2. American Hasidic Yiddish: Background and Corpora

Yiddish is currently maintained as a spoken community language only in some Hasidic groups, mostly in the US, Israel, the UK, and Belgium (Assouline Reference Assouline2018:472–473). In these close-knit communities, invariably united around a dynastic spiritual leader and preserving as far as possible the way of life of an idealized Eastern European past, Yiddish is a highly prestigious language, functioning as a powerful symbol of a distinct ethnic and religious identity (Isaacs Reference Bartal, Polonsky, Bartal and Polonsky1999). As a result, many Hasidim (plural of Hasid) continue to speak Yiddish and pass it on to their children, while also using the majority language (usually English, Hebrew or Flemish).

The exact number of Hasidic Yiddish speakers is hard to gauge. The estimated number of Hasidim worldwide in 2016 was somewhere between 700,000 and 750,000 (Wodziński Reference Assouline2018a:191).Footnote ⁴ However, not all Hasidim speak Yiddish. Some Hasidic groups have shifted to the majority language (for example, the biggest Hasidic group in Israel, Ger, has largely shifted to Israeli Hebrew), while in other groups Yiddish is maintained as a heritage language with limited use (for example, as a language used by men when studying sacred Jewish texts). The largest Yiddish-speaking Hasidic population is in the United States, where Yiddish is spoken in Hasidic sects such as Satmar, Bobov, Skver, Vizhnits, Belz, Munkatsh, and Pupa, located mainly in the Brooklyn neighborhoods of Williamsburg and Borough Park, in Kiryas Joel (Orange County, NY), and in Monsey and New Square (Rockland County, NY).Footnote ⁵ The most significant Yiddish-speaking group today is Satmar, the largest Hasidic sect (20% of the world’s Hasidim as of 2016; Wodziński Reference Assouline2018a:199), where Yiddish is maintained as a primary spoken language.

The American Hasidic sects were founded or rebuilt in the United States after World War II. Most Hasidic sects in the United States are “Hungarian” or “Galician” (Wodziński Reference Assouline2018a:209), representing dynasties of so-called “Habsburg Hasidim” (Biale et al. Reference Assouline2018:359–400). These reestablished postwar sects attracted not only their own original followers, but also other Jews, born in Europe or the United States, some of them with no Hasidic background (Fader Reference Fader2009:8–9, 223). Some possible linguistic consequences of the American Hasidic revival are discussed in section 4.

The following corpora were consulted in order to examine both spoken and written language.

1. (i) Spoken American Hasidic Yiddish: recorded between 2000–2011 from both men and women. Men: 12 radio interviews (from the Yiddish Kol Mevaser news hotline; Assouline & Dori-Hacohen Reference Assouline2017), featuring 12 male interviewees and 3 male interviewers (81,879 words). Women: recorded lectures and lessons by 12 women (77,546 words).Footnote ⁶

2. (ii) Written American Hasidic Yiddish: A corpus composed of two American Hasidic Internet forums, a total of approximately 85 million words: ivelt active since 2007 and kaveshtiebel active since 2012. These forums are designated male-only, yet participants contribute anonymously, and it is possible that some women participate under the guise of men. The Yiddish in both forums is heavily influenced by English and it seems that most participants are American.

3. (iii) Written Yiddish: the online database Yiddish Book Center’s Full-Text Search (10,055 Yiddish books as of December 2020; mostly prewar East-European Yiddish); and the online database of Otzar HaHochma (110,564 Judaic books as of December 2020, including about 6,000 Yiddish books and about 500 Judeo-German books). There is almost no overlap between the Yiddish texts in these two databases.

4. (iv) Spoken Israeli Hasidic Yiddish (recorded between 1995–2007): male and female speakers, 250 recorded hours of lessons, lectures, and sermons (Assouline Reference Assouline2017:27).

In the next section, I discuss the distribution of wh-imer constructions in the American corpora (i and ii above). I demonstrate that they occur in concessive conditionals, nonspecific free relatives, and as indefinite pronouns, and provide examples of each use.

3. Wh-Imer Constructions in the American Corpora

Wh-imer constructions were documented in the American corpora in concessive conditionals and in nonspecific free relatives. They also function as indefinite pronouns. Wh-imer patterns in the American corpora (in concessive conditionals, nonspecific free relatives, and as indefinites) were documented mainly among male speakers, and probably male writers too (since the analyzed Hasidic forums are supposedly confined to men). Wh-imer constructions may thus be identified with more elevated and formal Yiddish styles of public speech and writing. It is also possible that these constructions are more commonly used by men, yet the limited number of female speakers in the analyzed corpora prevents one from drawing such a conclusion. Only a single occurrence of wh-imer in the speech of a female speaker was documented in the spoken corpus.Footnote ⁷ The precise status and use patterns of wh-imer in the speech community can only be established by means of a more exhaustive study among American Hasidic speakers. At present it is not clear whether wh-imer patterns are also common in everyday spoken Hasidic Yiddish.

Furthermore, the word imer itself is not used in spoken Hasidic Yiddish in the sense of the German immer ‘always’ (rather, štendik and ale mol ‘always’ are used in Israeli Yiddish, and aybik/aybig ‘always’ in American Yiddish), so its uses in spoken American Hasidic Yiddish are restricted to the wh-imer constructions.Footnote ⁸ Note, however, that imer ‘always’ is commonly used in American Hasidic Yiddish texts, as are many other Germanisms (see section 4 below).

3.1. Concessive Conditionals

Three different patterns of concessive conditionals are attested in the written American corpus: a pattern with nor ‘only’, a wh-imer pattern, and a pattern marked by negation. These three constructions can also be combined (see the examples in 6 below). In the spoken American corpus, only the first two patterns (the only-pattern and the wh-imer) are attested.

Consider the following examples of the three patterns. First, there is a construction with the restrictive focus particle nor ‘only’ (documented in East-European Yiddish; Haspelmath & König Reference Haspelmath, König and van der Auwera1998:613), with nor usually following the wh-word or the verb. This is the most common pattern in the written corpus, appearing in about 60% of all concessive conditionals:Footnote ⁹

1. (3)

   

Second, there is a construction with wh-imer (vos imer ‘whatever’, ver imer ‘whoever’, vem imer ‘who(m)ever’, ven imer ‘whenever’, vu imer ‘wherever’, vi/vi[a]zoy imer ‘however’). This construction is used in about 30% of all concessive conditionals in the written corpus:

1. (4)

   

The third pattern is marked by negation (subjunctive + negation + infinitive), as shown in 5. This pattern is relatively less common in the written corpus and was not attested in the spoken corpus.

1. (5)

   

The three patterns can also be used together, as in 6.

1. (6)

   

The estimated proportion of the different patterns in the written American corpus (the ivelt and the kaveshtiebel databases together comprising circa 85 million words; see ii in section 2 above) is presented in table 1.Footnote ¹⁰

Table 1.

Frequency of concessive conditional patterns in the written American corpus (CI 95%).Footnote ¹¹

As can be seen from table 1, the pattern with nor is the most common in the American Hasidic written corpus, and the pattern wh-imer the second most common.Footnote ¹² The pattern with negation is the least common, with its proportion in the written corpus (including combinations with other patterns) about 7% of all concessive conditionals.

3.2. Nonspecific Free Relatives and Indefinite Pronouns

In both spoken and written American corpora, wh-imer constructions are commonly used in nonspecific free relatives, or “fused relatives” (Huddleston et al. Reference Huddleston, Pullum, Peterson, Huddleston and Geoffrey2002:1068), such as zey tuen vos imer zey viln ‘they do whatever they want’. Besides, wh-imer constructions also serve indepen-dently (as in vos imer ‘whatever’), alongside several other series of free-choice indefinite pronouns and parallel nonspecific free relatives (Haspelmath Reference Haspelmath1997:52, 55). Consider the following examples of indefinites and lexicalized nonspecific free relatives expressing free choice in the American corpora, all with a meaning similar to ‘anyone, whoever’, ‘anything, whatever’, ‘anywhere, wherever’, etc.

1. (7)

   

The free-choice indefinite pronouns in 7a are the most common type in the written corpus followed by the ones in 7b. The two lexicalized nonspecific free relatives in 7c,d are rarely used, probably due to their length and complexity, which restrict their use to specific syntactic environments. Thus, for example, while say ver ‘whoever’ was documented 775 times in the written corpus and ver imer (as the indefinite ‘whoever’) 105 times, the longer patterns were less frequent, with 9 occurrences of ver es zol nor zayn ‘whoever’ and 3 occurrences of ver es zol ništ zayn ‘whoever’. Note that these constructions can also be combined, thus emphasizing the sense of free choice that they imply (for example, say ven imer ‘whenever’, vu imer es zol nor ništ zayn ‘wherever’, say vos es zol nor zayn ‘whatever’). Besides, free-choice indefinites are expressed in both American corpora in additional ways, as, for example, with the borrowed eni ‘any’ + N: eni zax ‘any thing’.

The distribution patterns of the different free-choice indefinite pronouns in the American corpora are complex and are not discussed here. For the present purpose, it is sufficient to note that wh-imer free relatives and indefinites are commonly used in the American corpora.

4. The Origins of Wh-Imer in American Hasidic Yiddish

Having established that wh-imer is commonly used in the corpora of men’s written Yiddish and public speech, the next step is to consider the origins of wh-imer in American Hasidic Yiddish. It is possible that contact with English is affecting this Yiddish construction, and it appears that Yiddish speakers identify wh-imer with the parallel English wh-ever.Footnote ¹³ While English seems to affect contemporary patterns of use, however, it is possible that English was not the only source of this construction in American Hasidic Yiddish. I suggest that these constructions were not necessarily introduced to American Hasidic Yiddish as calques of the English wh-ever, and that they may have been accessible to speakers directly through Germanized Yiddish dialects, German-like formal Yiddish styles and contact with Judeo-German varieties (see section 4.1 below). In other words, I suggest that these Yiddish constructions may not have been innovations based on English patterns, but rather patterns already familiar to some Yiddish speakers due to the direct and indirect impact of German varieties. Their dissemination in American Yiddish, thanks to the impact of English, can be described as a gradual shift from a minor use pattern (for example, from a construction found only in certain German-like elevated Yiddish written texts) to a major use pattern, found in various contexts of use (Heine & Kuteva Reference Heine and Kuteva2005:44–62).

In order to examine this hypothesis, I first turn to the dialectal makeup of American Hasidic Yiddish and the possible impact of German and Germanized Yiddish varieties during its formation. Note that written American Hasidic Yiddish (mainly Satmar Yiddish) manifests German-like orthography and abounds in German loanwords (Krogh Reference Aptroot2014). This is also evident in the written corpus, which manifests a tendency toward Germanized orthography and contains hundreds of Germanisms such as alzo ‘so’, manxe ‘some’, umfarmaydbar ‘inevitable’, etc. Both the Germanized orthography and the use of German loanwords follow certain conventions and preferences of the so-called daytshmerish style of early 20th-century East-European Yiddish (Krogh Reference Assouline2018).Footnote ¹⁴ The present discussion suggests that German-like prestigious Yiddish written styles played a role in the rise of the wh-imer constructions and speculates that contact with spoken German and spoken Germanized Yiddish varieties might have also played a part in this process.

4.1. Multilingualism and Formation of American Hasidic Communities

Contemporary American Hasidic communities were founded in the United States after World War II. Several small Hasidic communities did exist before the war, but most Hasidic immigrants were Americanized and did not succeed in passing the Hasidic way of life to their children (Glazer Reference Glazer1957:143, Poll Reference Poll1962:19–20, 26–27). Substantial, close-knit Hasidic communities were established only after the war with the arrival of prominent Hasidic leaders and their surviving Hasidim, who settled in New York (Poll Reference Poll1962:29, Perlman Reference Perlman1991:132, Belcove-Shalin Reference Belcove-Shalin and Janet1995:8–9). Two studies conducted in Williamsburg in the 1950s shed light on the early stages of the Hasidic revival in the United States. Kranzler (Reference Kranzler1961) and Poll (Reference Poll1962) describe the formation of Hasidic groups in Williamsburg, mainly the dominant Satmar Hasidic sect led by rabbi Yoel Teitelbaum (1887–1979). After Teitelbaum settled in Williamsburg in 1946, Hasidim started to join him (a few dozen in 1948, and 860 households by 1961, according to Rubin Reference Haspelmath1997:47). Williamsburg had attracted Jewish immigrants before—mainly Galician and Polish Hasidim in the 1920s and 1930s, and refugees from Germany, Austria, and Czechoslovakia in the late 1930s (Kranzler Reference Kranzler1961:18–19). Kranzler describes the arrival of surviving Hasidim after the war, who chose Williamsburg because it allowed them to continue a Hasidic way of life, offering Hasidic educational institutions, etc. (Kranzler Reference Kranzler1961:14, 19). The main groups forming in Williamsburg were “Hungarian”, and they also attracted new members, including Hasidim who had come to the United States before World War II (Kranzler Reference Kranzler1961:19, Rubin Reference Haspelmath1997:47). Williamsburg gradually became a center of Hungarian Hasidic Jews (with about 10,000–12,000 Hasidim in 1959; Poll Reference Poll1962:30), who later settled in other neighborhoods as well (Kranzler Reference Kranzler1961:15).

Yiddish, Hungarian, and to a lesser extent German seem to have been the common native languages of Jews in Williamsburg in the 1940s and 1950s. The term Hungarian Hasidim refers to Hasidim from “Greater Hungary”, which includes those areas that belonged to Hungary in the 19th-century Habsburg Empire (Biale et al. Reference Assouline2018:624).Footnote ¹⁵ Hasidim who came from the so-called Unterland (a Jewish folk-geographic term, referring roughly to northern Transylvania and the mountainous portions of East Slovakia and Carpathorussia; Weinreich Reference Weinreich and Lucy1964:246, 249–250, Biale et al. Reference Assouline2018:390) spoke mainly Yiddish. Orthodox (usually not Hasidic) Jews who came from the Oyberland (a Jewish folk-geographic term referring to West Slovakia, West Hungary, and the “Seven communities” of the Burgenland; Weinreich Reference Weinreich and Lucy1964:246, 249) spoke mainly German (or Judeo-German) and Hungarian (Poll Reference Poll1962:16–17, 1965:130, Perlman Reference Perlman1991:63–65, Katz Reference Belcove-Shalin and Janet1995:12).Footnote ¹⁶ Some of these Oyberlender spoke Oyberland Yiddish, one of the last Western Yiddish varieties that survived into the 20th century (Weinreich Reference Weinreich and Lucy1964:251–252, Fleischer Reference Assouline2018:245–246).

There are no exact data about patterns of multilingualism in 1940s and 1950s Williamsburg or the relative proportion of Yiddish speakers in the Hasidic communities; only some impressionistic and anecdotal descriptions regarding language use, such as observations that many Hasidic women spoke only or mainly Hungarian (Kranzler Reference Kranzler1961:219, Poll Reference Poll1965:135).Footnote ¹⁷

When considering the possible impact of German during the formation of Hasidic Yiddish in Williamsburg, note that German could have influenced the language of Yiddish speakers in several ways. The impact of German is evident first and foremost in the Germanized Habsburg Yiddish varieties, including Unterland Yiddish. Yet another channel through which German constructions such as wh-immer could have been introduced to the nascent Hasidic Yiddish in Williamsburg was the Judeo-German speaking Jews.

Note that the term Judeo-German refers to “a variety spoken by Jews containing special vocabulary, but not otherwise differing form (local) German” (Fleischer Reference Assouline2018:239).Footnote ¹⁸ Judeo-German speakers in Williamsburg were mostly Oyberlender, such as the “Viennese Community” (Adas Yereim Vien, Kranzler Reference Kranzler1961:152, 209), and other groups of Orthodox, non-Hasidic so called Ashkenazish or Yekish Jews (lit. German Jews). Both terms, meaning ‘German’, refer in the Hungarian Jewish context to orthodox non-Hasidic streams such as Pupa, Nitra, and Kashoy. Today, such streams have become Hasidic or merged into dominant Hasidic groups (Weinreich Reference Weinreich and Lucy1964:255, Sadock & Masor Reference Assouline2018:98). The possibility of direct influence of Judeo-German is supported by metalinguistic comments such as “say ven (ven imer, oyf di yekiše šprax)…” [whenever [say ven] (whenever [ven imer] in the German language)…] (kaveshtiebel, Reference Assouline2018). Most speakers of Judeo-German (as well as German) were among the Jewish refugees from Germany, Austria, and Czechoslovakia that came to Williamsburg in 1938 and 1939 (Kranzler Reference Kranzler1961:95, 257–259).

5. American Versus Israeli Hungarian Yiddish

It seems likely that, as well as English, both Germanized Yiddish varieties and Judeo-German played a part in the rise of the wh-imer constructions, though the relative role played by each language is hard to establish. Yet even if in this specific case the impact of German may be limited or even uncertain, it is important to note that American Hasidic Yiddish has maintained many “Habsburg” or Hungarian-Galician features, such as the use of Germanisms. This feature of American Yiddish, mainly Satmar Yiddish, becomes more salient when compared to another contemporary Hungarian Hasidic Yiddish variety, spoken in Israel. While I hope that future comparative studies of contemporary Hasidic Yiddish dialects will shed more light on the subject, some preliminary explanations for this difference may be briefly presented.

To begin with, note that the communities that best maintain Yiddish in Israel are also Hungarian, mainly “zealous” extremist groups such as Toldot Aharon (originally founded in interwar Satmar; Biale et al. Reference Assouline2018:720–721).Footnote ²² However, the spoken language of such “Hungarians” is less Germanized than American Satmar Yiddish (for example, almost no German loanwords are attested in the Israeli Hungarian spoken corpus). Two possible explanations may be suggested for this difference. First, each community had its own formation history and, as a consequence, a distinct dialectal makeup and dialect contact setting. Second, the two communities show different literacy rates in Yiddish, which may also contribute to the presence of German features in the spoken language.

Speakers of Habsburg Yiddish (mainly Hungarian Hasidim), who came to the United States after World War II, during the late 1940s and 1950s, were a rather homogenous group that was free to choose where to settle. Many of them chose to settle in Williamsburg, where they formed their own organized communities, with Hungarian and Galician Yiddish as dominant varieties. They maintained strict communal boundaries, with very limited contact with outsiders, even if these outsiders were religious Jews (Poll 1962:38). Such isolated life of a homogeneous group in one place supported the maintenance of Habsburg and typical Hungarian features, which remain evident in American Hasidic Yiddish to this day.

While the Williamsburg community appears to be an isolated Hungarian enclave formed within a single decade in a single neighborhood in Brooklyn, the formation process of Hungarian Hasidic communities in Israel was more complex. Some groups of Hungarian orthodox (Ashkenazim) and Hasidim came to Ottoman Palestine before World War I and settled mostly in Jerusalem (for example in the neighborhood batey ungarin/ungeriše hayzer ‘Hungarian houses’, established in 1891). These “Hungarians” and other Habsburg Hasidim found in Jerusalem existing Hasidic communities, where Northeastern and Southeastern Yiddish varieties were widespread (that is, Yiddish dialects that are more “Slavic”, with Slavic loanwords and syntactic calques). This is because the founders of the Hasidic communities in Ottoman Palestine in the late 18th and early 19th centuries came mainly from Eastern Belarus (Wodziński 2018a:23). As for Northeastern Yiddish, it was also used in the so-called “Jerusalemite” ascetic, non-Hasidic groups, who maintain their “Lithuanian” Yiddish dialect to this day (Assouline 2010). As a result, even the contemporary Yiddish of a largely Hungarian group such as Toldot Aharon is mixed with Northeastern Jerusalemite Yiddish, which blurs its distinct Hungarian traits. Another factor that hindered the maintenance of Hungarian features in Israel was the fact that unlike Williamsburg Hasidim in the late 1940s and 1950s, Hungarian Hasidim who came to Israel after 1948 could not always choose where to settle. These Hasidim either joined family members in different locations in Israel or were sent to temporary immigrant camps, thus postponing the creation of homogenous Hasidic centers like the one in Williamsburg.

Thus, in the United States both the relative homogeneity of the Hasidic community as well as the concentration of Habsburg, mainly Hungarian Hasidim in one location in the 1940s and 1950s supported the maintenance of Habsburg and especially Hungarian features. By contrast, dialect contact in Jerusalem and the geographical dispersion of Hasidim in Israel of the 1950s hindered the maintenance of such features.Footnote ²³

The second explanation is based on the difference in Yiddish literacy rates observed within the two communities. Yiddish literacy rates are generally much higher in the United States, whereas Yiddish-speaking Hasidim in Israel read and write mostly in Hebrew (Assouline Reference Assouline2018:475). American Hasidim are thus more exposed to German loanwords common in written Hasidic Yiddish texts. As a result, higher Yiddish literacy rates may also support the maintenance of German features in the United States.

In conclusion, while contemporary Hasidic Yiddish varieties spoken in different countries are distinct mainly due to the different contact languages, their different dialectal makeup, formation history, and other sociolinguistic factors should also be considered in any comparative study of Hasidic Yiddish dialects. This is of course relevant not only to American and Israeli Yiddish, but also to European varieties. The impact of the majority language remains crucial and can explain many linguistic changes, but other factors may be important as well. For example, returning to the concessive conditional patterns discussed above, the fact that the Slavic pattern with negation is used by Israeli Hungarian speakers (while never attested in the American spoken corpus) may testify both to past and contemporary contact with speakers of the Slavic, North- and Southeastern Yiddish dialects, as well as to the impact of Israeli Hebrew, where a similar pattern exists.Footnote ²⁴

By contrast, in the United States the Slavic negation pattern is not supported by any parallel pattern: Contact with Northeastern Yiddish speakers is very limited, and a similar pattern with negation does not exist in English.Footnote ²⁵ As a result, the use of the negation pattern in the American corpora seems to be decreasing. Similarly, the use of wh-imer by American speakers can be attributed both to the impact of English and to the possible impact of German, due to the role played by Germanized Yiddish and Judeo-German in the development of Williamsburg Yiddish. Both factors contribute to American Hasidic Yiddish, which was in any case more “Germanic” and less “Slavic” to begin with, gradually becoming even more “Germanic”.

6. Future Research Directions

The study of Hasidic Yiddish is still in its early stages, and still suffers from the lack of scholarly interest in Hasidic Yiddish throughout the second half of the 20th century (Nove Reference Assouline2018). As a result, there is very little information about the formative years of contemporary Hasidic Yiddish dialects and the various factors affecting their development. This paper employed the construction wh-imer in order to speculate about the possible impact of German on the emerging American Hasidic Yiddish in Williamsburg. Further research is needed in order to assess the possible impact of past contact languages on contemporary Yiddish varieties. In the American case, due to the multilingual nature of the emerging Hasidic community in the 1950s, it would be fruitful to examine the possible impact of a Hungarian substrate, as well as that of a German superstrate (and possibly also substrate). Further studies in other Hasidic communities worldwide would inform our understanding of the interplay of past and present contact languages in the development of contemporary Yiddish varieties.