Epigraphy and Islam
Download as PDF (best formatting)
INTRODUCTION
Epigraphy is the essentially the study of inscriptions and graffito. In some ways, it is parallel to paleography. It involves the study of such writing and attempts to date them. Sometimes this is relatively easy – an official inscription that is dated can usually be safely attributed to a certain time. However, writings that are not official, and un-dated – such as graffito (the singular of graffiti, employed normally as an archaeological term) – pose greater challenges – not only of dating, but of authenticity. This has not stopped dawah activists from utilizing epigraphy, including graffito, as evidence for the historical authenticity of Islam.
Epigraphy as a historical discipline
Epigraphy is defined as follows (Sir John Edwin Sandys, Latin Epigraphy: An Introduction to the Study of Latin Inscriptions, Cambridge: at the University Press, 1919, p. 1 [Although the book addresses specifically Lain epigraphy, the same principles hold in general; we have omitted specific references to “Latin” in this quotation]):
The science concerned with the classification and interpretation of inscriptions is known by the name of Epigraphy, a term ultimately derived from ἐπιγραφή, the Greek word for an ‘inscription.’ …Strictly speaking, Epigraphy is a branch of Palaeography… Epigraphy may be defined as the science concerned with all the remains of the … language inscribed on durable materials, such as stone or metal, but inscriptions on coins which, under this definition, form a part of Epigraphy, are generally reserved for the domain of Numismatics… Palaeography is, in practice, confined to that which is written on less durable materials, such as papyrus, parchment or paper. Writings on tablets covered with wax may be treated as belonging to the domain of Epigraphy, but they are more closely connected with that of Palaeography. The province of Epigraphy is, in one respect, wider than that of Palaeography, for, while Palaeography confines itself to the study of the forms of writing found in ancient manuscripts. Epigraphy not only deals with the lettering, but is even apt to concern itself with the subject-matter of ancient inscriptions, thus unduly encroaching on the provinces of History, and of Public and Private Antiquities.
This is not simply true of Latin/Roman epigraphy but is valid generally. The other issue is consider methodology in epigraphy. Peter J. Brand, “The Historical Record”, (Vanessa Davies and Dimitri Laboury [Eds.], The Oxford Handbook of Egyptian Epigraphy and Paleography, New York: Oxford University Press, 2020, p. 60), points to related concerns about method – the dangers of arbitrariness and misreading:
An important caveat for using epigraphic data to support historical interpretation is the danger of misconstruing the motives or purposes of the ancient Egyptians in cases of erasures, iconoclasm, reworking, or other instances in which they altered monumental reliefs and inscriptions... In the modern history of Egyptology, scholars have often assigned visceral, personally antagonistic motives to cases such as Thutmose III’s systematic effacement of Hatshepsut’s royal monuments, which was often seen as arising from his supposed deep hatred of her. A more cautious approach to the evidence admits that our epigraphic sources are limited in what they can tell us about historical events. So, while an epigraphic approach may involve reconstructing Egyptian history as the record of “who did what to whose monuments,” it rarely can tell us unequivocally why they did it.
When used in conjunction with other types of source material — textual, iconographic, and archaeological — epigraphic data can augment the historical record. There is also considerable overlap between these methodologies. So, erased or altered hieroglyphic texts can be approached epigraphically, philologically (to reveal what the text says and what has been changed or removed), and paleographically.
In the same volume we find this by Claude Traunecker, “How to publish an Egyptian Temple?”, pp. 299-300, which attempts to suggest a methodology for correct reading, including collective reading:
Whatever the circumstances, I remain very attached to the method of copying by hand. For me, this phase of work, especially if we follow the procedure that I advocate, is crucial. It provides an invaluable base document and allows all field observations to be recorded live. Additionally, if we respect the three rules of the field epigrapher’s work, it is the product of a real group workshop.
1. All primary documents are prepared using fine lead pencil on graph paper. They must include reference to location, be signed by authors, and be dated. Do not forget indications of orientation and scale.
2. Work in pairs as much as possible, whether together or in turns. This method (“copying with four eyes”) generates a constant discussion of readings. All philological comments and observations, of structure, grammar, parts of translations, and other points, will be recorded on the copy sheet. One must not hesitate to attempt translation and philological commentary on the spot. At el-Qal’a, we always had a copy of the Wörterbuch with us to make headway in this difficult epigraphic situation. Today, information technology facilitates these consultations.
3. All copies and documents are collated systematically, if possible, by or with a colleague who did not participate in the initial copying. These collations are recorded on the copy sheet with a different colored pencil and also signed and dated... One of the methods of collation that Serge Sauneron taught me consists of systematically checking the text sign by sign, but backward, for example, from bottom to top for columns or, for lines, from end to beginning. This avoids the shortcuts and lack of attention of a reader drawn along by the content of the text and its particular orthographic conventions.
Constant conflicting views, especially in the case of difficult texts, is salutary for the work’s quality. Epigraphy should not be solitary work. It teaches epigraphic humility, even when we have a reading that seems satisfactory to us, to listen objectively to the suggestion of a colleague, whoever that may be. When one has identified a plausible solution, one does not tend to notice other possible paths.
For the process, I propose, at least in the case of epigraphy with modest means, the following three stages:
1. Set up a drawing at 1/20 scale of the wall on graph paper. Note stone joints, restored areas, and other features. Measurements are taken directly with a wooden yardstick of two meters and possibly a téléscomètre (rigid telescopic meter). This reference document, which does not replace the architectural survey possibly done by an architect or topographer, is easy to make and is immediately available. It will be very useful for establishing the overview drawings.
2. Establish a plan on graph paper of each epigraphic unit at 1/10 scale. Precisely record the dimensions of columns of text, frames, thrones, and other features. Note crowns and objects held.
3. Finally, proceed to manual copying of texts on the same lined sheet (cadrat of 1 cm). It will be used for collation and to note various remarks.
Another important example shows us the methodology of coping with broken inscriptions and common names (Christer Bruun and Jonathan Edmondson [Eds.], The Oxford Handbook of Roman Epigraphy, New York: Oxford University Press, 2015, pp. 4-5):
When Pflaum and Marec discovered the Suetonius inscription at Hippo in 1950 during excavations of an exedra in the portico on the E. side of the forum, lying face down, it was badly damaged. Of the original moulded plaque, just sixteen fragments survived. After a long and thoughtful discussion, Pflaum and Marec restored the text conservatively as follows (AE 1953, 73; Fig. 1.1):
C(aio) Suetoni[o] / [. fil(io) . . . (tribu)] Tra[nquillo] / [f]lami[ni–c. 10 letters–] / [adlecto i]nt[er selectos a di]vo Tr[a]/[iano Parthico p]ont(ifici) Volca[nali] / [–c. 16 letters– a] studiis a byblio[thecis] / [ab e]pistulis / [Imp(eratoris) Caes(aris) Trai]ani Hadr[i]an[I Aug(usti)] / [Hipponenses Re]gii [d(ecreto) d(ecurionum)] p(ecunia) p(ublica)
To C. Suetonius Tranquillus [son of ?, of the voting-tribe ?], priest of [??], chosen as a jury-panel member (?) by the Deified Trajan, pontifex of the cult of Vulcan, a studiis (in charge of literary and cultural pursuits), in charge of the libraries, in charge of correspondence of the emperor Hadrian. The inhabitants of Hippo Regius (erected this monument) with public funds by decree of the town council.
Enough survived of the text to stimulate the curiosity of the discoverers: in particular, the name in the first line. Names are always useful in inscriptions for a variety of reasons. In this case, C. SVETONI and TRA must have seemed so fascinating that Marec and Pflaum may well for a minute have neglected the important task of physically recording the stone and its full text. Instead they probably hurried off to consult standard works of reference in order to find out whether they could draw any conclusions from that name. Could it really be… the Suetonius, who is known from his own transmitted works and from Pliny to have borne the cognomen Tranquillus?
Before they could entertain the hypothesis of identifying the honorand with the famous imperial biographer, some background research on Roman naming practices needed to be carried out. In today’s North America, there are many men called William Clinton, not just the former U.S. President, and few of the Clintons one might encounter will even be related to the Bill Clinton known the world over. How could they find out about the distinctiveness of the name Suetonius in the Roman world?
The various corpora of Latin inscriptions include extensive indices of all the individuals mentioned, with separate lists of family-names (gentilicia) and surnames (cognomina). Similar indices can be found in the annual volumes of L’Annee Epigraphique (AE) and the Supplementum Epigraphicum Graecum (SEG), which register new discoveries and noteworthy discussions of previously found Latin or Greek inscriptions (Ch. 4). Today the various epigraphic databases (Ch. 5) allow for a rapid search of names, with the proviso that a name may appear in various grammatical cases and that such an automated search may not catch variant spellings.
A consultation of the indices of CIL VIII (covering North Africa) and Inscriptions latines d’Algerie I (1922) showed our scholars that the name Suetonius is indeed rather rare in the region; just three or four Suetonii are attested (ILAlg 3374–75, 3843, and possibly 3105).
This shows that great reserve must be taken with inscriptions; if we find a name like ‘Umar or Muhammad, we must not rush to judgment as to the identity of the individuals mentioned therein.
Official Inscriptions
Official inscriptions can confirm what we know of individuals from other media (Bruun & Edmondson The Oxford Handbook of Roman Epigraphy, p. 3).
On the basis of a paper they had presented on 15 January the same year at a meeting of that learned society, they were on the verge of causing quite a commotion among Roman historians and classicists. Who has not heard of Suetonius, the imperial biographer? His scholarly and somewhat sensationalist lives of the twelve Caesars from Julius Caesar to Domitian have influenced later Roman writers, the Middle Ages, and common modern perceptions of these Roman principes. Like many of the Roman authors we know so well from the literature they wrote, Suetonius used to be completely unknown outside of his own work, except for seven references to him in correspondence of the younger Pliny (Ep. 1.18; 1.24.1; 3.8.1; 5.10.3; 9.34; 10.94–95) and a few further remarks in some other later sources (cf. PIR2 S 959). Imagine the excitement, therefore, when the two French scholars in 1950 came upon a long lost inscription during excavations at Roman Hippo Regius, a coastal town in eastern Algeria (now Annaba, formerly Bone), which seems to give details of the life of the author Suetonius!
A more pertinent example is that of Pontius Pilate, mentioned in the New Testament, especially the Gospels; Philo, the Alexandrian Jewish writer; Josephus, the Palestinian Jewish historian who took Roman citizenship. Philo (c. 20 B.C.– c. 50 A.D.); Josephus, (37-100); Tacitus, (c. 56 – c. 117). The Roman pagan historian Tacitus makes some comment, but possibly more extensive comments have been lost.
Coins issued by Pilate exist, dated from the sixteenth (c. 28-29) to the eighteenth year of Tiberius (i.e. 31-32 A.D.) [Madden, Frederic W., Coins of the Jews, (London: Trübner & Company, 1881), pp. 182-183].
In 1961 the Pilate Stone was found at Caesarea Maritima, Roman administrative capital of Judaea. It is dated to 26-37 AD. It reads: To the Divine Augusti [this] Tiberieum… Pontius Pilate … prefect of Judea … has dedicated [this].
This confirms that Pilate was Prefect of Judaea during the reign of Tiberius.
3. Graffito
If we move specifically to graffito, a valuable comment is found in this statement (J.A. Baird and Claire Taylor [Eds.], Ancient Graffiti in Context, New York & London: Routledge, 2001, p. 3):
… this volume shows that many graffiti not only communicate a message to a reader, but also can be viewed as being part of a dialogue with one another. Graffiti may also mark time and space, and are one of the few forms of writing from the ancient world which preserve the material context of their production. Contextualising graffiti in this way highlights that, as a form of evidence, they have much to contribute to existing scholarly debates, for example those concerning literacy, orality, the relationship between text and image, the display of emotions, performance and the material construction of memory.
The same page makes further important observations:
The English word graffiti comes from the Italian verb to scratch, and in its simplest sense, graffiti are simply markings scratched onto a surface, whether of text or pictures. But graffiti themselves can be made by a number of means besides scratching: these include inscribing, using charcoal, ink, or paint (painted marks are usually termed dipinti and sometimes considered as a category of graffiti… Apart from shining a light on the tools used, the way in which a mark is produced — whether with chisel, charcoal, paintbrush, or other implement — is not the singular means, nor necessarily the most useful criterion, with which to define this practice.
Peter Keegan (Graffiti in Antiquity, New York & London: Routledge, 2014, p. 54) gives a Hellenistic-Egyptian example of religious graffito:
If we return briefly to Egypt, but move forward in time to the Hellenistic (or Ptolemaic) period (332–31 BCE), a variety of surfaces in and near the small rock temple of Seti I at El Kanais (east of Edfu in the Wadi Mia) displays a record of graffiti inscribed by Greek travellers, incised in relation to earlier Egyptian markings, later Roman messages, as well as post- classical and early modern writings. Like Wâdi el-Hôl, the El Kanais site was used as a way- station for military traffic and workers associated with mining operations. The facility of the site for such a purpose was the reason why Seti I (c.1294–1279 BCE) originally constructed the small temple: to mark the digging of a well to service one of the major routes between the Nile Valley and his gold- mining operations in the eastern desert. An inscription on the walls of the shrine records the sinking of the well shaft and the pharaonic foundation of the temple. Another inscription, recorded on a pillar in the temple a millennium after Seti I’s foundation, describes the place as the “Paneion”, reflecting the identification of iconography associated with the Egyptian god Amen-Re (namely, the attributes of the ithyphallic god of fertility, Min) which corresponded to the Greek god of wild places, Pan.
He goes on to give examples from the temple (pp. 54-55):
In relation to this cultural appropriation, Greek textual graffiti are directed for the most part to Pan – bearing the epithets Euodoi (“of the good road”), Euargos (“of the good hunt”), Epekoos (“who listens in prayer”), or Soter (“Saviour”) – and comprise expressions of thanksgiving for safe return from distant places, acts of adoration in relation to the dangers of desert travel, and a miscellany of personal signatures. Two of the 92 extant graffiti warrant special attention:
G3.12 “Pan of the good road, Zenodotos son of Glaukos has given you this [ornament? Altar?], having come back from the land of the Sabaeans [modern Yemen].”
G3.13 “[I dedicate] this to Pan of the good hunt who listens to prayer, who has saved me from the land of the Troglodytes [an epithet for the African inhabitants of the Red Sea], having suffered greatly in redoubled hardships, and from the sacred land which produces myrrh and from among the Koloboi. You saved us when we went off course on the Red Sea, and you sent a breeze to our ships when they were rolling on the ocean, whistling in shrill breaths in the reeds, until you yourself brought us to the port of Ptolemais, steering us with your hands, most skilful from the hunt. Now, friend, save the city which Alexander first founded in Egypt, the most famous of cities. I proclaim your power, friend Pan, having come back safely from Ptolemais [?] …”
This shows that we must not think of ancient graffiti the way we often consider modern graffiti – as defacement, street art or protest material. These examples give us indications of religious belief and practice.
Dating
Sandys, Latin Epigraphy: An Introduction to the Study of Latin Inscriptions, p. 200, presents several means to attest a date to inscriptions: “The date of an inscription may be determined (1) by its form, in respect to (a) the shapes of the letters, or (b) the spelling of the words; and (2) by its subject matter.”
Essentially this presents a paleographical approach to dating – examining the words and letters. It is worth quoting Metzger again (Bruce Metzger, Manuscripts of the Greek Bible, New York & Oxford: Oxford University Press, 1981, 1991, p. 3):
PALAEOGRAPHY… is the science that studies ancient writing, preserved on papyrus, parchment, or paper, occasionally on potsherds, wood, or waxed tablets. Epigraphy deals with ancient inscriptions on durable objects, such as stone, bone, or metal, while numismatics is confined to coins and medals. The distinctions are less superficial than it may seem, for the forms of letters were determined in part by the nature and the size of the material that received them.
Greek palaeography has three aims: first, developing the practical ability of reading and dating the manuscripts; second, tracing the history of Greek handwriting, including not only the form and style of letters, but also such matters as punctuation, abbreviations, and the like; and third, analyzing the layout of the written page and the make-up of ancient book forms (codicology).”
(p. 49, regarding mss. which are undated) Since most manuscripts, however, lack such chronological information, their approximate age must be determined on the basis of considerations of the style of the script. Now, the evolution of handwriting is a gradual process, and one form gives way to another almost imperceptibly. A considerable lapse of time is generally required to produce significant changes in the shapes of the letters and the general appearance of the script.
To return to Sandys, p. 200, he notes how the Latin alphabet developed (i.e. changed) over time:
The date of an inscription belonging to the Roman Republic may be partly determined by the character of the letters.” He gives examples on how “Under the Republic, forms of A… which are never constant, cease altogether about 184-174 B.C., and the same is true of the sibilant letter 𐌔, and of those forms of O which are open at the top or the bottom… 𐌖 is hardly ever found before 55 B.C. Z, which had appeared after 273 B.C. on an ancient coin of Cosa, was finally borrowed solely for the spelling of Greek words containing that letter, and was placed at the end of the Latin alphabet.
In terms of spelling, Sandys (p. 202) observes:
Turning to questions of spelling, we find that, in the final syllable of the inflexions of nouns and verbs such as tribunŏs, pocolom, donom, sacrom, dederont, coiraveront, o is superseded by u about 234 B.C. ... Similarly, in the final syllable of words like fruge, cepet, curavet, dedet, e is superseded by i. The final d of the ablative falls out of general use after 200 B.C., though it is retained in 186, in the Senatus Consultum de Bacchanalibus.
On p. 203, Sandys addresses ‘subject matter’:
The date may also be determined by the subject-matter, by the mention of consuls or other officials, whose date is exactly or approximately known. In imperial inscriptions the details of the titles borne by the emperor are generally conclusive. References to consulships, and to imperial titles, are often found in honorary inscriptions. Epitaphs are rarely dated by consulships, and it is very exceptional to find an epitaph in which the successive dates in a centurion’s career are recorded in terms of imperial titles. The lex parieti faciendo of Puteoli is dated ab colonia deducta anno xc, the year of the foundation of the colony being 194 B.C. An Umbrian monument of 32 A.D. in honour of Tiberius is dated 704 years from the foundation of Interamna; and the ‘restoration of liberty’ by Nerva on the death of Domitian in 96 A.D. was commemorated on the Capitol in an inscription dated in the year of Rome, 848. Such dates are generally confined to tituli sacri, and are very rare in Italy. But they are common in Asia, and are also found in Mauretania, where some public baths were dedicated in ‘the year of the province 157’, i.e. 196 A.D., and where a priest is described as having died at the age of 105 (‘more or less’) ‘in the year of the province 363’, or 402 A.D.
From this, we can see the importance of official inscriptions, which - generally – can be trusted as to date.
However, we must be aware of the possibility of forgery and propaganda (p. 204):
Modern criticism of Latin inscriptions began when the forgeries, of Pirro Ligorio (who died about 1586) were detected by Maffei and Olivieri. Inscriptions produced solely to glorify a particular family, or to support a particular opinion, are always liable to suspicion'. An inscription supporting the view that Basilice is on the site of the Samnian town Murgantia is discredited by the illegitimate formation of the adjective populus Murgantius, instead of Murgantinus or Murgantiensis, and (less strongly) by the unidiomatic use of the demonstrative in basilicam hanc.
A recent example comes from the Crimea (Chersonesus Taurica), which was under Roman rule from the first century B.C to the latter fourth century A.D. (Georgy Kantor, “Local Courts of Chersonesus Taurica in the Roman Age”, Paraskevi Martzavou & Nikolaos Papazarkadas [Eds.], Epigraphical Approaches to the Post-classical Polis: Fourth Century BC to Second Century AD, Oxford: Oxford University Press 2013, p. 69):
In 2005 Igor Makarov published a fragmentary inscription from the Chersonesus Museum collection containing provisions for the reform of local court system in Chersonesus Taurica in the early Imperial period. The decree deals with the reform of the local juries for private law disputes (whose survival into the Roman period has not been previously attested), necessitated by the lack of eligible jurors. Highly importantly for legal history, the provision for reiectio iudicum appears to have been taken over from Roman practice into an otherwise Greek model on the initiative of the Chersonesites themselves, without any suggestion of an intervention from the Roman authorities.
On its dating, Kantor comments (p. 74):
The appearance of the office of archons in the text of the decree suggests placing it in the period after the constitutional reforms in Chersonesus in the Augustan period, in line with all the other mentions of that office in Chersonesitan epigraphy.
This dating is supported by other indications. Makarov has convincingly demonstrated through parallels among Chersonesitan and Bosporan inscriptions that on palaeographic grounds the decree is most likely to belong to the second half of the first century AD. Consideration of the language of the inscription also points towards the early Imperial period… and the presence of the Doric forms precludes a date in the third century AD. The use of a triangular pediment to top a stele with the decree text is also typical of the northern Black Sea epigraphy from the second century BC to the early second century AD, though the absence of any surviving decoration elements makes more precise dating on artistic grounds impossible.
Bruun and Edmondson, The Oxford Handbook of Roman Epigraphy, pp. 14-15 addresses dating:
Epigraphic patterns and practices changed over time and it is important to establish the date of an inscription for it to be as useful as possible for enhancing our understanding of classical antiquity. Correspondingly, for the restoration of a damaged text and for its interpretation, it is helpful to know to which period it belongs. Hence, every textual edition should be accompanied by at least a tentative attempt at dating, even if no precise chronological indicators can be found.
In a few fortunate cases dating presents little problem, namely when a consular date is given or a reference to a known local era appears... Sometimes the mention of officials or magistrates either of the Roman state or of local municipalities, for whom the date of their holding of office is known, help date an inscription accurately. The occurrence of an emperor’s name and titulature is always useful, as the tribunician potestas and the imperial acclamations may allow us to date the text to a precise year, and at the very least the text’s chronology may be narrowed down to the reign of the emperor mentioned…
However, this is not always possible: “In the vast majority of inscriptions such helpful elements are unfortunately lacking. Nevertheless, after much scholarly discussion, which is still ongoing, some generally acknowledged dating principles have been established. As a result, editors often have to be satisfied with very approximate suggestions for a text’s date, such as “second/third century CE” or a terminus post quem, indicating that it belongs to the period after a certain event or emperor’s reign.”
They then suggest “Some of the most useful criteria are” (pp. 15-16):
• the formula D. M. or D. M. s., which is very common in funerary inscriptions, does not (with exceedingly rare exceptions) appear in Italy before the mid-first century CE and in the western provinces before late in that century.
• the appearance of known historical figures or events help to provide chronological orientation, as do the titles of Roman military units, which evolved over time and the history of which has been reconstructed from other sources.
• the appearance of imperial freedmen is helpful, as the beginning of the reign of the emperor who manumitted them is an obvious terminus post quem. However, it needs to be remembered that an Aug(usti) lib(ertus) may have lived on for up to fifty years after the death of the emperor in question.
• personal names can provide useful chronological hints... If a common Roman bears no cognomen, the text dates to before c. 50 CE, likely to the Republican or perhaps the Augustan period. Filiation started to be omitted with greater frequency as the Principate progressed, while in the Republic it was more common... The use of supernomina or signa (marked by the connectives qui et or sive) is a sign of a late date: second or, more likely, third/fourth century...
• the massive appearance of individuals bearing an imperial gentilicium such as Flavius, Ulpius, or Aelius is probably an indication that the text dates to a period after the reign of the emperor(s) in question. These individuals are likely to be descendants of manumitted imperial freedmen or newly enfranchised citizens who took the gentilicium of the reigning emperor or their descendants. In many parts of the Empire, the name Aurelius became particularly common after Caracalla’s grant of citizenship to all freeborn inhabitants of the Empire in 212 CE.
• in Rome, Italy, and the Hispanic provinces, the practice of using marble for inscribing a text is Augustan or later. In other regions the use of certain materials may also be a chronological indicator.
• the decorative elements of the monument on which the inscription was carved may help to date the text on archaeological or stylistic grounds: for instance, in the case of funerary monuments with portrait-busts, the hairstyles of those depicted can provide some chronological orientation.
• the circumstances of an inscription’s discovery may assist with its dating. The archaeological layer in which it was found or the construction to which it belonged may have been dated by the excavators. It is important to be aware of the danger of a vicious circle here. Archaeologists are sometimes keen on using epigraphic evidence for dating sites and archaeological strata, even just in a preliminary, tentative, and hypothetical way. When epigraphers subsequently base their dating on this foundation, little has in reality been achieved.
Lastly, letter-forms… are often used as a dating criterion. For identifying Republican inscriptions, the older forms of several letters are useful…
This becomes important, since the Roman Empire – especially in terms of what was later called the “Byzantine” Empire – immediately preceded the rise of the Arab Empire, and many of the territories constituting the latter Empire previously belonged to the former (e.g. northern Hijaz, Palestine, Syria, Mesopotamia, Egypt, etc.).
At any rate, the general principles hold across cultures and languages.
For example, as we have noted elsewhere, the absence of epigraphic evidence in Mecca, Medina and Khaybar – supposedly the sites of substantial Jewish presence according to the Hadith and Sira – contrasts with the clear evidence of a Jewish presence in places in the northern Hijaz (which were under successively Nabatæan and later Roman/Byzantine jurisdiction:
One tomb inscription at Hegra from 42/43 AD, in Nabatæan Aramaic reads: ‘This is the tomb which Shubaytu son of ‘Ali’u the Jew made...’ In al-’Ulā, we also find such Jewish tomb inscriptions in the same language, as with one from 307 AD. In nearby Tayma, there is a similar example from 203 AD which is important because it appears that it refers to a local headman, either of his ethnic group or of the town itself, and another example from Hegra in 356/7 also refers to someone who held an analogous position there. The significance of this is that these ‘are important texts for north Arabian Jewry, for they imply that some of them at least were members of the elite of the society.’ However, it should be noted that whilst the inscriptions cover ‘a large period of time, at the very least the first century BCE to the fourth century CE’, they are ‘relatively few in number’, and ‘not geographically very widespread, principally hailing only from al-Ula and Mada’in Salih.’ Hoyland comments that ‘the limited nature of this epigraphic crop’ is surprising, particularly given ‘the very frequent reference to Jews in the Qur’an.’ He then observes that there is ‘not a single clearly Jewish inscription’ that has yet been found ‘at Mecca, Yathrib or Khaybar despite quite a number of epigraphic surveys conducted at all three sites.’
In terms of dating graffito, the same basic principles of epigraphy/paleography remain valid (Karen B. Stern, Writing on the Wall: Graffiti and the Forgotten Jews of Antiquity, Princeton and Oxford: Princeton University Press, 2018, p. 27)
Methods of determining the antiquity of individual graffiti, in comparable ways, vary according to their precise archaeological contexts and contents. Few examples of graffiti associated with Jews include exact dates or regnal years, which might better assist their accurate dating. In certain cases, however, the stratigraphic record offers important clues for establishing chronologies of graffiti written by or around Jews. Graffiti from the Dura‑Europos synagogue, for example, were discovered in a sealed archaeological context that predated the destruction of the building sometime between 255 and 257 CE. In this case, we have a useful terminus ante quem for their ancient application.
Other graffiti deposits pose greater challenges for establishing patterns of relative dating. Graffiti in Beit Shearim, Palmyra, or along cliffs in deserts of Egypt, Sinai, and Arabia are located inside spaces that sustained ongoing reuse; this might prompt questions about their antiquity. While it remains impossible to establish secure dates for every graffito and dipinto discovered throughout these sites, considerations of regional customs, language, and iconographic typology help to better situate their chronology. Although methods of dating textual graffiti through paleography remain as tenuous as is dating pictorial art through patterns in iconography, deployments of Greek and Latin (and in some cases, Aramaic, Hebrew, and Nabataean) in graffiti may assist their dating to similar periods of Hellenistic, Roman, and Byzantine hegemony, which predate the early medieval period.
J.A. Baird, “The Graffiti of Dura-Europos: A Contextual Approach”, Baird and Taylor, Ancient Graffiti in Context, p. 61, also addresses graffito dating:
One house, near the centre of the town, had so many graffiti scratched into its walls it was named ‘The House of the Archives’ by the excavators… Many of the graffiti record transactions and commodities, accounts and tabulations. Some also contain dates, the earliest of which is 218 CE; the others, including horoscopes, cluster around 235 to 240 CE. One graffito from this house has played a part in the reconstruction of the historical sequence at Dura, as it refers to the Persians descending on the site, in 239 CE, some time before the demise of the town at the hands of the Sasanians, probably in 256…
Kristina Milnor (Graffiti and the Literary Landscape in Roman Pompeii, Oxford: Oxford University Press, 2014, pp. 15, 17) presents some of the issues in dating graffiti in ancient Pompeii:
It is challenging to date Pompeian wall writings, as most do not offer chronological information, and those which do overwhelmingly provide a day but not a year. Those which do record the Roman consuls — our best means for establishing a firm date — range remarkably widely, from the earliest of 78 BCE (CIL 4. 1842, from the basilica) to the latest of 60 CE (CIL 4. 4182, in the House of the Silver Wedding).
Islamic/Arabic graffiti faces the same problems in dating (Jouni Harjumäki & Ilkka Lindstedt, “The Ancient North Arabian and Early Islamic Arabic Graffiti: A Comparison of Formal and Thematic Features”, Saana Svärd and Robert Rollinger [Eds.], Cross-Cultural Studies in Near Eastern History and Literature, Münster: Ugarit-Verlag, 2016, p. 62): “The date of many Arabic graffiti is problematic. Only a few of them provide exact dates, so their dating must, to some extent, depend on paleography. However, there is no clear-cut or error-free way to do this, and some of the texts discussed below could be somewhat later than from the first two centuries of Islam.”
Ilkka Lindstedt, Fellow, The Helsinki Collegium for Advanced Studies, “Writing, Reading, and Hearing in Early Muslim-era Arabic Graffiti”, https://iqsaweb.wordpress.com/2017/01/02/writing-reading-and-hearing-in-early-muslim-era-arabic-graffiti/, January 2, 2017 writes:
Arabic inscriptions yield dated texts that are important, among other things, for the study of the developing Islamic identity. The majority of the surviving inscriptions are undated, but just to give one figure, there are according to my calculations almost 100 dated published inscriptions between the earliest Islamic-era one (23 AH) and the end of the Umayyad dynasty (132 AH). This is not a meager amount for a period for which otherwise dated (or datable) evidence is slight… Most of the extant Arabic graffiti are lapidary and engraved, although we have some painted graffiti as well.
Some of the methodology of addressing Arabic graffiti can be found in Younis al-Shdaifat; Ahmad Al-Jallad; Zeyad al-Salameen; Rafe Harahsheh, “An early Christian Arabic graffito mentioning ‘Yazid the king’”, Arabian Archaeology and Epigraphy, Volume 28, Issue 2, 24 November 2017, p. 315:
The rock inscription discussed here was discovered during the first season of the El‐Khḍerī archaeological and epigraphic survey project in north‐eastern Jordan. The inscription comes from as‐Samrūnīyyāt (site number S5), located 12 km south‐west of Qaṣr Burquʿ, and was the only text found at the site. In terms of archaeological remains, the site contains a tailed tower tomb on its southern slope and a number of stone‐circle structures scattered on its northern side. These ancient structures have been reused in recent times as seasonal camps by nomadic pastoralists. The inscription consists of a four‐word early Arabic graffito, perhaps from the sixth or seventh century, accompanied by a cross. The text contains several unique palaeographic features and a reference to a certain yzydw ʾl‐mlk or ‘Yazīd the king’.
(Next paragraph from p. 316):
2.1 Letter shapes
The letter forms are completely Arabic as opposed to the Nabataeo‐Arabic script, suggesting a sixth‐century date at the earliest. The dāls, however, have a dot, a survival of an Aramaic practice of distinguishing the two phonemes but a completely unnecessary relic given that the r at this time has a distinct shape. This practice is still found in the Nabataeo‐Arabic script but has not yet been attested in the Arabic script proper until now. The dot is not used to distinguish ḏāl from dāl, as in the later Arabic script…
(p. 318):
2.4 Formularies and orthography
The formula dkr ʾl‐ʾlh is unique to sixth‐century Christian Arabic inscriptions, probably attested in the Zebed inscription and in a new mid‐sixth‐century Arabic inscription from Dūmah. The verb dkr would appear to be an Arabic calque of the Nabataean dkyr, common in Nabataean and Nabataeo‐Arabic graffiti. The divine name ʾl‐ʾlh is only attested in sixth‐ or perhaps late fifth‐century Christian Arabic inscriptions, and is therefore probably a calque of the Greek ho theós. We can compare the cross accompanying the benediction to the Dūmah pre‐Islamic Arabic inscription (Nehmé, forthcoming, b) and to the crosses associated with the Ḥimā Arabic and Nabataeo‐Arabic inscriptions (Robin, al‐Ghabbān, & al‐Saʿīd, 2014).
It can be seen that the authors date the inscription on paleographical grounds – date of words, usage, itself.
Ian Morris (http://www.iandavidmorris.com/not-set-in-stone/), “Not set in stone”, March 19, 2014, observes the treatment of “a cluster of graffiti from Mount Sal‘ in north-western Arabia. A bunch of passers-by apparently signed their names in pre-classical script; one wrote the sort of supplicatory message we often see in old Arabian graffiti.
These would be no more than late-antique curiosities, were it not for four names that draw our attention: Abū Bakr, ‘Umar, ‘Alī bin Abū [sic] T[ā]lib, and Muhammad bin ‘Abd Allāh.”
He then observes:
For one thing, the Muhammad bin is on the line above ‘Abd Allāh, indicating that erosion has damaged the autographs of two distinct people, just as it’s damaged several other lines. But there are two more fundamental reasons to be sceptical.
1) These are very common names, and anyone who’s studied the history of the Arabs knows that Arab names are often duplicated.
2) Moreover, these four ‘recognised’ names appear in no discernable order alongside the names of people who are completely unknown. Nothing marks them out as special; as the heads of a revolutionary or reformist movement.
They could be anybody of any tribe: pastoralists, traders, soldiers on campaign or a bunch of teenagers on an adventure. The evidence simply isn’t strong enough to sustain the ideal identification.
We can see that dating inscriptions, especially graffito, can be difficult. The method is largely paleographic. The possibilities of misreading and even forgery needs to be noted. It follows that we cannot point to rock graffito as clear evidence for the historical authenticity of Islam.