compare to Huet, present participle parasmaipada

[funderburkjim]

This issue summarizes differences and similarities noticed in the comparison of present participle stems and declensions; the methodology is described in the readme for huetcompare/parts-prap.

[funderburkjim]

stem comparison

huet_stems_prap.txt summarizes the stems that appear in present participle (parasmaipada) forms in the SL_parts.xml file from Huet. For each stem is shown all class-pada (pada always P) combinations which have declined forms. Two noteworthy details are

• the same stem value is used, regardless of the gender of the declined form. Specifically, there is no specific mention of a feminine stem.
• the underlying root of any participle is NOT specified in the SL_parts file. Some computation (probably involving the roots appearing with present tense conjugational forms in SL_roots.xml) could backtrack to find the root from which a given participle stem was derived; but it would be useful to have this root identified within the records of the SL_parts file.

pysan_stems_prap.txt summarizes the stems and classes padas that appear in the present participle stems of
MW-verb-prap.txt of the pysan system.

The two lists of roots and padas are compared in
compare_stems_prap.txt.

This comparison is organized into categories:

• 432 cases of stems which appear in both sources, and with the same class-pada data
• 17 cases of stems that appear in both sources, but with Pysan showing conjugational classes in addition to those in Huet.
• 10 cases of stems that appear in both sources, but with Huet showing conjugational classes in addition to those in Pysan
• 42 cases of stems where the Pysan and Huet classes are disjoint.
• 1108 cases of stems which appear only in pysan.
• 204 cases of stems which appear only in the Huet source (Many of these have
  conjugational class of 11, which in the present tense conjugations,
  are identified as from denominative roots.)

[funderburkjim]

declension table comparison

huet_decl_tables_prap.txt contains 2157 present participle parasmaipada declension tables derived from Huet's SL_parts.xml.

Each of these appears on one line of the file.

For each of these Huet computations, a comparable computation from the pysan
system was prepared, taking into account the class-pada information used by pysan present participle parasmaipada declension computations. The result is
pysan_decl_tables_prap.txt.

compare_decl_tables_prap.txt summarizes the similarities and differences in the corresponding declensions from the two files just mentioned.
Some statistics resulting from the comparison:

• 918 (43% of the 2157) declensions are identical in the two systems.
• 829 (38%) of the Huet declensions have no comparable pysan declension. This large percentage has two major contributors:
  • The presence of denominatives in Huet, which are not included in the Pysan work.
  • Differences in the root conjugational classes of roots in the two systems.
• 410 (19%) of the remaining declensions differ in one or more details.

  • Most of these (396) are due to a systematic difference in the neuter
    declension. Here is a typical example:

    akzRuvat ppr 5Pn: declension differences @ 1d,2d 
    1d (huet) akzRuvantI,akzRuvatI != akzRuvatI (pysan)
    2d (huet) akzRuvantI,akzRuvatI != akzRuvatI (pysan)
    

As with the comparison of declensions of nouns, Huet does not include vocative inflected forms in the SL_parts.xml data. This systematic difference is ignored in the comparison.