Kaufman, Stephen A, “Paragogic nun in Biblical Hebrew: Hypercorrection as a clue to a Lost Scribal Practice,” in Solving Riddles and Untying Knots: Biblical, Epigraphic, and Semitic Studies in Honor of Jonas C. Greenfield. ed. Ziony Zevit, Seymour Gitin, and Michael Sokoloff (Winona Lake: Eisenbrauns, 1995), 95-99.
In Biblical Hebrew prose the imperfect forms that end with a long vowel (2fs, 2mp, and 3mp) sometimes occur with an extra nun on the end – the so-called paragogic nun. However, an explanation of why these forms appear has eluded Hebrew grammarians. The forms appear most frequently in older texts where they occur most frequently in pause. The appearance of a final nun also corresponds to the Aramaic/Arabic forms of the imperfect indicative יִכְתְּבוּן (yiktebūn) which contrast with the shorter jussive/preterite form יִכְתְּבוּ (yiktebū).
In Semitic languages it is common for a nun to be assimilated to the following consonant in the prosody of speech. Thus in the majority of cases, except where the verb occurred in a clause final position (as in pause) or before a consonant which cannot be doubled, the final nun would be lost to assimilation. Over time the imperfect form was reanalyzed as יִכְתְּבוּ and therefore, as occurred in later colloquial Arabic, the two forms would have fallen together in normal Biblical Hebrew prose so that the contrast between יִכְתְּבוּן and יִכְתְּבוּ no longer marked the difference between the imperfect (also called the long form) and the jussive/preterite (the short form).
This explains why a majority of the forms with paragogic nun are preserved in pause, however this is not the case with all of the forms. Further, in a few cases the paragogic nun occurs on a form other than an imperfect indicative such as the imperfect consecutive (which should be a preservation of the short preterite form) and even the perfect. Dr Kaufman suggests that the variation can be explained by hypercorrection and is evidence of a scribal tradition rather than a living linguistic phenomenon.
Hypercorrection is often the result of tension between a higher formal dialect and a lower colloquial dialect where a speaker applies a feature in the higher dialect by analogy to a situation where it should not occur, betraying the author’s lack of experience in the higher dialect. For instance, in English we have lost the use of a “case system” except for some personal pronouns. The 1cs nominative pronoun is “I” while the oblique case is “me.” Children often mistakenly use “me” as a nominative, “Me and Jack are going to the store.” Adults, however, weary of being corrected as children for using phrases such as “Me and Jack”, often misunderstand the rule as applying to compounds and tend to hypercorrect the pronouns in oblique cases where “me” is indeed the proper use, “Bob came with Jack and I.”
In Biblical Hebrew the situation arose where the higher formal dialect pronounced the final nun on 2fs, 2mp, and 3mp imperfect indicatives in situations where the nun could not assimilate such as contextual positions (such as pause). However, over time the lower dialect no longer pronounced the final nun at all. Scribes copying older texts in whose literary dialect final nuns were still included in the orthography would have to learn a set of rules for their use and non-use. By examining the scribal errors we can deduce the rules.
Dr Kaufman notes first that all of the “errors” are found in the books of Deuteronomy and Judges, perhaps reflecting a shared scribal history. In Deuteronomy 1:22, 4:11, 5:23, and Judges 8:1 and 11:18 the paragogic nun occurs on an imperfect consecutive. In Deuteronomy 8:3 and 8:16 the nun occurs on a perfect form (though both cases are the פ’’י verb ידע whose consonantal form may have been misinterpreted as an imperfect). What is significant is that in five of these seven cases the paragogic nun is followed by a word beginning with aleph. Thus it seems that the scribe was concentrating so hard to remember to use a paragogic nun when it cannot be assimilated to the following word that he forgot that it should only be applied to imperfect indicatives.