A Late Antique Schooltablet at Duke University

by Peter van Minnen


NB: To use the Greek filter of the Perseus project at Tufts University and display the Greek contained in this article the browser's fixed font should be a Greek font and its encoding should be user-defined.
The wooden tablet published in this article is kept in the Rare Book Room of Perkins Library at Duke University. It was donated to the library in December 1980 by Professor W.H. Willis, who had inherited the tablet in 1958 from the late Professor D.M. Robinson along with a collection of papyri.[1] Robinson had acquired the piece in 1910 on one of his travels to Egypt. No further information about its provenance exists.

Both sides of the tablet are covered with a white slip and inscribed in black ink with simple writing exercises. There are many parallels for such exercises in other schooltexts from Egypt, on tablets, ostraca, and papyri.[2] The particularly numerous examples from late antiquity graphically illustrate the process by which a relatively high level of literacy was attained in Egypt in that period.[3]

The Duke tablet is interesting for a number of reasons. It combines several forms of simple writing exercises on one tablet and it apparently also combines Greek with Coptic. The Greek subscription on the outside (line 17) suggests that the context is Greek, but in addition to the twenty-four letters of the Greek alphabet (written on each side in different order; lines 1-8 and 16) the tablet also contains the six additional letters used for Sahidic Coptic texts (written on each side in the same order, which differs from the expected order only in the interchange of janja and chima;[4] lines 1-6 and 9-14). Both from the subscription on the outside (line 17), which gives the date and the total number of syllables on the tablet (eighty-four), and from the use of a so-called chrismon on both sides marking the beginning (line 1) and the end (line 15) of the text one might infer that the text on the tablet was what the teacher gave the students to copy on a single day.

The scribe (or scribes) produced twelve columns[5] of seven mock syllables each by inserting the seven vowels of the Greek alphabet (a, e, h, i, o, u, w) in a combination of two consonants.[6] The first consonant is b eight out of twelve times and g in the remaining four columns. The second consonant varies with each column and follows the order of the alphabet.[7] The exercise starts with c, presumably because the previous consonants had already been covered on another tablet,[8] then skips o, because it is a vowel,[9] and then proceeds from r through x including u, although that produces awkward syllables. Maybe u was treated as a consonant (Egyptian w).[10] Of the last four columns of syllables, where the first letter is g, the first is a column without a second letter. The exercise then skips a, because it is a vowel,[11] and three columns with a second consonant from b through d follow. There is only one slip of the pen (line 7).

After finishing the first three columns on the inside the scribe drew a vertical line and penned down the twenty-four letters of the Greek alphabet in three columns and in random order.[12] This was considered good educational practice by Quintilian:[13] quae causa est praecipientibus ut, etiam cum satis adfixisse eas (litteras) pueris recto illo quo primum scribi solent contextu videntur, retro agant rursus et varia permutatione turbent, donec litteras qui instituuntur facie norint, non ordine. This particular permutatio was brought about as follows. Starting with the last letter of the alphabet, the first column proceeds by going back four letters of the alphabet spelled backwards.[14] After reaching d the series starts all over again with the last letter but one of the alphabet and repeats the procedure of going back four letters at a time until it reaches g. After this the series picks up with the last letter but two of the alphabet and again goes back four letters at a time until it reaches b, after which the series proceeds by going back four letters from the last letter but three of the alphabet until it reaches the first letter of the alphabet. The series should have looked like wupmqd, ytolhg, xsckzb, frniea, but on the Duke tablet the letter p was skipped in the first set and this apparently caused confusion when the scribe found out that the "alphabet" thus produced contained only twenty-three letters. The stroke below the final a is not a p, but may have been written to fill the "slot" of the missing letter.

After another vertical line, which is broken this time, a column with the six additional letters used for Sahidic Coptic texts follows. On the vertical lines were drawn between each column and a horizontal line was added to create a bottom register for another set of the 24 letters of the Greek alphabet (line 16), this time in the expected order. Next follows a subscription giving the date (March 16 of an unknown year)[15] and the total number of syllables on the tablet (twelve columns of seven syllables each make eighty-four syllables; see the note on line 17). In the process most of the lower ends of the vertical lines had to be redrawn.

The text was apparently written by two different scribes: the writing on the inside of the tablet is slanted to the right and more careful than the writing on the outside, which is slanted to the left. The difference may have been caused by the different texture of the surface of the tablet or by the fact that the scribe had to cram in three more columns on the outside of the tablet than on the inside, but it seems more likely that the inside was written by the more experienced teacher and the inside by a less experienced assistant. A likely palaeographical date for the text would be the sixth or seventh century, the heyday of the symbiosis of Greek and Coptic. The only diacritic used on the tablet is the diaeresis over the iota, sometimes in the form of three (twice in line 4, once in line 5) rather than two dots. Abbreviations occur in the date only. The second chrismon apparently also stands for the word stauro/s as it is followed by -os.[16]

At the top of the tablet in the middle two holes were drilled-visible on the outside only, because they curve upwards-to allow the tablet to be hung up on a string or to be bound up with another tablet (so as to create a diptych).[17] No trace of a string remains. The tablet is 1.2 cm thick. Its dimensions are 11.2 x 26.6 cm.

VII/VIII A.D. provenance unknown

Inside

+ bac bap bar w l b shai

bec bep ber u h f fai

bhc bhp bhr m g r hori

4 bic bip bir q x n chima

boc bop bor d s i janja

buc bup bur y c e ti

bwc bwp rwr t k a

8 o z -

4 bi+c, bi+p, bi+r 5 bor: o corrected; i+ 7 read bwr

Outside

bas bat bau baf bax ga gab gag gad shai

bes bet beu bef bex ge geb geg ged fai

bhs bht bhu bhf bhx gh ghb ghg ghd hori

12 bis bit biu bif bix gi gib gig gid chima

bos bot bou bof box go gob gog god janja

bus but buu buf bux gu gub gug gud ti

bws bwt bwu bwf bwx gw gwb gwg gwd +os

______________________________________________

16 a b g d e z h q i k l m n c o p r s t u f x y w

q
fam k e ind
/ pd onomata

12 bi+w, bi+t, bi+u, bi+f, bi+x, gi+, gi+b, gi+g, gi+d 15 read (staur)o/s 16 i+ 17 read *fam(enw\)q k, e i)nd(ikti/onos), pd o)no/mata

17 o)no/mata: the intended meaning of the word o)/noma here is "syllable" rather than the expected "word." As far as I know no parallel for this usage exists. Maybe the student assumed the syllables were "words," the meaning of which the teacher would reveal later on.

[1] W.H. Willis reported on this collection in "The New Collections of Papyri at the University of Mississippi," Proceedings of the IX International Congress of Papyrology (Oslo 1961) 381-392.

[2] For parallels see MPER XV = H. Harrauer and P.J. Sijpesteijn, Neue Texte aus dem antiken Unterricht (Wien 1985), T.Varie = R. Pintaudi and P.J. Sijpesteijn, Tavolette lignee e cerate da varie collezioni (Firenze 1989) and MPER XVIII = M.R.M. Hasitzka, Neue Texte und Dokumentation zum Koptisch-Unterricht (Wien 1992). Several lists of schooltexts from Graeco-Roman Egypt exist. For the latest see R. Cribiore, Writing, Teachers and Students in Graeco-Roman Egypt (diss. Columbia University 1993), which I have not seen. There are also two lists of wooden tablets from Graeco-Roman Egypt: W.M. Brashear & F.A.J. Hoogendijk, "Corpus tabularum lignearum ceratarumque Aegyptiarum [in German]," Enchoria 17 (1990) 21-54 and P. Cauderlier, "Les tablettes grecques d'Égypte: inventaire," Les tablettes à écrire de l'antiquité à l'époque moderne (Turnhout 1992) 63-94.

[3] See on this E. Wipszycka, "Le degré d'alphabétisation en Égypte byzantine," Revue des Études Augustiniennes 30 (1984) 279-296.

[4] For parallels see MPER XVIII 53, 4-5; 55, 2; 64, 5; 65, 3 and 69 verso.

[5] Columns seem the most natural way to present repetitive information of this sort, but occasionally one finds the information presented in horizontal lines, such as in MPER XVIII 82.

[6] There are numerous parallels for this. See especially MPER XVIII 207 for an exceptionally full example.

[7] In some school exercises the second letter is kept constant while the first letter varies from column to column. See e.g. MPER XV 8 and 10 and the text mentioned in the previous note.

[8] It is worth noting that the number of syllables for the series from ba etc. through ban etc. would be close to the number we find on our tablet: eighty-two if all vowels were skipped, eighty-four if, e.g., i were included. The preceding tablet could well have represented a day's work as well. Our tablet would be the second in a set.

[9] o is also skipped in MPER XVIII 74, 27-33 and 78, 1-7.

[10] Cf. A.E.R. Boak, "Greek and Coptic school tablets at the University of Michigan," Classical Philology 16 (1921) 189-194, no. 1 recto, which skips a, e and h, but not i. Maybe i was treated as a consonant too. Likewise MPER XVIII 78, which skips all vowels except i. I owe this observation to R. Cribiore.

[11] a is also skipped in MPER XVIII 74, 20-26 and 78, 1-7.

[12] In other writing exercises one occasionally finds the alphabet in reverse order. See MPER XV 5, 4-8 and 7, 2 and MPER XVIII 68, 2 and 74, 1-4.

[13] Inst. I 1, 25.

[14] In P. Sanz, Literarische Papyri christlichen Inhalts (Wien 1946) = MPER IV 24, Seite 14, 1-3 the order is achieved by starting with the first letter of the alphabet and then skipping three letters at a time. The second set starts with the second letter of the alphabet, the third and the fourth sets with the third and the fourth letters of the alphabet respectively. This produces four sets that mirror the ones on the Duke tablet: aeinrf, bzkjsx, ghlotc, dympuv. For this procedure see MPER IV, p. 46 and W. Brashear, "Lesefrüchte," Zeitschrift für Papyrologie und Epigraphik 50 (1983) 97-107 on pp. 98-99. The permutatio was originally brought about by setting out the alphabet in four or six columns as follows:

a b g d
e z h q
i k l m
n c o p
r s t u
f x y w

or

a e i n r f
b z k c s x
g h l o t y
d q m p u w

Instead of reading the horizontal lines or the vertical columns the teacher could have dictated the vertical columns or the horizontal lines. Other patterns are detectable in MPER XV 7, 5 and MPER XVIII 69 verso and 74, 1-4.

[15] For other dated schooltexts see P.Leid.Inst., p. 89 with footnote 9.

[16] There are many parallels in literary texts. See e.g. W.E. Crum & H.G.E. White, The Monastery of Epiphanius at Thebes II (New York 1926) no. 49, 4.

[17] These curved holes have so far not been satisfactorily explained. Most editors assume that all tablets with holes were bound together like wax tablets. In wax tablets, however, the holes go straight through the wood to allow multiple tablets to be bound together. (This also applies to the new tablets found at Kellis.) The curved holes would allow at most two tablets to be bound together with a string. It seems more natural to assume the string allowed the tablet to be hung up on the wall. This is the explanation given in the technical description in T.Varie, p. 208 and is implied in the description of the Fordham tablet, T.Varie, p. 169. Some tablets have no holes for a string at all, but one or more metal "handles." This suggests the tablets were either hung up or carried around by these "handles." See R. Cribiore, "A Homeric exercise from the Byzantine schoolroom," Chronique d'Égypte 68 (1993) 145-154 on p. 145 and again the description of the Fordham tablet in T.Varie, p. 169. (I have not seen P. Cauderlier, "Deux tablettes parisiennes en provenance d'Égypte, pour illustrer l'apprentissage des lettres grecques," Mélanges Étienne Bernand [Paris 1991] 141-153.) If the tablets were hung up on the wall, this does not necessarily imply they were all written by teachers to display in class. In good classical tradition students were supposed to display their work as well. Some tablets were, however, undoubtedly written by teachers. See e.g. F. Maltomini & C. Römer, "Noch einmal 'Ad Demonicum' auf einer Schultafel," Zeitschrift für Papyrologie und Epigraphik 75 (1988) 297-300.

(from Zeitschrift für Papyrologie und Epigraphik 106 (1995): 175-178; (c) Dr. Rudolf Habelt GmbH 1995, reprinted with permission from the editors)


Return to the papyrus home page

Return to Duke University Special Collections Library home page

Please see our page with contact information if you have any comments or questions about the Duke Papyrus Archive.

Last updated by Peter van Minnen on 3/25/96