilingualism except by Jews before the twelfth century indicates a high degree of language exclusivity by the people involved in commissioning and executing wall paintings and inscriptions. I have identified three major types of bilingualism in the public monuments of the later medieval Salento: intrasentential language mixing, which occurs within a single sentence or text; intramonumental mixing, in which a minority of texts are rendered in a language different from the one used extensively in a single monument; and fully bilingual monuments.⁷⁹ Linguists have shown that such “code switching”—embedding a syntactic unit (a letter, morpheme, word, or sentence) in one language in a different matrix language—was a socially meaningful act.⁸⁰

A handful of monuments exhibit intrasentential language mixing, in which an inscription, or titulus, begins in one language and ends in, or is interrupted by, another.⁸¹ An example is the titulus that identifies John Chrysostom in the apse of Santo Stefano in Soleto in the fourteenth century [113.sc.2]. It reads “S,” the Latin abbreviation for “Sanctus,” and “Ιω,” the beginning of John in Greek.⁸² A longer intrasentential text is the dedicatory inscription at Acquarica del Capo [1.A]: the eight-line paragraph begins in Latin but concludes with the painters’ names in Greek. The artists shifted to their native tongue to record their statement of authorship once the text assigned to them by the patron was complete. Code switching of the modest intrasentential sort may be unplanned, but it is not uninformative: it indicates that artists had some familiarity with a second language.

Monuments that manifest intramonumental mixing are the largest group. In these, one or two discrete texts are written in a language different from the one that dominates in that site. This group can be divided into three subtypes according to the kind of texts involved: didactic tituli; devotional and dedicatory inscriptions; and sacred speech acts. In the first subset, one or two short tituli interrupt an otherwise monolingual pattern of figural identification. Nicholas, who scarcely needed identification, received a double titulus more often than any other saint, although in the crypt church dedicated to Michael at Li Monaci it is the archangel and John the Evangelist who received the double label, done by the same hand [43.st.1]. As in this case, only rarely is the saint with the extra titulus the titular saint,⁸³ so the double name is not warranted by communal significance; nor is the doubly labeled figure the name saint of any patrons or supplicants or artists whose inscriptions survive (there is a low correspondence in the Salento between supplicants and homonymous saints).⁸⁴ Most likely the special tituli were connected with an individual’s desire to advertise personal devotion by visual means: more text attracted more attention because of the power of script itself. Sacred figures were made more powerful and more insistently present by graphic and epigraphic means.

A second type of intramonumental language mixing occurs when the lessused language is employed for a devotional or dedicatory text. Hence a layman and a Latin text are inserted into the otherwise entirely Greek program at the Orthodox monastery of Santa Maria di Cerrate [114.E–G; Plate 15]. Clearly, the Lord accepted prayers in both Greek and Latin, and language did not preclude someone from patronizing a church of a different Christian confession. The third type occurs when code switching indicates a speech act by a holy person. In the thirteenth-century Annunciation scene in San Pietro at Otranto, the titulus for the archangel Gabriel is in the church’s dominant language, Greek, but his salutation to the Virgin is rendered in Latin.⁸⁵ The open Gospel book held by Christ at several sites displays text in a language different from the majority inscriptional language in the church. It serves to distance him from the local speech community, and a few individuals likely reaped social benefits from this change of code: the local priest may have been seen as bridging the gap between sacred and secular because of his (relative) scriptural literacy, and the patron’s status would only increase if the language of the sacred citations were also his own public language. This was certainly the case with John of Ugento at Acquarica [1.A].

Only a few sites in the medieval Salento can be termed “bilingual monuments.” In these cases there is no dominant language; both Greek and Latin are used extensively and contemporaneously. Perhaps the best example is the dated crypt at Li Monaci, where the long dedicatory inscription is in Greek [43.A], two saints are identified in both Greek and Latin [43.st.1], the protagonists in the Annunciation and the Crucifixion are labeled only in Latin, and the Crucifixion scene in the apse has the titulus “VICT[OR] MORTIS” [43]. The date in the dedicatory text (1314/15) applies to all of the extant images on stylistic and paleographic grounds. The French-surnamed patron employed a father-and-son team of painters, one of whom used artistic models intended for Roman-rite places of worship even though this site was intended for Orthodox use: it contains an exclusively Orthodox saint, Onouphrios, in the left corner. In the left apse niche, an elderly John the Evangelist—the Byzantine type, not the youthful evangelist favored in European art—holds a Gospel book that has around its edge Ἐν ἀρχῇ ἦν ὁ λόγος (“In the beginning was the Word,” John 1:1) [43.st.1]. Above him, a partly preserved fragment of a fish is the one that swallowed Jonah, identified by the letters omicron plus a superimposed pi and rho, the beginning of ho prophetos, “the prophet,” in Greek. The connection between Jonah and John is both liturgical and textual. In the Orthodox liturgy the book of Jonah was read at Vespers on Holy Saturday while John 1:1 was read the next morning, Easter Sunday. The text of Jonah begins “The Word of the Lord came to Jonah” and John’s Gospel begins “In the beginning was the Word.”⁸⁶

Adjacent to Jonah and John at Li Monaci is the Annunciation, done by a different hand and labeled in Latin [43]. Several Byzantine hymnographers linked Jonah with the Virgin, including an eighth-century ode by Cosmas the Melodist that also references the Logos, as on the Gospel book held by John.⁸⁷ Farther to the right is the apse Crucifixion with its exceptional titulus referring explicitly to Christ’s victory over death rather than to the usual “Jesus of Nazareth King of the Jews” [43.A]. Such an explicitly salvific title reinforces the message of the Jonah image and strongly suggests a funerary function for the church. The presence of the Virgin and John flanking the cross brings the image in line with the deesis imagery more typical of apses in the Salento. At Li Monaci, then, the languages serve to introduce the complexity of the cultural space.

Literacy

The preceding discussion of bilingualism raises the question of literacy, a term that is notoriously difficult to define. Does it imply reading complex texts, or only recognizing one’s name? Does it require the ability to write or only to read? How many spelling or grammatical errors are permitted before one’s supplication is considered illiterate? However it is defined, there are no studies of medieval literacy that focus on public texts in the Terra d’Otranto. Among private texts, historians who have studied notarial documents identify as illiterates those parties or witnesses who cannot sign their names and instead make the sign of the cross.⁸⁸ In Taranto, the notarial acts in Greek from the tenth to thirteenth centuries reveal approximately 70 percent literacy, 27 percent semiliteracy, and only 3 percent illiteracy.⁸⁹ At the same time, some of the monks at the important Orthodox monastery of Saint Nicholas at Casole could not read Greek.⁹⁰ A study of Latin usage in medieval Bari found a high level of literacy among upper- and middle-class laymen and a lower level among ecclesiastics, two-thirds of whom did not surpass the elementary level.⁹¹ While functional literacy seems to have been more prized in Byzantium than in Europe, this was no longer true, at least in Italy, by the thirteenth century.⁹²

Jack Goody wrote, “Where writing is, class cannot be far away.”⁹³ Literacy was a source of social power, and the very fact of an inscription connoted status; this was true in antiquity and it remained true in the Middle Ages and beyond.⁹⁴ The literate, or those who could afford to fabricate evidence of it by commissioning texts, dominated the illiterate majority through the power of the word. “Monumental texts may exercise power through their location in space and the way they look,”⁹⁵ and many of the local dedicatory inscriptions—solid, framed blocks of words—were inscribed permanently, or at least durably, in prominent interior and exterior spaces. Most are in a church apse or over a doorway, obvious focuses of viewer attention. In this way messages and status were broadcast widely. The multiplication of monumental texts, such as the repeated strips of information in the Otranto mosaic pavement [86.C–G], make grand statements that attract the eye, as do texts supplemented by figural imagery. Devotional and funerary texts are rarely as long, as prominent, or as sizable as dedicatory inscriptions, but when they are disproportionately large or noticeable they command proportionally greater attention. Anyone entering the Carpignano crypt would be drawn to Stratigoules’s burial site, which has the longest funerary text in the Salento in any language and a unique arcosolium setting [32.J]. With its composition underscored by color, its intimate physical connection with sacred figures on the intrados who shelter the tomb, and its literary pretensions (regardless of its dodecasyllabic defects),⁹⁶ this text for a dead boy mostly helped his father stake his place in the local community, despite his modest title of σπαθάριος.

Linguistic Identities

La‘az is the generic term in Hebrew for a non-Hebrew language, including one that a Jew might speak.⁹⁷ In the Arukh, a Hebrew dictionary compiled in Rome circa 1100, la‘az is glossed with the vernacular barbaro—stranger, non-Jew, from the Greek term for “barbarian.” It is a cultural term as well as a linguistic one. Three centuries later, in an extensive glossary compiled by Judah Romano in Rome to clarify difficult terms in Maimonides’s Mishneh Torah, the meaning of la‘az is again limited to “foreign” language, probably synonymous with “vernacular.” Other terms are introduced here as well: Latino refers to someone who does not speak Hebrew, but likely refers to Romance rather than Latin.⁹⁸ In the Arukh and elsewhere, the expression lashon romi does not mean “speech/language of Rome,” which would be a literal translation referring to some form of Latin, but rather “a hybrid language, strongly graecicizing, perhaps that spoken in the central-southern regions of Italy under Byzantine influence.”⁹⁹ Rom in medieval Hebrew sources is Byzantium, the medieval Roman empire and not the ancient imperial city; Byzantine (and South Italian) Jews were Romaniotes. In general, the non-Hebrew lo‘azim, including Italian volgare, were derided by learned Jews as notzrì, Christian (from “Nazarene”), and this attitude meant that it was not used as a creative literary language by Jews before the sixteenth century. Nevertheless, short glosses in medieval texts indicate that the Italian Jews were developing their own written and spoken vernacular, which we now call “Judeo-Italian,” based on vulgar Latin.¹⁰⁰ Jewish women, and not a few men, would have profited from translations of the Scriptures and the prayer book written specifically for them in the thirteenth and fourteenth centuries.¹⁰¹

Greek speakers are found in today’s Salento only in nine communities south of Lecce, but the medieval Hellenophonic zone was much larger.¹⁰² The language spoken in the Grecìa salentina is a particular dialect called by its users griko or grika: “milume grika” means “we speak grika.” This word is not Greek; it corresponds neither to Latin graecus nor ancient Greek γραικός. Gerhard Rohlfs suggested that it was the term that the ancient South Italians, speaking an Italic language related to Latin, called their Greek-speaking neighbors in Magna Graecia.¹⁰³ It remains unclear whether the dialect represents a survival of ancient Greek or a medieval phenomenon.¹⁰⁴ Regardless of its antiquity, the term and the dialect survived through and beyond the Byzantine period.¹⁰⁵ After unification in 1861, when Italian was imposed as the country’s official language, griko speakers in the Salento continued to use their traditional dialect among family and friends, while the Romance vernacular was used for everyday business and Italian only for official matters and largely unavailable higher education.¹⁰⁶

Griko is illuminated by considering its use among Jews. After being expelled from Spain in 1492 and from the Salento by the Spanish rulers in 1541, the affected Jews went mostly to Thessalonike, part of the Ottoman Empire, or to Corfu, under Venetian rule. In both places they found Romaniote (formerly Byzantine) Jews, Italian Jews from Rome, Ashkenazim, and Sephardim practicing their distinctive liturgical rites. Visitors to early modern Corfu record that there were communities of Jews of diverse origin that included both gregi—Jews from the Salento who spoke griko—and others from Apulia who used pugghisu, “Puglian,” the Salentine Romance vernacular.¹⁰⁷ These communities had names derived from their languages: qehillah apulyanit (the Apulian community, using Romance) and qehillah griqa, or griga, using Salentine Greek.¹⁰⁸ The linguistic term was thus a cultural signifier for both Jews and Christians.

The ancient Greeks labeled those who did not speak their language barbaroi, “barbarians,” and this term is also used to describe the Libyan heathens in the Byzantine dedication of the rebuilt walls of Taranto [139].¹⁰⁹ Today, ppoppiti, with the same kind of staccato syllables as barbaroi, is used to describe the inhabitants of the southern Salento by those who live along and beyond its northern limit and speak an Apulian rather than a Salentine dialect.¹¹⁰ Ppoppiti has also come to connote boorish, unlettered peasants, just as speaking a non-Greek tongue once implied other kinds of cultural and behavioral barbarisms. As usual, when the term is adopted by those who have been identified pejoratively—when it becomes an emic rather than an etic label—it loses much of its negative force.¹¹¹

This chapter has demonstrated ways in which language is a linchpin of identity. Hebrew users were at least bilingual because they were always part of a larger community that did not share their language. Greek and Latin speakers, especially those who lived in a monolingual village, lacked such linguistic pressures, but their verbal interaction is apparent in their public texts nonetheless. Despite the erasure of the Jewish communities of the Salento by the sixteenth century, both the Jews and their sacred language have left traces in the local record. In addition to the toponyms that refer to Jewish streets or neighborhoods, we noted in the previous chapter the derogatory labels Sciuteì and Sçiudèu applied to the inhabitants of two southern towns. In Taranto, during the procession of the “Perdoni” characteristic of Holy Week, those seeking pardon from sin walk in pairs to venerate tombs in the city’s churches and are greeted in the street with u salamelecche, a respectful inclination that surely derives either from the Arabic salaam aleikum or the Hebrew shalom aleichem, peace be with you.¹¹² Few today understand the origin or meaning of the phrase.¹¹³ While I cannot prove that Hebrew words penetrated the local dialects, they did in Rome and may have done so here.¹¹⁴ In any event, the Salento vernaculars mix linguistic elements from Greek and Latin, and this is one of the features that conferred a unique regional identity. Even now, the inhabitants of one town in the Salento can identify those of another by their dialect. As was the case in the Middle Ages and earlier, language continues to be used as a method of inclusion and exclusion.