Relationships among the witnesses
 to 3.1 Dial., 1

John Kilcullen

The witnesses are: Mz, Fr, Frm, Pz, Lm, Ly and Gs. A collation of these witnesses appears in the bottom frame (which can be enlarged or reduced by dragging the border between the frames).

In William Ockham, Opera Politica, IV (Oxford University Press for the British Academy, 1997), p. 7ff, Professor Offler has analysed the relations among the witness to the text of Compendium Errorum, which are the same as for 3.1 Dial. As he remarks, "The weakness of CE's textual tradition is not auspicious for the future editor of IusIIae Dial." (i.e. 3.1 Dial.) (ibid. p. 11). Offler postulates as sources for our witnesses several non-extant MS, but in my opinion his arguments are not strong. In particular, I am not persuaded that Lm had any other source than Pz, or that Ly had any other sourse (for 3.1 Dial.) than Pz (though I do believe that the editor had access to other sources for 1 Dial. and 3.2 Dial. --- see here.). For criticism of Offler's arguments, see note.

Gs and Ly

All the variations are consistent with the hypothesis that Gs is a reprint of Ly without "contamination" from any other witness. The variants found in Gs and not in Ly, including the few in which Gs improves on Ly, are all explainable by accident or conjecture. They are as follows (the link is to the top line of the collation): dominis, statuit, se, iudicio, naturali, quia, hec, imponere, maiori, nominantur, ducimini, et, qui, conversionem, habeat, 20, 23, sicut, debent, quemquam, quam, apostolorum, alienus, non, eciam, erant, and rerum.

Pz, Lm and Ly

The most likely hypothesis is that in 3.1 Dial. Lm and Ly derive from Pz and no other source. The best evidence for the hypothesis that Pz is the source of the others is at ecclesie and successoribus. In Pz these points come respectively at the first and the last line of the same page. I suppose that by some printing shop accident the type set for the first line (ecclesie ad huiusmodi sancte sedis examen) was moved down to be after what should have been the last line of the page, ecclesie being replaced by cessoribus (to complete suc- at the end of what should have been the last line), which had already been set at the top of the next page. (See image of page opening, which shows the tops and bottoms of the two columns of the left hand page and the first few lines of the left hand column of the facing page.) Lm and Ly agree with Pz in this transfer of material, except that Ly  (followed as usual by Gs) tries to make better sense of the second occurence of cessoribus by a conjectural amendment to ab antecessoribus. It is only in Pz that these two passages are found at the beginning and end of a page and no explanation offers itself besides a printing accident. Without supposing that Pz is the source of the others there seems no way of explaining why these MSS agree at these points.

There is no need to suppose that Lm and Ly in 3.1 Dial. have any other source besides Pz. Lm sometimes differs from Pz, but in every case the difference is easily explainable as accident or conjecture: dicent, qui, ex, que, 9us, possunt, apostolis, negociis, auctoritatem, leviter, vel, exigendi. In many more cases Ly differs from Pz, but again the differences can be explained as accident or conjecture. The editor of Ly seems to have thought about the text carefully and made many changes (sometimes erroneously). In some cases he may have consulted a copy of the Canon Law or the Bible, but most of the changes seem to be conjectural (prompted for example by grammatical anomalies in Pz: videtur, de quarum... utramque, habeant, nitere, regimini). The possibilty that the editor consulted another manuscript of the Dialogus cannot be excluded, but there seems to be nothing to compel or support such a supposition. Some of the cases where Ly differs from Pz are as follows.

3.1 Dial., 2.11-12 provides a very interesting case in which Ly differs from Pz, Fr and Mz, in which these two chapters are not found at all. Those chapters are somewhat confused, which makes one doubt their authenticity, and they could well have been written by the editor on the basis of "back-references" in later chapters. However, the possiblity cannot be absolutely ruled out that the editor of Ly found them in some source other than Pz. Similarly, part of 3.2 Dial 1.6 ("Discipulus: Ut expressius . . .", and the Master's answer) is found only in Ly, and it seems very possible that the editor wrote it himself.

Offler mentions (ibid., p. 10, n.6) as confirming the view that the editor of Ly consulted a now lost manuscript Baudry's comments on Mz (Baudry, Guillaume d'Occam, p. 259). Baudry points to four places in which Mz and Fr contain notes warning that there is something missing from the text. In the third of these Ly does not contain any of the missing material, but in the other three contains material not found in Fr or Mz. One of these has just been discussed (3.1 Dial. 2.11-12). In another place Ly adds a few words to complete a sentence, something the editor could well have done out of his own head. The fourth passage corresponds with the five last lines in Goldast's version of 3.1 Dial. 4. The extra material found here in Ly and Gs is found also in Pz, from which Ly took it --- it therefore does not confirm the view that Ly had some other source besides Pz.

In many other places Ly and Lm follow Pz where it seems obvious that Pz was wrong. This confirms the hypothesis that they derive from Pz, since the more implausible the error the less likely it is that the agreement could be coincidence. These errors include: wrong or incomplete references for Ephesians, 1 Timothy, Titus, 1 Peter, misquotation (an extra esse) of Titus; obvious defects in canon law references (misreading of "2", resulting in there being no causa number, and omission of "di"), some defects of sense, e.g. alius, propter, specialem, and the transfer of material already noted (ecclesie and successoribus). There are also many other not-so-obvious errors in which Lm and Ly agree with Pz.

Mz, Fr, and Frm

Mz seems to be the best witness to 3.1 Dial. 1, but it contains many readings that must be rejected. In some places it has a gap, presumably because the scribe could not read his exemplar. In each of these cases the sense seems to require something and we have tried to supply it by conjecture or from some other source, not always succeeding. See: itaque, potestatis, infringere, silam, ieronimi, explicite, formidanda.

Fr often agrees with Mz and not with Pz even when Mz is clearly wrong, which might seem to suggest that Mz and Fr have a common ancestor (let us call it "X") not among the ancestors of Pz (or else that Mz is an ancestor of Fr -- see below). See et, huius, potestati, scit, profecto, quemlibet, summo, plenitudinem potestatis, ponit, enormium. However, these are all obvious errors easily corrected by conjecture or accident or by consulting the text of the canon law. It is therefore possible that they were all in an ancestor common to Mz, Fr and Pz but were corrected in Pz (or some intermediate MS) and not in Mz or Fr. There is, therefore, no strong reason to postulate X.

Is it possible that Mz is an ancestor of Fr, or vice versa? Not vice versa, since Fr is dated late; also the gaps in Mz do not correspond to gaps in Fr, and there are a number of short passages omitted in Fr that are found in Mz (e.g. pagano... cuicumque, in... quid). Mz is probably not the source of Fr, since there is one short passage in Fr that is not in Mz (see preterea), though it could conceivably have been supplied in Fr by conjecture. If we assume that the peculiar errors in Fr are likely to have taken a number of copyings to accumulate, then Mz is probably not early enough to have been an ancestor of Fr. But the possibility cannot be altogether ruled out.

Fr contains quite a few implausible readings not found in any other witness: meo, studebis, divelli, temporalibus, eciam, interpretacionis, and about as many again. On the other hand, Fr twice contains plausible readings not found in other witnesses: proficiendi, minoris. In some places the scribe of Fr (or of one of the ancestors of this MS) corrects mistakes by adding et or vel and the correct word: habetur et dicitur, mulier seu uxor, dei sive eiusdem Christi, etc. In other places he corrects by crossing out, e.g. non obviant.

The marginal and interlinear corrections in Fr may be the work of a "proofreader" checking against the exemplar, or they may be conjectural. Some of them are wrong. None of them suggests that the corrector referred to any MS other than the exemplar of Fr: incipe, comprehendere, est...obediendum, priusquam, papa... humanas, omnes, potestatem, nec...divinam, evangelica, quia, veteris... oneribus, 17, sed... mortalis, non, minime. Mz likewise contains a few corrections, including some chapter numbers, probably also made by checking against the exemplar or by conjecture: omnis...Iordanem, non, cap 3, cap. 4, cap. 5, se, non, cap 12, cap. 14.

Fr and Pz

There are a number of places in which Fr agrees with Pz against Mz. This might suggest that Pz and Fr have a common ancestor not an ancestor of Mz (let us call it "Y"). These places are as follows: de, subditi, habeant, debeant, ibidem, proprietatem... rerum, temporalium rerum, vester servus, potestates, potestatem a Christo. Four of these are transpositions, which easily happen, and the agreement could easily be coincidence. Three others are obvious errors which the scribe who wrote Mz could easily have corrected by conjecture, as the editor of Ly did. In the remaining case the reading potestates in Mz is doubtful and could in any case have been supplied from the canon law. There seems, then, that there is no good reason to postulate Y.

Conclusions

  1. Nothing in the MSS gives reason to postulate any particular pattern among the ancestors of the extant witnesses.
  2. Marginal and interlinear alterations in Mz and Fr do not suggest reference to any MS except their exemplars.
  3. In 3.1 Dial. Pz is probably the sole source of Lm and Ly.
  4. Ly is the sole source of Gs.

The independent witnesses are, then, Mz, Fr and Pz; it is not possible to say whether any two of them had an ancestor that was not an ancestor of the third. Mz is the best witness, though it contains gaps and errors. Ly is worth consulting as a source of intelligent conjectural emendations.

Return to Relations among manuscripts and editions