Yiddish final devoicing

The history of Yiddish devoicing is one example of such evidence. In the oldest forms of German, represented by Old High German, there was no restriction against word-final voiced consonants, so Old High German had words like tag ‘day’ ~ taga ‘days,’ gab ‘he gave’ ~ gābumes ‘we gave,’ sneid ‘he cut’ ~ snīdan ‘to cut,’ hand ‘hand,’ land ‘land.’ Between 900 and 1200 in the Middle High German period, a rule of devoicing was added, which resulted in tac ‘day’ ~ tage ‘days,’ gap ‘he gave’ ~ gāben ‘we gave,’ sneit ‘he cut’ ~ snīden ‘to cut,’ hant ‘hand’ ~ hende ‘hands,’ wec ‘road’ ~ weges ‘roads.’

Around this time, Yiddish began to develop as a language separate from German, and would have shared this devoicing rule. Devoicing of final consonants in Yiddish is attested in manuscripts from the thirteenth century where the word for ‘day’ is written , using the letter kuf [k] and not gimel [g]. In some dialects, such as Central and Western Yiddish, this devoicing persists up to today, where you find tak ‘day’ ~ tag-n ‘days,’ lant ‘land’ ~ lend-ər ‘lands,’ with the stem-final voiced consonants of /tag/ and /land/ undergoing final devoicing in the singular. In some dialects such as the Northeastern dialect of Yiddish, the devoicing rule was lost from the grammar, so that dialect has tog ‘day’ ~ tog-n ‘days,’ where the originally voiced consonant reappears as voiced. This process where an earlier sound change is dropped from the grammar is known as reversal of sound change: consonants revert to their original state found before the sound change applied.

There are mysterious exceptions to restoration of original voiced consonants. One case is the word gelt ‘money,’ which derives historically from geld with a voiced consonant. The reason for the different treatments of gelt and tag, words which both ended with voiced consonants at earlier stages of the language, is the difference in the presence or absence of phonological alternations within the paradigm of a word. In the case of tag, the plural form had a suffix -n, and so while the singular was subject to devoicing, the plural was not: this word had the paradigmatic alternations [tak] ~ [tagn]. On the basis of these alternations, a child learning the language would have no problem discovering that the underlying form of the stem is /tag/. It is expected that once the final devoicing rule is lost, the underlying form /tag/ resurfaces since there is no longer a devoicing rule.

In the word gelt, the situation was different. There was no inflectional ending which followed this particular noun. At the earliest stages of the language, a child learning the language only encounters [geld], and there would be no basis for assuming that the underlying form is anything other than /geld/. When the devoicing rule was added to the grammar, the pronunciation of the word changed to [gelt]. Since this particular consonant was always word-final, the devoicing rule would have always applied to it, so the stem only had the phonetic form [gelt]. Although either /geld/ or /gelt/ as underlying form would yield the surface form [gelt], there is no reason to assume that the surface and underlying forms are different. A priori criteria may support one decision or the other, but what we need to know is, what independent test tells us that our reasoning is correct? The loss of the devoicing rule provides exactly the needed empirical test: it allows us to know what underlying form Yiddish-learning children must have assumed at this earlier stage. Knowing the actual underlying form provides an important insight into the learning strategies that children make during language acquisition.

When the devoicing rule was added, there were no alternations in gelt so a child would have no reason to assume that the underlying form of the word is anything other than /gelt/. The child never hears geld, and has no reason to think that the underlying form is different from /gelt/. At an even later stage, the rule of final devoicing is dropped from the grammar of certain dialects. This allows the underlying and historically original voiced consonant of tag to be pronounced again, since it is no longer subject to devoicing and thanks to the paradigmatic k ~ g alternation the underlying form was established as being /tag/. This rule loss has no effect on gelt, since despite being derived historically from a voiced consonant, the final consonant of the stem had been reanalyzed as /t/ – a reanalysis predicted by the presumption that an underlying form is different from the surface form only if there is good reason for assuming so. Because there are no alternations for this word, there was no reason to assume an abstract underlying form.

Another important kind of exception to the reversal of devoicing is seen in the adverb avek ‘away.’ This word was originally aveg, with a voiced consonant. This adverb also had no inflected relatives which allowed the underlying voicing of the final consonant to be unambiguously determined, so once the devoicing rule was added to the grammar, it was impossible to determine whether the underlying form was /avek/ or /aveg/. Again, starting from the assumption that underlying forms do not deviate from surface forms without reason, there is no reason to assume that phonetic [avek] derives from anything other than /avek/, since the word is actually pronounced [avek]. The fact that the underlying form is directly revealed as avek in the dialects which dropped devoicing supports this decision.

The example also reveals something interesting about what might (but does not) constitute a “reason” for abstractness. The adverb avek is historically related to the noun veg ‘way.’ The voicing of the last consonant in the noun stem can be recovered within the paradigm given the earlier alternations vek ‘way’ ~ vegn ‘ways,’ because the singular and plural forms of the noun are clearly related to each other. The evidence from the plural noun had no impact on the child’s selection of the underlying form for the adverb, since there is no synchronic connection between the adverb and the noun – no process derives nouns and adverbs from a unified source, so nothing connects the words for ‘way’ and ‘away.’ The divergence of veg and avek in Yiddish points out that you cannot freely assume that any two phonetically and semantically similar words are actually derived from a single underlying form.