transcription estimate and best practices inquiry

[Rankin, Robert L]

> Fourth, words in isolation can sometimes be even worse.  The problems of list intonation and the influence of near-homonyms recently discussed arise.  Since I went through the Dorsey Kaw dictionary in alphabetical order, inflection of one verb often contaminated discussion of the next verb in line.  

Let me give an example of what I mean here.  In Kansa there are two different conjugations that begin with /ba-/ followed by the (often identical) verb root.  These correspond to Omaha verbs in /ba-/ 'by pushing' and /maa-/ 'by cutting with a blade'.  Their conjugation is completely different.  The 1st and 2nd person agent prefixes are inserted in different places and have totally different forms (allomorphs).  In Kaw one conjugation is 1sg ba-a-, 2sg ba-ya.  The other conjugation is 1sg p-pa-, 2sg s^-pa-.  The only difference in the 3rd person forms is accent and maybe vowel length.  Unfortunately these near homophones come right before/after one another in the dictionary.  I didn't have the sense to randomize them in my elicitation, and I managed to confuse poor Mrs. Rowe time after time as to which verb in the pairs we were discussing.  Thus, a lot of erronious conjugated forms were recorded.  Luckily most of these can be sorted out in retrospect, but a little more attention to detail on my part could have prevented the problems.

Bob