Two Transcripts, Three Divergences, One Editorial Principle
A candidate is collating two transcripts of one Maharashtri verse for a school edition. The transcripts diverge at three points: (a) Ms A reads puttassa, Ms B reads puttaho, both genitive singular of putta-; (b) Ms A reads bhavati, Ms B reads bhavadi; (c) Ms A reads kadua, Ms B reads kaduam, both meant as the neuter nominative singular of kadu- 'bad'. The style sheet says: follow the norm of the text's own tradition rather than imposing a single pan-Prakrit standard, record the rejected variant in the apparatus, and argue editorial choices in the introduction. The text is Jain Maharashtri copied in the sixteenth century. There is no third manuscript. Two of the three divergences can be settled by grammar; one cannot.
1. Which of the three divergences cannot be settled by grammar alone, and what must decide it instead?
Read the worked answer
That is (a), puttassa versus puttaho. Both -ssa and -ho are recorded genitive singular endings in Prakrit, neither is a Sanskritism, and neither creates an impossible form. Grammar therefore underdetermines the choice, and a candidate who insists that one is 'the' Maharashtri ending is over-generalising from a single grammarian's paradigm.
What must decide it is the usage of the text's own tradition and, where available, the manuscript stemma. Because there is no third witness, the candidate cannot weigh the readings by external evidence. The honest procedure is:
- survey the rest of the transcript to see which ending predominates in this text's own usage;
- print the reading consistent with that usage;
- record the other reading in the apparatus;
- state the criterion explicitly in the introduction, so that a reader knows the choice was editorial, not evidential.
This is exactly the case where the style sheet's instruction replaces a rule of grammar.
2. For the two divergences that grammar can settle, state the reasoning and the reading you would print.
Read the worked answer
(b) bhavati / bhavadi. Maharashtri regularly voices intervocalic -t- to -d-, so inherited bhavati appears as bhavadi. The form bhavati is the unmetathesised Sanskrit shape and is exactly what a scribe trained in Sanskrit would produce almost without thinking; it is a Sanskritising intrusion, not a Maharashtri reading. Print bhavadi.
(c) kadua / kaduam. The word is a neuter a-stem adjective. The Prakrit neuter nominative singular ends in -ṁ, normally transcribed as -ṁ or -ṃ. kadua without the nasal is either a script slip or a sandhi spelling before a following vowel. Standing alone it is not the normal neuter nominative. Print kaduam, and add a note that if the next word begins with a vowel the nasal may be realised as anusvāra or assimilated in pronunciation.
Note the direction of the argument: each verdict is read off a regular sound law or an inflectional ending, not off frequency, and each is stated in the introduction as a reason rather than asserted in the apparatus.
3. Draft the apparatus entries you would actually print, in the style sheet's required form.
Read the worked answer
Apparatus entries are mechanical: rejected reading first, witness siglum, accepted reading, short diagnosis where useful. Keep argument out of them.
- (a) puttaho A : puttassa B - print the reading supported by the text's own usage; add 'var. lect., cf. introduction'.
- (b) bhavati A : bhavadi B - print bhavadi; entry: 'bhavati A, Sanskritising'.
- (c) kadua A : kaduam B - print kaduam; entry: 'kadua A, scribal'.
Two disciplines follow from this. First, the apparatus records what a reader needs in order to reconstruct the witnesses; it does not relitigate the choice. Second, because (a) was settled editorially rather than grammatically, the introduction must carry the argument: which reading dominates the manuscript, what the criterion was, and what a future editor with a third witness should check. Entries (b) and (c) need only a one-word diagnosis, since the grammar does the arguing.