Why some people say git and some say gut: The mechanism of Yiddish dialect differentiation

  The sounds of a language can change in different ways. The most commonly described change is a sound shift. A classic example from the history of English is that a long U sound inherited from ancient Germanic became a diphthong the we find in words like 'cow.' Another way that sounds can change is when the words they are found in are borrowed from one language to another. The speakers of the new language adapt the pronunciation of the word to fit in with their own habits of speaking. This second process is what underlies most of the differences that we find in pronunciation between Yiddish dialects.
  To illustrate this we can take an example from the Yiddish dialects historically spoken in Ukraine, Poland, and Belarus and the Baltic countries. The word for 'good' is pronounced either as git or gut. To explain this fact we have to explain two things. One is how the difference came about and the other is why which pronunciation is historically preferred in which areas. In what follows, I will set questions of geographical distribution aside and deal with the origin of the difference. I will also not delve into variations in the pronunciation of I vowel among people who say git as these are not relevant to this particular problem.
  To begin with we are looking at a very big pattern. The alternation of u and i is found in words of Germanic origin like git and gut as well as words of Slavic origin like the word for 'beetle' that can be zhik or zhuk and words of Semitic origin like the word for 'crazy' that can meshige or meshuge. In all cases a speaker who uses a u or i in one of these words uses the same vowel consistently in the other, When linguists encounter a pattern like this, the first explanation they think of is that there was a sound shift. These words originally had one of these sounds, perhaps an I sound, perhaps an U and then some speakers regularly changed their pronunciation from one form to another. At first historical facts seem to support this. All these words once had an U in some forms of Germanic, Slavic and Semitic.
  A closer look at the facts allows us to regard such a sound shift unlikely. A sound shift has a limited duration. During the time it is happening all the words with appropriate sound undergo the shift. Then the shift comes to an end. After the shift has run its course new words borrowed into the language no longer undergo the shift. If they have the sound affected by the shift, say I or U, they keep it. Since Germanic, Slavic and Semitic words were borrowed into Yiddish over a long period extending into recent times we would expect, if there were a shift from U to I, that the shift would have come to an end at some time and words borrowed later that had U would keep it even among speakers who say git.
  We do not find this. People who say git use an I sound consistently in words where other people say gut even in words that came into the Yiddish language in historically recent times. Another explanation is required. This would be that the changes in pronunciation are the result of borrowing. It was borrowing of a specific kind and to see how this happened we have to take a closer look at the structure of Yiddish.
  Max Weinreich described Yiddish as a fusion language. A fusion language consists of components of different etymological origins. Each component borrows material from a specific class of languages called its determinant. Yiddish has a Germanic component with a Germanic determinant, as well as a Slavic and a Semitic component.
  While all elements of Yiddish share common linguistic structures, each component also has its own structural features. Plurals, for example, are generally constructed differently in nouns of the Germanic, Slavic and Semitic components.
  Each component also has its own ways of adapting words from its determinant into the Yiddish language. Typically, for example, borrowed nouns are assigned new plurals. The plural of noun borrowed from Germanic often has a different ending than it had in its Germanic determinant. This is often true in the case of nouns from the Slavic component and, occasionally,  in nouns from the Semitic component as well. This fitting words from determinants into components is the characteristically Yiddish form of linguistic borrowing.
  Ways of adapting words from the determinants to the components can vary somewhat within Yiddish. For example, words from the Germanic components can be assigned different plurals in Lithuania and Belarus than they might be assigned in Poland or Ukraine.
  The same principles apply to pronunciation. The Germanic, Slavic and Semitic components have characteristic ways of adjusting the pronunciations of words from their determinants to fit in with Yiddish patterns. As with the case of plurals these ways of adapting pronunciation can vary somewhat within the language.
  Now we can see how the git and gut pronunciations could have emerged. Languages vary in the kinds of U sounds they can have. Some like French and German have two kinds. One kind is usually spelled 'ou' in French and 'u' in German. In phonetic notation it is written /u/. The other kind is usually 'u' in French 'u' with an umlaut in German. It is written /y/. English has only one kind of u. In standard English it is /u/ but in dialects like the one spoken in Philadelphia it is /y/.
  It is possible to infer from old manuscripts that Yiddish in medieval Germany was one of the languages that distinguish /u/ and /y/. Bu those Yiddish speakers who migrated into eastern Europe lost this distinction. Linguistic distinctions are often lost in migrations.
  Once the distinction was lost, Yiddish speakers had to opt either for /u/ or /y/ as their one preferred pronunciation. Some communities opted for /u/ and others for /y/This had an effect on the way words from the Germanic, Slavic and Semitic determinants were incorporated into the components. Communities that used /u/ assigned /u/ to words with U sounds in the determinants while those with /y/ assigned /y/ to the same words.
  In eastern Europe Jews found themselves in different linguistic environments. Hungarian had both /u/ and /y/ sounds while Polish, Ukrainian and Belarusian had only /u/. Yiddish speakers in the territory of modern Hungary and adjacent Burgenland, Austria and western Slovakia arrived with the  /y/ variant and kept it. In the Slavic lands some settlers founded communities with the /u/ variant. They maintained it in their new homes. Others founded communities with the /y/ variant but these lost it under the influence of surrounding languages. They shifted from /y/ to the sounds closest to it which were I sounds.
  The shift from /y/ to I sounds changed the way words from the determinants were incorporated into the components. While previously these communities had incorporated words with U sounds from the determinants as having /y/ they now incorporated them with I sounds. As a result Yiddish speakers in the Slavic lands either said git or gut. (An exception was the town of Bransk, Poland which still has /y/.)
  This is an illustration of the power of the patterns that govern the way words from the determinants are incorporated into the components. As long as Yiddish remained a fusion language these patterns were strong enough so that words with U were taken into the components as having I vowels. The principle mechanisms by which Yiddish dialects arose were not sound shifts but rather variations in way elements from determinants were adapted to fit into components.

Spectacular Genomic Confirmation of Max Weinreich's Babylonian Renaissance

In his "History of the Yiddish Language" (1973) Max Weinreich regarded the 13th century as a turning point in the history of the Yiddish language. He identifies it as the boundary between the Early Yiddish and Old Yiddish periods. More specifically he writes about a 13th century Babylonian Renaissance characterized by a change in the Ashkenazic norms of Hebrew pronunciation. He says that the change was centered around Rothenburg and involved scholars who bore names that were previously rare or unknown among German Jews but were used by Jews in the Middle East. The name Bablyonian Renaissance comes from Weinreich's beliefs that the pronunciation norms came from Mesopotamia and that the scholars who brought them migrated from there.
A look at studies of the history of Jewish settlement in Germany shows that the 13th century was a period when settlements increased rapidly in number and geographical extant. Judging from the maps the center of this spread was the Main Valley. Rothenburg is in this area.(Maps in Michael Toch's "Jews and Peasants in Medieval Germany" 2003 and Alfred Haverkamp, ed., "Geschichte der Juden in Mittelalter" 2002).
Genomic studies have looked a uniparental markers among the Ashkenazim, These are genetic features that are inherited exclusively from one parent. Mitochondrial genes are transmitted only from mothers and genes on the Y chromosome come only from fathers. According to one study Ashkenazic Jews predominantly show mitochondrial genes that are characteristic of European populations (about 40%) and Y chromosomes genes that are characteristic of Middle Eastern ones. ("A prehistoric European ancestry amongst Ashkenazi lineages" Martin Richards et. al, in "Nature Communications" 4, article number 2543 (2013) )
Another study looked at the whole Ashkenazic genome.("Sequencing an Ashkenazi reference panel supports population targeted personal genomics and illuminates Jewish and European origins" Itsik Pe'er et al. "Nature Communications" 5, article number 4835 (2014) ). This makes it possible to infer some aspects of the history of the population. One thing that stands out is that the population looks like the result of an even admixture of a European and a Middle Eastern population (European ancestry estimated at 46-50%). The date of this admixture was between between 600 and 800 years ago.
The same study inferred a population bottleneck in the history of Ashkenaz. That is a period when the population dropped before expanding. As a result of this bottleneck modern Ashkenazim are descended from only between 250 and 420 ancestors.They lived between 25 and 32 generations ago. This would put the bottleneck at about the same time as the admixture.
If I am interpreting these data and inferences correctly, the majority of German Jews alive before 800 years ago did not contribute ancestors to the modern Ashkeanzi population. The exception was a subset comprised mostly of women who married Middle Eastern immigrants. The descendants of this admixed population expanded rapidly.
One explanation of this rapid expansion relative to the unadmixed population may be that the admixed population was disproportionately involved in the expansion of settlements. Some support for this idea comes from a study of the European settlement of Quebec. Settlers who founded new settlements had larger families with more married children and their descendants were also more apt to found new settlements, ("Deep human genealogies reveal a slective advantage to be on an expanding wave front" Claudia Moreau, et, al."Science" 25 November 2011 v.334 pp.1148-1150)
Another possible factor was suggested to me by Lou Cleveland. The Black Plague greatly reduced the German population in 1349. People of Middle Eastern origin may have had a degree of genetic resistance to this disease.
Update
A recent   article estimates that the admixture event involving a population with Middle Eastern genetic affinities was actually earlier than the 13th century, I'll write more about this in a new post.

The Role of Crimean Gothic in the Formation of the Eastern Yiddish Dialects


As readers of this blog know, my research has led me to some unusual conclusions about the history of Yiddish. Yiddish is most commonly believed to have originated in Germany about 1000 years ago and to have been carried from there to eastern Europe where the majority of its speakers lived. Two things that have that generally been considered relevant to the history of Yiddish are Crimea and the Gothic language. Crimea is a peninsula of Eurasia which is not known to have had a Yiddish-speaking population before the 19th century when Jews began to settle there from other parts of the Russian Empire. Gothic is a Germanic language spoken by a people who migrated from northern Europe to the shores of the Black Sea around 250 CE. It gradually stopped being spoken on the European mainland after about 600 but continued to be used in Crimea at least through the 1500's.

In previous posts I have stated my general conclusions that there was once a Gothic speaking Jewish community in Crimea which played a role in the history of Yiddish. Recently, I have been able to come up with a more detailed historical narrative. In future posts I plan to supply the reasoning behind this reconstruction but here I just want to present it in broad outlines.
Around the 850's there was an active trade route between Austria and east Germany in the west and the Black Sea coast. Crimean Jews played a role in this trade. The Jews involved in this trade spoke a form of Gothic. By making inferences from later Yiddish it is possible to reconstruct some aspects of this language. It incorporated words borrowed from Hebrew and Aramaic. Often these were incorporated into Gothic by attaching Gothic elements to the Hebrew roots.
After about 900, this trade route declined. The Gothic speaking Jewish community now split into two groups which had, for a long time, relatively little communication.
One group continued to live in the Austrian and East German area. They were surrounded by speakers of German, a language that was fairly closely related to Gothic and they gradually shifted from speaking Gothic to speaking a language that was largely German but with an underlying structure that was retained from Gothic. This was an early form of Yiddish. The modern Yiddish dialect which resembles it most closely is the one that Uriel Weinreich named West Transcarpathian Yiddish.
Back in Crimea, the Jewish Gothic spoken there eventually took a new approach to Hebrew loans. Gothic morphemes were no longer attached to Hebrew roots. Hebrew nouns were given Hebrew plural endings and the roots of Hebrew verbs were used along with separate Gothic auxiliary verbs.
Around 1300, trade between the German lands and Crimea revived. Jews from the eastern parts of Germany and Austria re-established communication and some migrated to Crimea. They brought their language, the ancestral form of West Transcarpathian Yiddish  along with a new approach to Jewish culture that had evolved in central Europe. Some Gothic speaking Crimean Jews adopted these cultural features and also learned Yiddish from these immigrants.
In this new environment Yiddish was no longer surrounded by German speakers but by speakers of Gothic. A new dialect, Crimean Yiddish developed under this renewed Gothic influence.
By the mid 1300's,  major trade routes from Crimea led north into the developing Duchy of Lithuania and Polish Kingdom. Yiddish speaking Crimean Jews moved north along these routes, settling along them and bringing their Crimean Yiddish. In the Duchy of Lithuania it developed into the Northeastern Yiddish dialect.
In the Polish Kingdom the development was more complicated. There settlers from Crimea encountered Jewish settlers from eastern Germany and Austria who spoke the ancestral form of West Transcarpathian Yiddish.. Contact between this dialect and Crimean Yiddish led to the development of the Central and Southeastern Yiddish dialects.

In summary the eastern dialects of Yiddish formed as the result of three language contact events. The first involving a form of Jewish Crimean and German occurred along the the trade routes between the German speaking lands and the Black Sea around 850. This produced a dialect ancestral to modern West Transcarpathian Yiddish. Subsequently, the central European and Crimean branches of this trading community were separated The second event was renewed contact between speakers of this Yiddish dialect and speakers of Jewish Crimean Gothic which occurred around 1300. It produced the ancestor of modern Northeastern Yiddish. The third event occurred in the 1300's in the lands that became the Kingdom of Poland. There the contact was between the Crimean Yiddish ancestral to Northeastern Yiddish and the dialect ancestral to West Transcarpathian Yiddish. The products of this was the ancestor of the modern Central and Southeastern Yiddish dialects.

A sketch of the history of the Yiddish dialects

I want to give a quick overview of how I see the Yiddish dialects emerging from a Gothic system. Early Yiddish developed out of Gothic in parts of Ukraine. Until about 1200, Early Yiddish remained in eastern Europe. The spread of Yiddish into Central Europe began in the 1200's reaching Austria first. There speakers of Bavarian German learned Yiddish. The Yiddish they learned was essentially a Gothic system but in the course of acquiring it they incorporated many Bavarian features. For example, the vowels in the words inherited from Gothic were redistributed to approximate the Bavarian vowel distribution. The resulting Bavarianized Yiddish constituted the ancestral form of modern West Transcarpathian Yiddish.
This Bavarianized Yiddish spread into Germany where it was learned by speakers of Central German dialects who modified it by incorporating  material from Central German. The German Jews of the Rhineland also incorporated their unique and very old Hebrew vocabulary.  This was the origin of the family of West Yiddish dialects that came to be spoken in Germany, Alsace-Lorraine, Luxemburg, Holland and Switzerland.
In all these dialects the incorporated German material was modified owing to its integration into the basically Gothic system of Early Yiddish. The nature of these transformations is a rich source of information about the structure of Early Yiddish. This particularly valuable as we have no texts in Early Yiddish.
A migration of West Yiddish speakers into eastern Europe began in about 1400. There these West Yiddish speakers came into contact with resident speakers of Early Yiddish. The newcomers acquired  the Early Yiddish system while adding West Yiddish elements. The new forms of Yiddish that developed out of this process were the ancestral East Yiddish dialects.
This process had interesting parallels to the earlier development of West Yiddish. In that case a vowel  distribution approximating that of a German dialect replaced the Gothic vowel distribution of  Early Yiddish. In this case, the Gothic vowel distribution was replaced by one from West Yiddish.
Just as the transformations undergone by German elements integrated into West Yiddish testify to the structure of Early Yiddish, the incorporation of West Yiddish elements into East Yiddish gives
additional testimony about Early Yiddish structure. In both cases the transformations reflect structural processes that Early Yiddish inherited from Gothic.

Yiddish from 400 to 1400 C E

Yiddish is well represented by written texts starting at about 1400. There are some earlier traces of written Yiddish that go back about 1000. It would seem that the trail goes cold at that time. However, if we accept the evidence presented on this blog that Yiddish is descended from Gothic we can venture further into the past, as far back as somewhat before 400 when Gothic is documented by Wulfila's translation.
Around 400 Yiddish would not have been very different from Gothic as represented by Wulfila. By 1400, Yiddish shows very strong influences of German. What can be said about the intervening period?
Let's look first at geography. Around 400 Yiddish speaking territory would have coincided with at least a part of Gothic territory. This territory embraced the northern shore of the Black Sea, including Crimea, and extended into the Balkans. By the 800's Ashkenazic Jews, presumably Yiddish speakers were living in the Carolingian Empire, in Austrian and Bavarian territory, notably in Regensburg. If this community was an extension of the Gothic Jewish community it mostly likely spread there from the Balkans.
In the course of the following two centuries, Jewish communities spread into East Central German towns like Magdeburg, Halle, and Erfurt and in the central Rhineland around Speyer, Worms and Mainz.
Next let's look at social conditions. During the late Roman and early Medieval period up to about 800, Jews were generally well integrated into European society and Judaism even attracted many converts. Later conditions deteriorated, particularly after the late 11th century.
These changes very likely had linguistic consequences. During the early period up to 800, Yiddish may have served the role of a missionary language designed to recruit converts. Traces of this early function still survive in Yiddish as I plsn to demonstrate in a later post. When Jews came in contact with German-speakers, chiefly after 800, Yiddish which was then based on Gothic would have been close enough to early German for Jews to be able to communicate with their neighbors by making small modifications in their speech. This would have chiefly meant that Geman sounds would be substituted for Gothic based sounds. Among themselves. Jews would have continued to use the Gothic based pronunciation.
This state of affairs where a community uses one system internally and communicates with neighboring communities through sound substitutions is called a diasystem. It is likely that this diasystem persisted until about 1100. Around this time the social status of Jews began to fall and the German sounds came to be used all the time, even in inter-Jewish communication. This collapse of the old diasystem left a very strong mark on later Yiddish, something that I plan to make the subject of future posts.

New Directions

I last posted on this blog in October, 2009. Since then a number of irresistable new opportunities have appeared which have competed for my time. For one thing, thanks to the initiative of Bob Scott at the Digital Humanities Center of the Columbia University Libraries there is a real possibility that a critical component of the Language and Culture Atlas of Ashkenazic Jewry data base will be published online. I have been part of the effort to accomplish this.
A new research opportunity has also appeared with the development of inexpensive and widely available genomic testing since 2008. The potential for using genomic information as a source for Ashkenazic history has long been recognized but pioneers have had, until recently, to base their research on the limited data provided first by classical markers and later by mitochondrial DNA and Y chromosome haplogroups. Extensive newly available autosomal data is now being added to provide a much richer and firmer basis for historical inferences.
My research on the history of Yiddish has also taken on some new directions thanks to hints provided by teachers, Marvin Herzog and Mordtkhe Schaechter.
Herzog called attention to the fact that when groups migrate they tend to lose cultural and linguistic traits. This observation which goes back his mentor Uriel Weinreich is now becoming more widely appreciated as can be seen from a recent article by Quentin D. Atkinson in 'Science' which uses lose of phonetic features to trace the spread of human languages back to Africa.
Uriel discussed the loss of features in small settlements in his article on 'Western Traits of Transcarpathian Yiddish.' The specific example he gave was the development of Northeastern YIddish but in other parts of the same article he describes a much wider pattern of progressive trait loss. He describes a number o features unique to Transcarpathian Yiddish. A smaller set of these can be found in Central Yiddish. A still smaller set are shared by Central and Southeastern Yiddish but not found in the remaining the East Yiddish dialect, Northeastern Yiddish.
The global pattern is then one of maximum richness of traits in Transcarpathian Yiddish with a gradual falling off of traits towards the northeastern part of East Yiddish territory. Uriel does not specifically say as much but this pattern implies that the settlement of East Yiddish territory spread out from the broader Transcarpathian area (this area would include Austria and the Czech lands which are to the west of the Carpathians) to the northeast. We can infer from this that the common root of Transcarpathian and East Yiddish originated in the Transcarpathian area. This is consistent with other evidence that the ancestor of these dialects coalesced in Austria around the the 11th century.
The spread of East and Transcarpathian Yiddish from a focal area in early medieval Austria is also implied by a line of research that was suggested by the work of another of my teachers, Mordtkhe Schaechter. Schaechter identified a specific kind of linguistic borrowing in which a language borrows standards of correctness rather than specific items from another language. The borrowed standard is used to select preferred variants within the internal repertoire of the borrowing language.
As I have argued elsewhere in this blog, early Yiddish was an East Germanic language derived from Gothic. Modern Yiddish is, in comparison, much more similar to German. Yiddish has come to resemble German partly through the mechanism of ordinary linguistic borrowing but the mechanism described by Schaechter has actually played a greater role. Specifically, early Yiddish borrowed standards from a form of literary German. This form was the Middle High German used in Austria.
As to the question of how the original Gothic-derived Yiddish got to Austria, I currently favor the hypothesis that from at least 500 CE on, Gothic was spoken by the Jews of northwestern Balkan Jewry. Medieval Austria was settled by Jews from the Rhineland and East Franconia as well as other areas but I think that early Yiddish was brought there by settlers from the area of the Save River valley.