1 For Octavius and Candidus see the discussion in the introduction.
3 a Marino: the name is well attested, but has not so far appeared elsewhere in the Vindolanda texts. We have considered and rejected the possibility of reading a name Amarino.
nerui: presumably genitive singular, the use of the word indicating "animal tendon etc., used as material" (OLD, s.v.2): cf. Vitruvius 1.1.8, per quae tenduntur suculis et uectibus e neruo torti funes; Tacitus, Ann. 2.14.3, non loricam Germano, non galeam, ne scuta quidem ferro neruoue firmata. One hundred pounds of this material seems a considerable quantity but it is not out of keeping with the quantities of other commodities mentioned in this letter. A.R.Birley (1991), 92 raises the possibility that we should read neruio and understand it as an ethnic. The putative o, however, is demonstrably an offset from q of quam in ii.14.
4 explicabo: our statement in the ed. pr. that this is a problematical usage was perhaps over-cautious and the evidence of its use in other Vindolanda texts (see below) provides some reassurance. explico had a well-established financial use (TLL V.2 1731.17) of sorting out, settling a debt, financial obligation or difficulty, which would fit the context here. The letter is full of financial terminology (cf. lines 12, 23, 39ff., perhaps 31-2). In this sense explico is used absolutely, or with a sum of money as object, or with a variety of words as object, indicating the debt, burden, account, etc. It is common in Cicero's letters to Atticus, e.g. 5.5.2, 12.24.3, 12.31.2, 13.29.1, cf. B.Alex. 34.2, Suetonius, Dom. 12.1, Digest, 42.1.31. There is a good deal of flexibility to this usage. The object of the verb need not define the debt in strict financial terms, but may merely express the general "burden" or "business" to be "sorted out" financially (note Cicero, Att. 12.24.3). It occurs, moreover, in this sense in two other contexts in the Vindolanda texts with reference to items which cannot, literally, be unravelled (note especially 301.i.5, 316.margin 2, 3). This, together with the recurring preoccupation with financial matters in the present text, weakens the suggestion by A.R.Birley (1991), 92 that we should understand the verb in its literal sense.
4 e quo: sc. tempore, "from the time when"; Augustan and later, cf. TLL V.2 1090.65. The end of the line after hac has been left blank because of the tie-hole; similarly lines iii.29 and iii.31. Lines ii.16, 19, iv.40 and 42 are indented at the start for the same reason.
5 scripseras: epistolary pluperfect, cf. line i.7, scripseram, line iii.30, fuerat, line iv.35, constituerat.
ne: = ne ... quidem, a vulgarism mentioned by Quintilian 1.5.39, who calls it detractio. It is also found in 311.i.6.
mentionem: the reading is inevitable, although on is somewhat difficult; o is made in three, rather straight strokes, left open at the bottom and ligatured to n, which makes the letter combination look rather like cul; but no such word exists.
7 spicas: the reading is not quite certain - i is not easy and p is very oddly written if correctly read; nevertheless the word must be accounted very probable in the context. Normally it means ears of corn, which we assume to be meant here, but it can refer to other cereals (see OLD, s.v.). Does he use this rather than frumentum because he is buying it unthreshed (cf. lines iii.26-9)?
8 prope: the reading of op is far from certain because the leaf is badly defaced by offsets at this point; indeed of the word as a whole only the final e is beyond question. 5000 modii is a very large quantity of grain (though he may have bought it unthreshed, see previous note). It is unfortunate that there is no indication as to where Octavius got it from, but the quantity and the financial transaction suggest that army supply was not always a straightforward matter of requisitioning annona. For some further evidence in the Vindolanda texts see 180, 213.
9 necessari: = necessarii. A standard contraction in the colloquial language: for further examples in the Vindolanda material see Tab.Vindol.I, p.73, and gladis in 164.3.
10 nisi mittis mi aliquit: in the ed. pr. we considered mittas a possible but less good alternative; we are now confident that mittis is correct. The future mittes might have been expected, but in fact conditionals of this type (with present indicative in the protasis and future in the apodosis) are not uncommon, particularly after nisi. The present stresses the need for immediate action to avert disaster; it is well-suited to threats and warnings, e.g. Cicero, Verr. 4.85, Rab.Post. 18, Nepos 15.4.3.
12 quod arre dedi perdam: for the monophthongisation of ae in arre cf. que (ii.15), illec (ii.19) and probably male (ii.21, see note ad loc.). The terminology is financial. For arram perdo see Digest 18.3.8 ut arram perderet, and for arra associated with do see Gaius Inst. 3.139 ... cum de pretio conuenerit, quamuis nondum pretium numeratum sit, ac ne arra quidem data fuerit. nam quod arrae nomine datur....
13-4 erubescam: no doubt idiomatic (if to be understood as financial embarrassment), but we have found no precise parallel.
14-5 ita rogo ... mi mitte: rogo (plus) imperative is colloquial syntax, with the direct construction (the imperative) used instead of subordination, see e.g. Petronius, Sat. 67.1, 75.3, Martial 2.14.18, P.Mich.VIII 469.17 = CEL 144, [m]erca minore pretium [sic], rogo. 15-8. A.R.Birley (1991), 94 infers from this passage that Octavius was writing from somewhere to the north or west ("east" is a typographical error, as Prof.Birley kindly pointed out to us) of Vindolanda since the latter lay between him and Catterick, but what Octavius writes here need only imply that Candidus was in a position to instruct someone at Catterick to send the hides and the wagon to Octavius; one possible explanation for this is that Candidus was in the military and Octavius was not.
15 For que = quae see above, ii.12 note.
16 Cataractonio: see Tab.Vindol.I, pp. 72f. on Luguualio in 250.i.9 and cf. above, pp.43-5 and Adams (1994). For Catterick as a centre for leather-processing see the introduction.
16-8 scribe dentur mi et karrum de quo scribis: the plain (jussive) subjunctive without ut follows the governing verb; for another example in this letter see lines iii.31-3, desiderabat coria ei adsignarem. If the meaning is "write that they should be given to me, and that the carrus about which you write (should be given to me)", karrum could be the neuter by-form of the usual carrus (see B.Hisp. 6.2 carra complura ... retraxit and cf. 315.2 note). Alternatively, if it is masculine accusative, the construction would be a constructio ad
sensum, with the accusative determined by the underlying idea that someone should "give" the carrus to Octavius.
18 quit: cf. aliquit (i.10, 14), si quit (iii.28), contrast d in quod (i.9, ii.12).
19 illec = illaec, neuter plural, cf. ii.12 note.
petissem: in the ed. pr. we preferred to read cepissem; there is little to choose but we now think that the third letter looks more like t than p. Either offers acceptable sense.
20 nissi: for this spelling see also Tab.Sulis 32.7 and 65.10 and notes ad locc.
iumenta non curaui uexsare: for curo with the infinitive (as distinct from ut) see TLL IV 1499.43ff. and cf. iv.36-7, nec curauit accipere. uexo was idiomatic of a horse or other quadruped hurting itself in the course of work or a journey, see Schol.Iuu. 8.148, Pelagonius 216.2, Mulomedicina Chironis 671. In uexsare the a and r are written very close together and there is some abrasion, but there is no doubt that both letters are there (pace A.R.Birley (1991), 95).
21 dum uiae male sunt: we take male as a misspelling of malae (see note to i.12). It is unlikely to be the adverb: the usual complement of male est is a dative (of a thing or person affected), expressed or understood (TLL VIII 237.7, OLD, s.v. male 1b).
21-2 uide ... de: for this idiom cf. Cicero, Att. 11.24.2, 12.6.1, 15.8.2.
Tertio: a possible alternative reading is Certio but the name Tertius is very common (see A.R.Birley (1991), 94).
Fatale: perhaps the same person as ].io Fatale, the sender of 349. A.R.Birley (1991), 93 suggests an identification with Claudius Fatalis who is known to have served three terms as a centurion in British legions, but if so he is not identical with the sender of 349 since his gentilicium cannot be Claudius (see back 1 note).
23 non illos mi accepto tulit: between mi and accepto the writer has left a blank space of 30 mm., presumably because the piece now missing from the bottom of the tablet was already missing when the letter was written. The expression is a variant of the classical financial formula aliquid mihi acceptum (re)ferre (for which see TLL I 314.13, OLD, s.v. fero 24b). accepto is the substantivised participle, acceptum = "thing received", pecunia accepta (so TLL I 321.82). The case can only be predicative dative.
24 scito mae explesse: the reading of the first two words is far from certain, but the sense produced is exactly that required. For spellings showing ae for e see Coleman (1971), 186-90, esp. 189, citing sae = se in CIL 3.8412, cf. ChLA X 416.ii.19 and Tab.Sulis 94 with Adams (1992a), 10. For scito me in letters cf. Cicero, Fam. 2.15.1, O.Wdi Fawkhir 1.14-5 = CEL 73. For explesse cf. O.Wdi Fawkhir 1.7 = CEL 73, explesti iiii matia.
25 bracis: for other examples of this Celtic word at Vindolanda see 191.16, 348.2. It is a kind of cereal used in making Celtic beer (see Tab.Vindol.I, p. 96, cf. Pliny, NH 18.62), but of "genre inconnu" (André (1985), 37).
excussi: excutio (lit. "shake, strike (something) out of (something else)") could be used of threshing, cf. TLL V 2 1309. 32-6: see especially Columella 2.10.14 nam semina excussa in area iacebunt, superque ea paulatim eodem modo reliqui fasciculi excutientur, cf. Varro, Rust. 1.52.1. It was not a technical term for one particular method of threshing, but a general term for the separation of one part of a crop from the rest by shaking, striking or the application of pressure.
26 In the ed. pr. we were fairly confident about the reading of the numeral as cxi, written twice, although we pointed out that the dittography is unexpected (the dittography in lines 32-3 is much more easily explicable). We now think it possible that most of the extra traces which we were reading are offsets and that cxix is more likely.
fac ... mi mittas: fac (plus) subjunctive is common in the colloquial language and epistolography from early Latin onwards; see e.g. Cicero Att. 2.14.1, cf. CPL 256, fac itaque emas.
26-7 possi/m: the word is inescapable in the context but the ending is difficult to read. If we suppose the writer put im at the beginning of line 27, both the letters are oddly made (for the word division cf. uentur/um in lines 35-6). We now prefer the alternative possibility which is to take the trace at the end of line 27, which goes through the o at the end of the line below, as ink (and not offset) and read possi/m; such a word division is very bizarre but perhaps credible in a letter written by someone of limited literacy.
27 spicam: at i.7 the plural was used but here the writer prefers the collective singular. Botanical terms are frequently used in the collective singular (e.g. faba: for singular and plural examples see TLL VI.1 2.52 and for singular examples at Vindolanda 192.3, fabae, 302.1, margin 3, fabae and oliuae).
27-8 in excussorio: excussorium is a neologism but its formation is regular and its meaning clear. It is synonymous with area, indicating the place where the act of separating grana from spicae takes place; at Varro, Rust. 1.52.1 excutio is used of this process, and area is used of the place where it is carried out.
autem: here it seems to be close to enim in meaning. On autem = enim (mainly in later Latin) see Hofmann and Szantyr (1965), 490-1.
29 perexcussi: a possible alternative reading is ter excussi, which is supported by the gap the writer has left between the r and the e, but which is, we think, on other grounds a less attractive reading. This is by far the earliest occurrence of perexcutio.
31 ei: in the ed. pr. we understood this as = sibi (so also CEL, note ad loc.); for another example at Vindolanda see 250.i.4. A.R.Birley (1991), 93 points out that it could equally well be taken as referring to Frontius, i.e. Frontius' contubernalis asked for the hides to be assigned to Frontius. We see no way of deciding between these alternatives.
36 interuenit: this seems to mean "(did not) turn up", though it usually refers to an unexpected or chance arrival (e.g. Terence, Phorm. 91, cf. TLL VII.1 2299.73). In 291.i.5-ii.8 Severa looks forward to the interuentus of Lepidina.
38-9 Frontinium Iulium: a new sentence must begin at this point. In the ed. pr. we understood the name as an example of the archaising inversion of gentilicium and cognomen, but A.R.Birley (1991), 91 points out that Frontinius is a good example of a "fabricated" gentilicium and that Iulius is often found as a cognomen.
38-41 This sentence is baffling. pro coriatione is a major crux and it is by no means the only difficulty.
There is an active liceo and a deponent liceor. The active has two main uses: (a) = "to be for sale" (TLL VII.2 1357.61), e.g. Cicero, Att. 12.23.3, de Drusi hortis, quanti licuisse tu scribis, id ego quoque audieram; with the genitive of value quanti here, contrast the ablative of price at Seneca, Contr. 1.7.3, magno licet (magno in iv.39 must be this same usage); (b) with a personal subject = "have for sale", with accusative of the thing offered for sale (TLL VII.2 1357.71), e.g. Pliny, NH 35.88 quanti liceret (pictor) opera effecta. The deponent means "bid for" (at auction) (TLL VII.2 1357.81), e.g. Seneca, Contr. 1.2.4, in auctione nemo uoluit liceri. Octavius has presumably used the first verb (rather than the deponent in an active form). If comparauit means bought, as seems likely (for this common sense see TLL III 2011.26, OLD, s.v. 3b; cf. e.g. Spanish comprar), there must be a contrast expressed between buying at a certain price and selling, offering for sale, at a different (higher) price. If so, the meaning of magno licere pro (assuming that pro is the preposition) would be something like "is asking a high price for". It is a difficulty that liceo is not attested with a pro-expression as complement, but on the other hand the verb is not frequent, and one cannot be certain about its range of uses.
coriatione would then have to be explained as the ablative of a noun coriatio, unattested in Latin at this date. The word coriatio is cited in RMLW (15th cent., no reference given), with the meaning "covering with leather". This is discussed by Petersmann (1992), 288, leading A.R.Birley to withdraw his suggested emendation ((1991), 95, VRR II, 60). The word would presumably have to be understood as an abstract verbal noun which had taken on a concrete meaning, a commonplace semantic development. Compare, e.g., uulneratio: this will at first have meant literally "the making of a wound", and then have acquired the concrete sense "a wound that is made" = uulnus; similarly coriatio (which would be a derivative of corio(r)) might originally have meant "the making of leather" (abstract) and then acquired the concrete sense "leather that is made"; perhaps cf. also tensiunculam, 239.3 and note. There still remains the problem of quem following. This would have to be taken as a masculine relative form used for the feminine. It used to be thought that this usage was late (see, e.g., Löfstedt (1911), 132). However, there is now a parallel in one of the tablets from Murecine, dated to AD 39 (TP 18.2.8.-9, Wolf and Crook (1989), no.5), in which C.Novius Eunus writes "quem suma iuratus promissi me ... redturum" [sic], where quem suma = quam summam, an expression which occurs in the more correct version of the document (TP 18.5.7). The masculine form was eventually to replace the feminine entirely (cf. Fr. qui, masc. and fem.). See also CEL 7.II.20 with Cugusi's note ad loc. The expression comparauit (denarios) quinos seems to contain an accusative of price ("which he bought for five denarii apiece"), a construction which occurs at Petronius, Sat. 43.4, uendidit enim uinum, quantum ipse uoluit, and perhaps in P.Mich.VIII 469.17 = CEL 144, with Adams (1977), 40-2, where further bibliography is cited. It is odd that the distributive quinos is used in association with the singular relative quem, though the antecedent may have a collective sense, and the relative clause would then contain a sort of constructio ad sensum. In any case distributives were by this time used for cardinals. This explanation of these lines was offered in the ed. pr. with some diffidence but we cannot suggest any substantive improvement. It is far from satisfactory to have to bring into existence at this date a verbal noun with concrete meaning which is attested only later (based moreover on an unattested verb corio(r)), a new use of licere, and an early example of quem = quam. The discovery of a fifteenth-century example of the word unfortunately throws no light on the major problems posed by our passage, namely the sense of the whole sentence and the presence of the masculine relative form immediately after coriatione. The reading of the lines seems more or less certain, but it is possible that a different word-division might throw new light on the passage. In the ed. pr. we considered the possibility of taking cori in coriatione as a genitive, but saw no obvious noun which could be extracted from the following letters on which cori might depend.
42-3 Spectatum: A.R.Birley (1991), 94 points out that there are no other examples known in Britain. This strengthens the case for identifying this man with Spectatus of 180.5, who must have been at Vindolanda. If this is correct, it makes it virtually certain that the letter of Octavius was sent to Vindolanda (see above, pp.43-5).
I.../rium: no doubt another proper name. In the ed. pr. we read the second letter as m and we still think this is what it most resembles; however, the only name we find attested which would suit this beginning and ending is Imbrius and this is certainly not a plausible reading here. A.R.Birley (1991), 91 has suggested the very common cognomen Ianuarium, accepted in CEL. The writing after i could be taken as a in this hand but it is very hard to reconcile what follows this with nua. Despite the palaeographical difficulties we would not wish to discount the possibility that Ianua/rium should be read and we have nothing else to suggest.
Firmum: presumably to be identified with Firmus of 180.23.
44 Gleucone: cf. A.R.Birley (1991), 94.
45 ual(e): probably this is all that was written in this line. The traces of ink at the right are offset and it is almost certain that the word was abbreviated, cf. 312.ii.12, 349.margin 1, 505; see also ChLA X 428.10, on which Schubart's original view is preferable to that of Marichal (so also CEL 3).
46 uindol: we are now uncertain whether this is written by the same hand as the letter on the front. The abbreviation is paralleled in 338.back 1 (see note).