In The Practical Talmud Dictionary, by Yitzchak Frank, one can read that there are two kinds of “don’t read” expositions: one “not only in accordance with its Masoretic vocalization, but in accordance with its spelling that allows for a different vocalization as well”; and the other “not only in accordance with the way it is actually spelled, but… also as if two of its letters were transposed.”
Frank’s example of the first kind involves the addition of a Vav to a word, so apparently he refers to full and defective spellings here; and his example of the second kind is straight-forward.
Then he adds a note that in one instance a word is read as if two letters were replaced by two other letters.
I just don’t get what he’s saying.
First, there are lots of changes in the Gemara that are neither of the kinds Frank mentions, that is, neither the addition (or omission) of vowel letters nor the transposition of two letters. There are examples with absolutely no change of letters (such as Ex 32:16 at Avos 06:02; Ex 23:25 at Berachos 48b; Num 21:14 at Kiddushin 30b; Isaiah 02:22 at Berachos 14a; Psalms 46:09 at Berachos 07b; and Proverbs 19:23 at Berachos 14a). There are changes that involve only the addition of a space within a word, as with Gen 01:01 at Succah 49a. There are examples that involve only the change of a single letter (Num 11:05 at Berachos 32a; and Psalms 29:02 at Berachos 30b), or involving more than two letters other than the singleton that Frank mentions in his afterthought, Deut 07:07 at Kiddushin 30a: Deut 08:09 at Ta'anis 04a; 1 Samuel 02:02 at Berachos 10a; and Ecclesiastes 08:10 at Gittin 56b.
But secondly, I just don’t get why this kind of classification is at all helpful. Is the distinction that Frank makes, between vowel-letter changes on the one hand and two-letter transpositions on the other, somehow useful in understanding the Gemara?