Vigiliae Christianae 2 (1948) pp. 185-200




The discovery at Troyes of a new manuscript of Tertullian dating to the 12th century and originating from Clairvaux has been without doubt the most important event for many years for the criticism of these works of this author.  The manuscript contains five of his works, the Adversus Iudaeos, De carne Christi, De resurrectione mortuorum -- otherwise titled De carnis resurrectione --, De baptismo and De paenitentia, in the order given.1 The value of the manuscript is evident in the Adversus Iudaeos and De carne Christi, where the readings occupy an intermediate position between those given by the best witnesses (the Agobardinus and the Fuldensis) and those from the rest of the manuscripts.  The detailed comparison of the different readings is still to be done 2.

The importance of the new discovery is unparalleled for the short treatise De baptismo as this survived in no manuscript: the editions of this work were all based upon the Paris edition of 1545 (B), attributed to Jean Gagny (Gangneius), but due in reality to Martin Mesnart, as noted by Kroymann.  Unfortunately the text of De baptismo is not complete in T: the end, from ch. 18,2 occurrit in tempore, is lacking because some folios of the exemplar followed by the copyist were missing.  But the majority of the work is preserved for us in T, and allows us to actually recover for the first time in many places the exact text 3.

However we can certainly indicate some places where the choice between the readings of T and those of B appears difficult; where both readings, sometimes very different, give a satisfactory sense and can be defended as originating with Tertullian himself 4.  So it is at the start of our text.  In B there is: : Felix sacramentum aquae nostrae quia ablutis delictis pristinae caecitatis in vitam aeternam liberamur! Non erit otiosum digestum istud ('this work', i.e. the one that follows) etc.  But in T the text must be read as : De sacramento aquae nostrae, qua5  ablutis delictis pristinae caecitatis in vitam aeternam liberamur, non erit otiosum digestum istud. On the face of it both readings offer an excellent sense: the exclamation Felix sacramentum in B is paralleled in other places; cf. in De baptismo itself (ch. 15, 3): Felix aqua quae semel abluit! etc.6. The only objection to this reading, which appears more elegant than that of T, is that the next phrase non erit otiosum digestum istud follows in a manner which is a little abrupt and odd compared to that preceding, and objection which disappears if  we examine the version of T, where otiosum ('useless') is connected to de (sacramentum) as is found elsewhere in our author: cf. De oratione 25, 1 (Diercks): ): De tempore vero non erit otiosa extrinsecus observatio etiam horarum quarundam 7.  It can be seen that the two readings, very different as they are, can be defended by reference to other places in Tertullian's own works.  This is what makes the choice difficult.  

Also, based on these places and other parallels, the hypothesis has been put forward that in De baptismo, just as in the Apologeticum for example -- according to current opinion anyway 8 -- we are dealing with two successive editions of the text from the hand of Tertullian himself 9.  And at once the question arises of which of the two must be considered as the later, which the author considered definitive.  However we cannot doubt that the version given by B, whether by Tertullian himself or another, must in every case be considered as the latter of the two: however difficult the decision may be in some places, there is one significant passage which gives certainty.  In ch. 18, 1, the instruction of St. Paul in 1 Timothy 5, 22: Χεῖρας ταχέως μηδενὶ ἐπιτίθει μηδὲ κοινώνει ἁμαρτίας ἀλλοτρίαις 10 is rendered by B in the following manner: : manus ne facile imposueris ne 11 participes aliena delicta; T on the other hand has: manus ne facile inposueritis nec amartiis alienis communicaveritis. Here the version B is without doubt a later correction, as it would be unlikely that the writer of B would have changed it later in order to introduce the word amartiis borrowed from the Greek, which is almost never encountered in all Latinity -- the Thesaurus Linguae Latinae only gives a single example 12  -- in place of an original delicta.  On the contrary any author, whether Tertullian or a corrector of a more recent date, would be inclined to replace this odd word with the normal latin expression. 13 14.

A second passage almost as indicative has been indicated by H. Koch 15.  In Bapt. 3, 2, Genesis 1, 1 is cited as follows, if we follow the redaction of B: in primordio inquit fecit deus caelum et terram. But T gives: In principio in primordio inquit fecit caelum et terram 16.  However Koch draws our attention to the fact that Tertullian himself prefers primordium as in our chapter a few lines earlier17, but wherever he quotes exactly18 he always writes in principio, 'wie offenbar in seinem Schrifttext stand', adds Koch: it admits the obvious point that Tertullian here is using a Latin translation.  The citations from the Gospel of St. John 1,1 offer a good parallel: he always cites this passage literally in the form: : in principio erat sermo 19; on the other hand when he uses his own words, he gives in primordio 20.  It is evident that in our passage the reading in principio must be considered the original, changed later, either by Tertullian himself or by someone else into in primordio, a modification preserved in T by means of the duplication of the reading.

The redaction of B must therefore taken as the younger of the two.  The interest of this information for the constitution of the text is above that it is evidence: if it goes back to Tertullian himself, one must as far as possible follow B; if on the other hand, its variants result from the activity of a corrector, it is necessary in general to prefer T, or at least to examine carefully each reading that is different from T.

Before we proceed to examine in detail the different passages, the question must be posed whether it is really probable that Tertullian produced two successive editions of De baptismo.  There would be nothing strange in him doing so: he tells us himself that he has already handled the same material in Greek, or at least part of it 21.  On the other hand it cannot be denied that our text, even though it contains brilliant passages, perfectly appropriate to the great stylist which our author was, presents some negligence in composition, of the sort which many scholars have supposed indicate a work delivered orally 22.  In fact, a statement like that of ch. 15, 1: Nescio si quid amplius ad controversiam baptismi ventilatur. Sane retexam quod supra omisi, ne inminentes sensus videar interscindere is hardly encountered in a work of which the composition has been previously conceived and established with care: the author would be expected to correct such an obvious negligence in a subsequent revision.  The second paragraph of the same chapter also gives the impression that it was done in haste.  In ch. 12, 3 (Et nunc illis ut potero respondebo) we likewise encounter a transition which is a little brusque and arbitary.  Further passages could perhaps be found which render a second edition by Tertullian himself extremely problematical 23.

It is now necessary to commence a comparative examination of all the variants which are of interest.  We have already indicated 24 two passages where in B the original text has been modified as can be seen from evident reasons.  At the start in ch. 1,1, in speaking of those who come to embrace the Christian faith, the author says: qui . . . temptabilem fidem per imperitiam portant.    Thus the reading of T 25  which makes excellent sense: it is exactly to the young inexperienced Christians that the faith is temptabilis, still exposed to temptation, since they can still be easily ensnared by the objections of adversaries.  In B on the other hand we read in place of these words cited: qui ...intentatam probabilem fidem per imperitiam portant, therefore quite otherwise: it almost seems a polemic against the reading of T!  How can this be explained? Probably some superficial reader, unable to understand that anyone could call the Christian faith temptabilis, reputed incorruptable, has substituted here the words given in B, and that they have penetrated into the text of the manuscript used by Mesnart.  It is difficult to believe that Mesnart himself could have made so poor a correction 26. -- A second example of arbitary correction in B is to be found in ch. 17, 1, where the question of who has the right to administer baptism is treated.  Tertullian says: Dandi (i.e. baptismi) quidem summum (om. B) habet ius summus sacerdos, si qui est, episcopus; dehinc presbyteri et diaconi, so that the supreme right belongs to the most senior ecclesiastical representative present, i.e. the bishop if present; in his absence, to priests, to deacons, and in the last resort, as shown in paragraph 2, even the laity.  Probably following mispunctuation this passage has been poorly understood;  from this we read in B: summus sacerdos qui est episcopus.  That the omission is not accidental is clearly shown by T, where the third 27 hand has also modified the text, suppressing the word est using points and placing dem above it, giving summus sacerdos siquidem episcopus.  The omission of summum in B is the consequence of the correction qui est episcopus, summum no longer making proper sense.  But the real reading is preserved for us by the first hand of T.

Finally a third example of conscious modification of the text 28 is given by B at ch. 10, 2, where it is said of penitence that it is in hominis potestate. T on the other hand reads: in hominis voluntate, a reading admitted as of right by Fr. d'Alès in his edition.   In his explicatory notes he made the following observation: 29: Voluntatis humanae partes Tertullianus pro sua in stoicismo institutione fortiter tuetur, nedum extenuet. Quid humanum sit in paenitentia, plane videt: minus quod divinum, nec partes gratiae distincte assignat. It can easily be understood that a reader, finding that by the words of Tertullian the divine grace is remote as a factor in penitence has changed here voluntate into potestate to leave open the possibility of the influence of this grace in penitence.  But it is necessary to take the trouble to examine more precisely our passage in its entirety. Ch. 10 discusses the baptism of St. John the Baptist.  According to Tertullian (10,2) Jesus himself had already submitted to the Pharisees the question  30 caelestisne is baptismus esset an vero terrenus. But these did not know how to respond utpote non intelligentes quia nec credentes. But, continues Tertullian, since we other Christians have the faith, we can realise that for us divinum quidem eum baptismum fuisse, mandata tamen, non et potestate, quod et Iohannem a domino missum legimus in hoc munus, ceterum humanum 31 condicione: the baptism of John the Baptist was from God, it is true, but only in so much as concerned the mission given to the Baptist, and not in that which concerned the effect, because, apart from being sent by God, he was in other ways an ordinary man.: (10,2) nihil enim caeleste praestabat sed caelestibus praeministrabat, paenitentiae scilicat praepositus quae est in hominis voluntate. Tertullian considers therefore here penitence in its purely human form as it falls within the competence of St. John just as for example the distribution of corn by a Roman official 33.  And it is upon human volition that this penitence depends.Denique legis doctores et pharisaei, Tertullian continues in ch. 10, 3, qui credere noluerunt, nec paenitentiam inire voluerunt 34: which is why (denique) the Pharisees did not repent.  Why not?  Because they did not want to believe 35; 'did not repent' has the logical consequence of 'did not want to believe': he who doesn't want to believe cannot repent, not even want to repent: the repentence does not fall within the domain of his volition, non est in eius voluntate. Tertullian could have said simply: 'the Pharisees did not repent because they did not want to repent'.  But he related at the same time the refusal to repent of the Pharisees with its cause, their refusal to believe.  Seen in this light it cannot be doubted that the reading voluntate at 10, 2 is the correct one which renders rightly Tertullian's idea.

The three passages which which we have treated give cause for reflection, since in all three we see that the original reading, coming from Tertullian himself and safeguarded intact in T, has been modified badly in B, for reasons which we can still understand but which must nevertheless be labelled completely arbitary.  Also the three passages form a solid argument against the theory of a double edition and for the superiority  of the version in T.  There are still some more passages which seem to confirm this superiority.

In the preface of his work (digestum), Tertullian says in ch 1,1 that he is instruens tam eos qui cum maxime formantur quam et illos qui similiter (so T; simpliciter B) credidisse contenti, non exploratis rationibus traditionum, temptabilem 36 fidem per imperitiam portant: the work is addressed both to those who are in the process of instruction in Christian principles and those who have in fact already received this instruction and have been admitted into the community of believers, but whose faith is unsteady because of lack of experience and more detailed knowledge of divine things.  However, we are convinced that the reading of T, similiter, is the right one and we suspect that Mesnart or whoever has modified arbitarily this reading into simpliciter, undoubtedly an attractive word at first sight, because in two ways he has not understood the text.  Firstly the specifically Christian sense of credidisse has escaped him.  We can do no better in this area than to cite the words of Miss Mohrmann who says in an article entitled: 'Traits caractéristiques du latin des chrétiens' 37: 'In the Latin of the Christians one finds from earliest times a credidi with an ingressive nuance and with a sense specifically Christian: "take up the faith, become a believer"  38; she sees here the translation of the Greek aorist e0pi/steusa.  However in our passage Mesnart or another has badly interpreted this credidisse as equivalent to credere; then he has commited the fault of placing similiter -- if this was found in his manuscript also -- in agreement with credidisse, which loses the sense.  This is why he changed similiter into simpliciter, a very ingenious solution at first sight.  But if we join similiter to portant, adding a comma, we obtain a phrase which does make sense: Tertullian distinguishes two groups of people who need to be addressed by his work, namely those under tuition and those with a shallow faith who had contented themselves with becoming Christians, without penetrating more deeply into the reasons for the traditions.

-- At the end of chapter 3 Tertullian says that he does not want to insist on the authority (auctoritas) of water, ne laudes aquae potius -- in the manner of earlier orators 39 -- quam baptismi rationes videar congregasse, as a more details digression can present the advantages.  We juxtapose the text of B and T as follows in chapter 4.


Sed ea satis erit praecerpsisse, in quibus et ratio baptismi recognoscitur, prima illa quae iam tunc etiam ipso habitu praenotabatur ad baptismi figuram, dei spiritum qui ab initio supervectabatur super aquas intinctorem oraturum.


sed ad ea satis erit praeripuisse in quibus et ratio baptismi recognoscitur prima illa qui iam tunc etiam ipso habitu praenotabatur baptismi figurandi spiritum qui ab initio super aquas uectabatur. super aquas instinctorem moraturum.

The text of B is not regular and before the discovery of T attempts were made to correct it using various conjectures which we won't list here.  Let us compare instead B and T.  In place of ea T reads ad ea, which gives an excellent sense ( = πρὸς ταῦτα), when one joins the object required by praecerpsisse (praeripuisse), i.e. prima illa, rightly interpreted in our view by Kroymann as  40 as primordia illa: in Greek one says τὰ πρῶτα ἐκεῖνα.  The choice is difficult between the praecerpsisse of B and the praeripuisse of T; all the same, the syllables ripuisse being written in a correction it is possible that in T also praecerpsisse was to be found in the original 41.  On the other hand one would not hesitate to prefer qui to the quae of B, since the subject of praenotabatur must be singular, i.e. spiritus, while here in general one would put the verb into plural if ea or prima illa is the subject.  Likewise the baptismi figurandi of T is to be preferred to the ad baptismi figuram dei of B: probably there was only (baptismi) figuram di (in place of figurandi) in the codex of Mesnart, and he had to add ad to make sense of it.  The final employment of the genitive gerundive is very common in Tertullian as in Tacitus and many others 42. It goes without saying that in the following the reading of T spiritum qui ab initio super aquas ferebatur, super aquas ... moraturum suppresses, as if by magic, all the difficulties.  And finally, should we read intinctorem with B, a ἅπαξ εἰρημένον according to Hoppe 43, or instinctorem with T? Intinctorem is not without meaning, as one could say that it is the Spirit that baptises.  This is why the conjecture of Rigault intinctorum44 is so often substituted, which is no more satisfactory: it is improbable, in fact, that Tertullian wanted to say that the spirit of God demeurely reposes only on the waters employed by those who administer baptism -- an impossible idea, however phrased! --, but that it reposes on the contrary on all the waters, and that it is by that that the water of baptism can exercise its salutary effect.  However, the text takes immediately the sense which we read with T instinctorem ('instigator, inspirer') 46.  All this passages furnishes afresh a good proof of the superiority of T, although a conscious modification in B is not so evident here as in the preceding cases.

One may say the same thing about the end of ch. 17 (§ 4 f.). Again we compare the text of B and T.


Petulantia autem mulier quae: usurpavit docere utique non etiam tinguendi ius sibi pariet nisi si quae nova bestia evenerit similis pristinae ut quemadmodum illa baptismum auferebat, ita aliqua per se conferat. Quodsi quae Pauli perperam scripta 47 sunt scrip-|p197 


Petulantur aut(em) mulieri[e]s; que usurpant (nt in ras.) docere; utiq; n(on) etiam tinguendi ins sibi rapiet; nisi si q(uae) noue bestie ueneru[i]nt similes pristinae; ut quemadmodum illa baptismum auferebat; ita aliqua p(er) se conferat. Quodsi-|p197 

tum Theclae ad licentiam mulierum docendi tingendi quae defendant sciant in Asia presbyterum . . . confessum id se amore Pauli fecisse loco decessisse.

que acta Pauli que p(er)peram scripta sunt. exemplum tecle ad licentiam mulierum docendi tinguendi que[q;] defend(un)t; Sciant in asia presbites . . . confessum. id se amore pauli fecisse loco decessit;

Here again the text of B is irregular, as is that of T; the latter shows evident traces of modifications, easily explained however: the fault petulantur at the start has been cause for a correction (by a second hand!) from mulieris to mulieres, like that from usurpavit into usurpant. Presbites and decessit at the end are also arbitary changes, caused by the inversion of some folios of the archetype of T 48.  Leaving these aside the text of T shows itself infinitely superior.  At once at the start the original reading of T mulieris, already conjectured by Orsini, is just, contrary to that of B 49; the author of the reading mulier in B had probably seen an ablative in petulantia.  Then the rapiet of T is certainly preferable to the pariet of B: this is clear from the numerous examples of rapere in this sense in Tertullian, assembled by Thörnell 50.  The choice appears more difficult between the singular nova bestia of B and the plural in T; we believe however that it is necessary to prefer here still the reading of T novae bestiae in view of the word following aliqua, which is better understandable if one reads the plural 51. In the following paragraph the greater fidelity of T is shown with still more evidence, in the first place by the words acta pauli, by which is meant the apocryphal Acts of Paul, which are effectively brought to light, and by exemplum tecle as opposed to the scriptum Theclae of B 52.  But also the word quae (after pauli) which is additional in T shows the same thing, seeing how it completes and reestablishes the whole passage.  From this the quae following quodsi is interpreted as a neuter plural; however, it is evident that it is a feminine plural from which is revealed in a single stroke the connection with the preceding paragraph.  Acta Pauli becomes the object of defendunt; it would be better to see as a parenthesis the words exemplum Theclae, which Rigault wanted to exclude from the text.  There only remains to locate the subject of sciant, which is obviously the women who defendunt!   The whole passage now runs like this: Quodsi quae (mulieres sc.) Acta Pauli quae perperam scripta sunt (-- exemplum Theclae!--) defendunt, sciant etc.

-- We see therefore that T gives a better text than B in all the passages we have examined so far -- and one could augment their number further! 53 --, with the exception of 1,1 where it appears that the two readings, that of T (De sacramento) and of B (Felix sacramentum), can both be reasonable defended 54.  We are nevertheless right to admit here also, if that is possible, that the reading of T is that of the original, since elsewhere T has also conserved the original text.  But how do we explain the manner in which B acquired here a reading so totally different, which cannot derive in any manner from those words which are given by T? There is only one explanation we can give: it must perhaps be admitted that in the manuscript employed by Mesnart he did not find either Felix sacramentum or De sacramento, but only sacramento or e sacramento, since a space had been left blank for the miniaturist painter to add later the word De or the letter D 55.  It was probably Mesnart himself who invented the phrase Felix sacramentum, based in the first place on the model of ch. 15, 3 Felix aqua.  Why he did not adopt the reading of T, if he knew the latter, is explained perhaps by the fact that in T there is a heavy punctuation mark after the words De sacramento of the kind which, on a superficial reading, gives the words the appearance of a title for the whole work.  Whatever it may be, in other cases also Mesnart has mistaken the real readings of T.  We thus arrive at the end of our enquiry which, it seems, demonstrates the superiority of the codex Trecensis, compared to that manuscript followed by Mesnart which we know from his edition, and the extreme improbability of a second edition by Tertullian himself.  It follows that all those to come who will prepare a new edition of De baptismo 56 will do well in principle to follow everywhere the readings of T in the part of this work which is preserved in that manuscript, even in the few passages where the reading of B, taken by itself, offers a satisfactory sense; only in places where T shows obvious defects can B be preferred.

La Haye, Waalsdorperweg 217.

[Notes  have been moved from the bottom of the page to here]

1 The manuscript (no 523, indicated below by T) was discovered by the benedictine monk André Wilmart in 1916.  Cf. his article 'l'Ancienne bibliothèque de Clairvaux' in Mémoires de la société académique d'agriculture, des sciences, arts et belles lettres du département de l'Aube, 1917, p. 167; then the same author in Anal. Boll. 38 (1920), pp. 241-284.

2 Cf. provisionally Kroymann in CSEL vol 70 (1942), pp. XIII and following.

3 With the aid of T we have reconstructed the correct reading for several passages in Philologische Wochenschrift 51 (1931), col. 251--55, followed in our edition published at Leiden in 1931.  Without knowing this Fr. d'Alès published a new edition of De baptismo (Rome 1933) in which he also used T.  We regret to have to say that his edition does not show very much progress compared to ours.  In fact the author has not absolutely understood the value of T, so that he has omitted many variants of T and has not profited from those he has mentioned.  It may be said that he felt himself that he had misunderstood the value of T, since he says on p. 5 of his preface: Quod si quis censuerit ex illis lectionibus quas negleximus potuisse unam aut alteram palaeographo magis exercitato videri non indignam mentione, non acriter contradicemus.

4 W have assembled a certain number of these passages in the preface of our edition, p. 7 ff.

5 This qua (as opposed to the quia of B), already conjectured by Orsini, is of course the correct reading.

6 Cf. in other works: De carnis resurrectione 26 (p. 63, 7 Kroymann): Felix nimirum fides si ea consecutura est etc.; De oratione 2,3 (1. 8 Diercks): Felices qui patrem agnoscunt! Ad Uxorem I 3 (1. 44 Kroymann) : Felicem illum qui Pauli similis extiterit!

7 With an infinitive Apol. 10,6 (1. 24 Hoppe): Otiosum est etiam titulos persequi; De pat. 5 (p. 6,11 Kroymann): procudere disputationem de necessariis fidei non erit otiosum.

8 This opinion has been defended above all by the Swedish scholar Thörnell in his Studia Tertullianea IV (1926); cf. Hoppe in the preface of his edition in CSEL 69 (1939), pp. XXXVII--XLVII. It seems to us that this hypothesis is not so certain that one would not like to see today a fresh examination. -- One thinks also of the books Adversus Marcionem where a portion of the first three books has been rewritten by the author no less than three times.  But in this case there was each time a very good motive for a new edition of the text.  Cf. the latest study on this matter in Quispel, De bronnen van Tertullianus Adv. Marc., Leiden 1943, pp. 3--21.

9See F. J. Dölger, Antike u. Christentum 2 (1930), p. 118.

10 In the Vulgate: manus cito nemini imposueris neque communicaveris peccatis alienis.

11 For the second ne one should certainly read nec.

12 Located in a Christian inscription originaly from the cemetary of Callistus in Diehl Inscr. lat. Chr. veteres 1558 1. 10: ut possit amartias meas indulgere. Cf. also the title of a poem by Prudentius: Hamartigenia.

13 We cannot resist remarking that if one part of our passage seems to strongly support the opinion of those who like G. J. D. Aalders, Tertullianus' citaten uit de Evangeliën en de oud-latijnsche bijbelver-talingen, Amsterdam 1932, and G. Quispel, op. laud., pp. 104 et ss., believe that Tertullian translated directly from Greek, on the other hand the same passage and those which we examine after make us look with reserve at the parallel research into the uncertain character of the text of the citations of the Bible.

14 Another demonstration of the later date of B would perhaps be the singular imposueris and participes as in the Greek text, against the plural in T.  In fact it seems more probable that a correcting hand, whether the author himself or another, has tried to make a faulty translation agree with the original which says the opposite.

15 In a review of our edition of  De baptismo in Theologische Literaturzeitung 1932, col. 589.

16 In T deus is therefore missing, omission without importance for our point.

17 And again in 4, 2 and 5, 4.

18 Adv. Hermog. 3 (p. 129, 13 Kroymann); 20 (p. 149,5 Kr.); 26 (p. 154,9 Kr.); Adv. Praxean 5 (p. 233,3 Kr.).

19 Adv. Hermog. 20 (p. 149, 5 Kr.); Adv. Prax. 13 (p. 247,13 Kr.); 21 (p. 263,23 Kr.).

20 Adv. Prax. 5 (p. 233,15 Kr.); 16 (p. 256,11 Kr.).

21 See ch. 15,2: Sed de isto plenius iam nobis in graeco digestum est.

22 Cf. E. Noeldechen, Die Abfassungszeit der Schriften Tert.s. T. u. U. 5, 2 (1888), p. 366; P. Monceaux, Histoire litt. de l'Afrique chrétienne I (1903), p. 366; J. A. Knaake, Die Predigten des Tert. u. Cyprian, Theol. Stud. u. Krit. 1903, p. 631.

23 This is why we have now rejected ourselves the theory of Dölger of a double edition by Tertullian himself (see the preface of our edition, p. 8) and the other, namely that the equivalent variants of B and T go back to Tertullian himself who wrote in his own exemplar variants below or next to the next; later these variants penetrated into the text, sometimes in combination with the original reading, in later copies of the text.  It is clear that for the constitution of the text it comes to the same whether one prefers to follow Dölger's theory or our modification of it.

24 In the preface of our edition, p. 7.

25 Only T gives imperia iam in place of imperitiam, a paleographical fault easy to explain.

26 In T also there are such faulty correction, e.g. in ch. 3, 1, where Tertullian says:  Nihilominus quam stultum et impossibile sit aqua reformari rettactemus (so Bm apart from the correction by the conjecture proposed by us of retractemus in place of fractemus T or tractamus B).  In T the middle words are changed into stulte sed possibile by a reader who didn't understand the irony of the original reading. But sed (s;) has been written by another hand in a space!  Probably T has therefore the correct reading. -- The reader can find other examples of faulty corrections in T, even by the first hand, in the preface of our edition p. 6.  But they result always from a temptation to eliminate a reading considered as corrupt, e.g. in ch. 13, 2, where T gives nec potentiam habuit sine, where B says: nec potest iam sine, an expression proper to the style of the author which has not been understood.

27 The variants of this hand have no value for the recovery of the text: see the preface to our edition, p. 7.

28 A modification which we have misjudged elsewhere: see our edition, p. 7. 29 p. 36.

30 Matth. 21,24 f.

31 humanum refers to Iohannem, not to munus!

32 A little further on (10,3) Tertullian says himself: Quodsi paenitentia humanum est ('a human matter'), et baptismus ipsius (paenitentiae sc.) eiusdem (i.e. humanae) condicionis fuerit necesse est.

33 Note the term praepositus!

34 Thus T; B gives nec paenitentiam agete. But one may easily understand that the more familiar  expression paenitentiam agere has replaced this less used expression, which was found in the original.

35 Earlier on, in ch. 10, 1 Tertullian had said: quia nec credentes. It is here alone that the voluntary element is introduced.

36 On the reading temptabilem see above, p. 5.

37 In Miscellanea Giovanni Mercati, Studi e testi 121 1 (1946), p. 25 of the extract.

38She cites as an example ch. 17, 3 de la Passio Perpetuae et Felicitatis -- a work often attributed to Tertullian --: ita omnes inde adtoniti discedebant e quibus multi crediderunt.  From Tertullian himself she cited apart from our passage ch. 13, 3: tunc et Paulus, ubi credidit, tinctus est.

39 Cf. the letters of St. Jerome 69, 6, 1: de schola rhetorum aquarum laudes et baptismi, which imitates here our work of Tertullian as he does regularly in this letter.

40 Quaestiones Tertullianeae criticae, Innsbrück 1894, p. 74.

41 Tertullian has used this word elsewhere in Adv. Prax. 3 (p. 230, 23 Kr.): at ego, si quid utriusque linguae praecerpsi, monarchiam nihil aliud significare scio quam singulare et unicum imperium and De anima 7,4 ( Waszink) : si quid tormenti sive solacii anima praecerpsit in carcere ... inferum, thus in a different sense. -- There is praeripere in Adv. Marc. III 3 (p. 379, 12 Kr.), likewise in a different sense which does not help with this context.

42 Although Hoppe in his Syntax u. Stil des Tert., Leipsic 1903 does not discuss the word. Often found in combination with a substantive as in Ad Nationes I 2, 2(Borleffs): praesides extorquendae veritatis, De anima 10,5 (Waszink) : pabuli trans mittendi ... membra and elsewhere; but also more by itself, e.g. De anima 19, 1: viam sternunt postea inducendi eius. A good example is found in De paenit. 7, 13: iterandae valitudinis iteranda medicina est as is read in the manuscript of Troyes; the other manuscripts and the editions give iteratae valetudinis etc.

43 Beiträge zut Sprache und Kritik Tert.s, Lund 1932, p. 137.

44 Other conjectures are intinctos of Gelenius and in tincto by Kroymann.

45 moraturum; here again T has conserved the true reading, established however long ago by conjecture.

46 Instinctor, a word found after Tacitus (Hist. I 22,3; IV 68,4) in Ammianus Marcellinus (XXI 12,20; XXX 1,2) and others, found also in Tertullian Adv. Marc. II 10 (p. 348,4 Kr.): in diabolum ... ut in instinctorem delicti.

47 The reading scripta (in place of inscripta), corroborated now by T, has been defended by  Löfstedt (Zur Sprache Tert.s, Lund 1922, p. 79), who compares Ps. Asconius in divin. 25: in libris qui De oratore scribuntur. He was also able to cite Suetonius, life of Nero 11: inducta Afrani togata quae Incendium scribitur, which since Erasmus is corrected to inscribitur, and from Tertullian himself De praescr. haer. (1. 25 Kr.): (Apelles) quae ab ea (Philumena) didicerat, Φανερώσεις scripsit, from which one would greatly hesitate to adopt the plainly arbitary  alterations brought into the text  by Kroymann, and De carne Chr. 12 (1. 28 Kr.): libellas quem scipsimus De testimonio animae. Compare also St. Jerome, letter 112,3,2: legisti enim et graecos et latinos qui vitas virorum inlustrium descripserunt, quod numquam Epitaphium huic operi scripserint, sed De inlustribus viris; ... Epitaphium autem proprie scribitur mortuorum. Here St. Jerome construes even the verb with a dative!

48 See our edition, p. 4.

49 This woman is the de Caina haeresi vipera of chapter 1, 2, who professed that baptism must be suppressed.

50 Op. laud., p. 118 f. It is enough to cite, among the passages which he has listed De anima 25, 3: (fetus) rapiens sibi iniurias matris and Adv. Marc. III 8 (p. 388,28 Kr.): Marcion sibi eam rapuit praesumptionem.

51 On the other hand, if Tertullian used the singular he probably didn't write aliqua but haec (opposed to the illa which precedes it).

52 It is true that exemplum is found already in the editions following Gelenius who printed it from a marginal note in the edition of Mesnart. But these notes in B (in De baptismo at least) come from T which Mesnart consulted together with a more complete manuscript which he followed for preférence. See our edition, p. 5; Kroymann CSEL vol. 70, p. XX.

53 Starting with those we have treated already in Philologische Wochenschrift: v. p. 1* n. 3.

54 See above p. 2.

55 The omission of the first letter of a book or of a new paragraph in the manuscripts or in the  incunables is a phenomenon so frequent that it is hardly necessary to cite examples of it. We will restrict ourselves to cite a single one.  In the so-called Quadratus of Lucretius, in the library of the university of Leiden (Vossianus. l. 94, of the 10th century), this phenomenon is encountered repeatedly.  Thus e.g. I 1052 (f. 10r a) is found llud, without the required i, V 1 one reads uis potis est, lacking  the Q, and so on. Also the missing letter has often been added in a small character by another hand at the beginning of the line, apparently with the intention that the  miniaturist could design a new letter later, which never happened: see for example II 89, 142, 221 etc.; on folio 13v we have five examples at a time (II 388, 392, 398, 404, 408). At the start of book III the title and the first word have been omitted, so that the first verse begins with tenebris; another hand has places an A before it, however wrongly, as the real reading is probably E tenebris.  It is curious to note that in the codex Oblongus of the same author likewise found at Leiden ( Vossianus F 30), one finds in this place O tenebris, which isn't right either: from which one may deduce that the word E was already missing in the archetype of OQ, which must certainly be datedmany centuries earlier.  -- Sometimes the little letter remains in the text together with the big letter in fact added by the miniaturist: the Trecensis of Tertullian gives us a good example.  In this manuscript the treatise De baptismo finishes with the following words of ch. 18, 2: scriptum ipsius fidei. Then one reads: explicit de baptismo. Incipit de penitentia; hoc genus hoies. quod (et) ipsi retro fuimus. These are the first words of De paenientia; the rest of the line has been filled by a line[? 'trait'].  However the remaining part of the phrase follows on a new line: Cceci sine domini lumine etc., as if the text of De paenitentia only began here.  The first C has been drawn very large by the miniaturist, the second is the same height as the normal letters, but thinner.

56 A parallel edition from us is in preparation from Daamèn,La Haye (with De patientia et De paenitentia).

© Brill Academic Publishers, 1948.  All rights reserved.  Reproduced by permission. Greek text in unicode.

This page has been online since 18th June 1999.

Return to the Tertullian Project Original Language About these pages