Contamination

A major challenge to cladistics and, more in general, to all textual critical attempts, is contamination. This consists in cases such as the following: the copyist of manuscript B, while copying it from manuscript A, counter-checked manuscript C. Hence, B is a contaminated manuscript. Contamination may regard specific portion of the text, i.e. one checks manuscript C whenever A is hardly legible, lacks some folios, is not authoritative, and so on. In other cases, contamination may be a sort of collation: the copyist chooses the best reading among the availble ones (dual or plural).

It is easy to see that to reconstruct a stemma when one has many contaminated manuscripts is almost impossible. Moreover, contamination is not always easy to detect. There is also parallelism, that is, «the phenomenon that identical mistakes affect different lines of transmission independently and by chance» (Maas 2009-2010, quoted in yesterday's post).

What to do?

One may adopt the same kind of approach one uses in the problem of authenticity in visual arts. As well known, in order to decide who is the author of a picture, one does not look at its most salient traits, but rather at its marginal ones. At nails and not at eyes in portraits, for instance. In manuscripts, one might look not at the sūtras embedded in a commentary (which might be counter-checked in another text), but rather at variants in less significant sentences and of not so direct significance, but of textual critical relevance. For instance, let me quote a text I am presently critically editing, Rāmānujācārya's Tantrarahasya:

evaṃ ca parasparam ākāṅkṣāvaśāt phalavākyenetareṣām ekavākyatayā yugapad anvaye prasakte tatra prathamam āgneyādivākyeṣu kālasambandhapratīteḥ darśapūrṇamāsaśabdavācyatvasambhavād idam adhikāravākyenaikavākyaṃ bhavati.

Out of my (contaminated) witnesses, two (O and the younger P) read prathamādivākyeṣu instead of prathamam āgneyādivākyeṣu. This is clearly wrong and can, hence, hardly be the result of parallelism. It is, on the other hand, easy to imagine that one copied from the other a syntagm which, at first sight, is not patently wrong. Hence, the copyist did not feel the need to check a different manuscript. Hence, one might infer that –at least as far as this portion of the text– the copyist was basically copying O and counter-checked other witnesses only in more important cases.