Vindolanda Tablets Online Tablets Exhibition Reference Help

Volume One (TVI) material for tablet 346 (TVI tablet number 38)


Publication: Bowman-Thomas-Wright, Britannia v (1974), pp.474-7, pl. XLIV.

References: Birley, Vindolanda, pp.132, 153, pls. 76-7; Bowman-Thomas, VWT, p.24, pls. I, XI; Bowman-Thomas, Historia 24 (1975), pp.474, 477; Wright, Proc.XIVth Int. Congress of Papyrologists, p.355; J.N. Adams, BICS 22 (1975), pp.20-1; Birley, PRB, p.93

This tablet was the first to be recognised as such in the excavation of 1973. It was unavoidably damaged in an early attempt at conservation and we are therefore completely dependent for our readings on the photograph. The two fragments were discovered adhering face to face with the writing on the inside. The letter must have been written on a diptych in the usual two-column format. We stated in the editio princeps that each fragment is complete at the foot and incomplete at the top and right. We now think that this may be incorrect. Something may be lost at the foot, since it is possible that the writer may have added a phrase such as uale frater (cf. No. 22.17 (250)); it is clear that in line 11 we have the closing greeting of the main body of the letter. It seems probable that column ii is complete at the right since little or nothing seems to be lost at the ends of lines 10 and 11; similarly, column i is probably complete at the left. The loss, which we cannot really estimate (though comparison with the breadth of No. 25 (247) suggests that it cannot be more than about 4 cm) is therefore in the middle of the diptych; and we suppose that the left-hand column was probably broader than the right (cf. Nos. 21 (248), 22 (250)). The loss at the top is perhaps about 5 cm. There are faint traces of about six letters in address script on the back of the right-hand section.

The text is written in a competent hand with a pronounced slope to the right. The letters are made with rather thick pen-strokes, but are not without some style. Ligature is regularly found after e and occasionally elsewhere, e.g. with ar (paria, line 2, solearum, line 3), or (line 4), us (line 10) and ua (line 11). Note particularly the form of b in line 10 (twice), cf. Vol. 1., Ch. 4. There may be examples of medial point in lines 4 and 5 (see notes ad locc.).

The letter was evidently sent to a soldier serving at Vindolanda. In column i the writer mentions the despatch (or bringing) of various items of clothing. The content is not at all dissimilar to that of some letters on papyrus from Egypt (note particularly CPL 251); from Vindolanda, No. 44 (521)might also be comparable.



1. Only one long descender (r or a) and the feet of two or three letters survive from this line.

2. ram: we are now much more confident of this reading than in the editio princeps. We regard r as certain and a as very probable (both strokes of the letter are in fact visible and the apparent horizontal stroke joining them to m is presumably not ink). This may well be a pluperfect with the force of a perfect, a use which is paralleled in P.Oxy.32.4-6, iam tibi et pristine commendaueram Theonem amicum meum, and the obvious restoration would be miseram. Another possibility, which would refer to the intention of the writer, is afferam, cf. CPL 305.13 ff., dic Serapiadi si u<u>lt (denarios) xv accipere afferam illei (1. illi).

paria udon[um: cf. CPL 251.23 ff., rogo . . . ut mittas mihi inde caligas cori subtalare<s> ed udones par (the writer of this letter was serving in the fleet at Alexandria). Udones, of which the etymology is obscure, seems to mean ‘felt socks’ or ‘woollen socks’; certainly it relates to a garment which came over the feet. Digest says of them, usum calciamentorum praestant; Martial 14.140 says of udones Cilicii, non hos lana dedit, sed olentis barba mariti (i.e. the he-goat). See also Lauffer DP, ad 7.47. The end of this line will have contained a number word. In the editio princeps we tentatively suggested that line 3 might begin with et, but we no longer consider this possible. We are fairly confident that the first letter of line 3 is t (it does not slant sufficiently for e in this hand, and we are now sure that what we previously took to be the top of e is just part of the downstroke of a in the line above); the next letter might just be e, but we think it just as likely to be i (which usually has a pronounced serif in this hand). If line 3 does begin ti, then the restoration which naturally comes to mind is uigin]/ti, but this seems to be a very large number of pairs of socks for someone to be sending to a soldier.

3. t.: see previous note.

ab Sattua: in the editio princeps we suggested that Sattua might be an alternative spelling for Sattia, remarking that A. Holder, AS, quotes numerous examples of the names Sattius-Sattua. We recognise, however, that the well-known interchange between i and u attested, for example, in such words as contibernales (line 10) and optimus/optumus (cf. Väänänen, LVIP,3 pp.25-6) is not a linguistic parallel for the change here suggested. Nevertheless, the reading is certain; so we must therefore regard this either as an alternative form of the name Sattia or as a hitherto unattested proper name.

solearum [: paria must be supplied after the break (or perhaps solearum[que paria). For soles see RE 2.IA, 2257-61 and 2.IIA, 754-5. Well-preserved sandals have been found at Vindolanda (see Britannia iv (1973), pl. XLIB, Birley, Vindolanda, pp.125 f.). Dr. R. W. Davies reminded us that a relief of a legionary soldier from Britain wearing sandals is to be found in RIB 292 (Wroxeter). We can also cite a tablet from Vindonissa which has the instruction, soleas clauatas fac mittas nobis ut abeamus (O. Bohn, Anzeiger für Schweizerische Altertumskunde n.s. 27 (1925), p.14).

4. duo: there is a medial point after this word like that used elsewhere for interpunct (cf. Nos. 31 (297), 41 (323)). If it is intended as a word-divider it is very strange that it does not occur elsewhere in this letter (but see note to line 5, below). It would be equally strange if it were intended as a punctuation mark, and perhaps (despite appearances) it is no more than an unintentional mark.

subligariorum [: presumably paria is to be supplied after the break. The usual nominative form is subligar/subligaria; on this form of the genitive see J.N. Adams, BICS 22 (1975), pp.20-1. This is an undergarment worn on the lower half of the body, see RE 2. IVA, 481-2 and cf. Pliny, NH 12.59 (subligaria worn by workmen in the frankincense factories in Alexandria).

5. It is possible that there is a medial point after duo (see note to line 4) but it is less distinct than that in line 4 and probably ought not to be taken seriously.

6. The possible traces of ink at the bottom of the fragment were not noted in the editio princeps. It is difficult to be quite certain of them since the photograph suggests that there is still some dirt adhering in this area. The clearest traces seem to be at the left-hand side. It should be noted that, as a consequence of this addition, the line-numbering of the text from this point onwards differs from that in the editio princeps.

7. salutare: we are now more confident of the reading of the first six letters than we were in the editio princeps; alu must be reckoned as virtually certain, as must the second a; although the tops of both s and t are missing, both readings are possible. The letter following saluta could equally well be a or r, and we incline to prefer e for the letter following that (unless the apparent cross-stroke is not ink). We therefore no longer favour the restoration saluta amicos meos, suggested in the editio princeps. For the preceding word we cited CPL 255.23, c[r]ebrum salutat te Claud[i]us, noting that the first word is doubtful. What follows will then be the names of individuals to whom greetings are being sent. It would seem more natural for the writer to have separated each name from the one following by using et (for a list of names in Greek without consistent use of the copula see P.Oxy. 2435 verso. 35 ff., perhaps based on a Latin original).

8. ]ndem: presumably another name. Neither eu]ndem nor ta]ndem is attractive.

Elpidem: the reading is secure and can hardly be other than a proper name. Elpides is very poorly attested (P.Flor. 64.5 in BL II). The Romans regularly use the name Elpis or Helpis (sometimes Latinized as Elpidius or Helpidius), see H. Solin, Beiträge z. Kenntnis griechischer Personennamen (Soc. Scientiarum Fennica, Commentationes Humanarum Litterarum 48, 1971) i, pp.111, 115-6, who cites Helpis as the normal form in the imperial period. The regular genitive form is Elpidis so Elpidem would be quite acceptable as an accusative. In Greek the name is normally, but not always, feminine (see O. Masson, ZPE 16 (1975), p.36 and 20 (1976), p.232); the Romans used it as a masculine name (Solin, loc.cit., cf. Pliny, NH 8.57, Elpis Samius natione) and it is no doubt masculine here. It is perhaps surprising to find a Greek-derived name in north Britain at such an early date.

Iu[: possibly Pu[. Presumably another name. Dark marks visible on the photograph after these letters are probably not ink.

9. ].enum: the trace of the letter after the break best suits r and there is probably room here for Se]renum.

Tetricum: in the editio princeps we read Tetrecum but we are now confident that the supposed cross-bar of e is no more than a chance mark which may not be ink, and that the correct reading is Tetricum with i having a marked serif at the top (cf. uiuas, line 11).

10. c]ontibernales: the spelling with i is common in inscriptions, see the list in TLL s.v. and cf. Väänänen, LVIP3, pp.25-6. The expression is commonplace in the closing sections-of letters, cf. e.g. CPL 251.61-2, saluta omnes contubernales nostrous (sic); and since the Vindolanda text is from a military environment it should be taken quite literally as the members of the addressee’s contubernium (for which see Vegetius 2.13, contubernium autem manipulus uocabatur ab eo, quod coniunctis manibus pariter dimicabant). It is probable that nothing is lost after omn[es; if a supplement were required meos or tuos would suit.

11. o]pto: cf. P.Oxy. 32.28-30. Since our note in the editio princeps this latter text has been improved by J.F. Gilliam, BASP 13 (1976), pp.53-4, opto te felicissi[mum] . . . ben[e ualere. The expression is commonplace, cf. No. 22.14-16 (250). For the use of opto + subjunctive we may compare RMR 98.2.3, opto bene ualeatis.

uiuas: the expression felix uiuas is common on rings and drinking-cups (see TLL s.v. felix, col.444.9 f. and Gilliam, loc.cit.) but we have found no example of the superlative or of the addition of opto (the fragility of ILS 6870, [opt]amus te felicissimum be[ne. uiue]re, is immediately obvious).