The Loshn Koydesh Component in Contemporary Hasidic Yiddish


The loshn koydesh (Hebrew and Aramaic) component has historically influenced the development of Yiddish lexis and grammar. We examine its contemporary use among 26 native speakers of contemporary Hasidic Yiddish from Israel, New York, and London using a written questionnaire examining the gender of loshn koydesh nouns, periphrastic verbs with a Hebrew/Aramaic element, and adjectives derived from the loshn koydesh element of periphrastics. Our findings show that there are differences on both the geographical and gender axes, many of which are consistent with the speakers’ varied exposure to Modern Hebrew, English, and loshn koydesh. We also found that the loshn koydesh component has developed since the pre-War stage of the language in ways that seem to affect contemporary Hasidic Yiddish usage in all locations and for both genders. We take these developments to provide evidence for the existence of this newly emergent variety of Yiddish – Contemporary Hasidic Yiddish.


Introduction1
In this article we seek to examine the role that the traditional loshn koydesh, i.e., Hebrew and Aramaic, component of Eastern Yiddish (henceforth referred to simply as Yiddish) plays in the speech and writing of contemporary male and female Hasidic Yiddish speakers living in Israel, where the majority language is Hebrew, in comparison with that of male and female Hasidic Yiddish speakers living in the United States and the United Kingdom, where the majority language is English. An additional aim of the article is to ascertain whether there are differences between male and female speakers in each language environment due to differing levels of familiarity with and exposure to the traditional pre-Modern Hebrew of classical Jewish texts on the one hand, and with spoken Modern (Israeli) Hebrew on the other hand.2 To this end, we investigate three different areas where the loshn koydesh component has formed an integral part of Yiddish grammar, namely 1) the gender of loshn koydesh nouns, 2) periphrastic verbs with a loshn koydesh element, and 3) adjectives based on loshn koydesh passive participles. In each case, we seek to investigate the effect of the contact languages on the forms and constructions in question, in addition to possible interlinked gender differences. This study is necessary because there has thus far been little scholarly attention paid to the role which the traditional Semitic component of Yiddish plays in the present-day varieties employed by Hasidic speakers and the ways in which the dominant co-territorial contact languages Modern Hebrew and English have influenced it, as well as the extent to which speakers' gender is relevant to this process. The only works we are aware of are Assouline (2010), where she discusses periphrastic verbs with a Hebrew-origin component in contemporary Israeli Haredi Yiddish, Assouline (2017:125-154), discussing lexical borrowing from Modern Hebrew in Israeli Haredi Yiddish, and Assouline (2019), providing a comparative overview of the Hebrew component in Israeli and American Haredi Yiddish. Assouline's main findings are that Modern Hebrew has exerted a significant influence on the Modern Hebrew contributed to a higher rate of use, or greater productivity, of the periphrastic forms and passive adjectives, and to a higher rate of recognition of the traditional morphological gender of loshn koydesh nouns (where that corresponds to the Modern Hebrew gender), among Israeli Hasidic Yiddish speakers than among their New York and London counterparts, who have less exposure to Modern Hebrew and greater exposure to English.
In addition to the geographical relocation of the speech communities and ensuing exposure to different contact languages, various other sociolinguistic factors have also potentially affected the communities to different degrees. One such factor is speaker gender. This is an issue worthy of investigation for two reasons. First, among Israeli Hasidic Yiddish speakers, there is a pronounced gender distinction whereby women are more likely than men to work outside of the Yiddish-speaking community and as such often have a higher rate of exposure to Modern Hebrew (in contrast to loshn koydesh, to which men typically have greater exposure), as well as an increased tendency to employ the language actively on a regular basis (see Bogoch 1999;Isaacs 1999a:119, 199b:15;Assouline 2017:93-123).4 As such, it is possible that female Israeli Hasidic Yiddish speakers will exhibit a higher degree of influence from Modern Hebrew than their male counterparts in the grammatical categories under examination. Second, in New York and London, where Modern Hebrew is not a dominant everyday language, male and female Hasidic Yiddish speakers may display divergences from each other in these respects due to their different levels of familiarity and use of loshn koydesh and English: men have a greater rate of exposure to traditional Hebrew and Aramaic texts as a result of their studies in yeshiva (religious college) and kolel (a house of study for married men), whereas women tend to speak more English and have less intensive engagement with loshn koydesh (Glinert 1999:32;Fader 2009, especially 135; see also Assouline 2018:47). Thus, it is possible that women will exhibit greater influence from English whereas men will exhibit a higher level of familiarity with loshn koydesh grammatical elements and will be able to use them more productively. Moreover, because men tend to travel internationally more than women and to relocate for marriage between Hasidic communities, they have more occasion to use Yiddish as a lingua franca; in addition, as they spend more time mixing with Hasidic Yiddish speakers in different countries, their Yiddish is more likely to converge regardless of their country of residence. For these reasons, the second major question to be examined in this article is the construction of passive and inchoative periphrastics, e.g., ‫ווערן‬ ‫נעלם‬ nelem vern 'to disappear' and ‫ווערן‬ ‫מגולגל‬ megulgl vern 'to be reincarnated.' Less frequently, the light verb ‫מאכן‬ makhn 'to do, make' can be employed, e.g., ‫חוזק‬ ‫מאכן‬ khoyzek makhn 'to mock.'8 The invariable loshn koydesh element may be a noun (some of which are also used independently as nouns in Yiddish, while others exist in Yiddish only as part of the periphrastic verbal construction); examples of such nouns are ‫מורא‬ moyre 'fear' and ‫חתונה‬ khasene 'wedding.' More frequently, the invariable loshn koydesh element is a masculine singular active or passive participle which typically exists in Yiddish only within the periphrastic verbal construction; examples include ‫מסכים‬ maskem, from the Hebrew masculine singular participle meaning 'agree' and ‫ממשיך‬ mamshekh, from the Hebrew masculine singular participle meaning 'continue.' The loshn koydesh component of the verbs in question can be drawn from the biblical stratum or from postbiblical Hebrew texts such as the Mishnah and medieval literature such as the siddur and maḥzor. Periphrastic verbs are a commonly attested and well-established feature of Yiddish (see Mark 1978:311-315;Katz 1987:172-177;and Jacobs 2005:210-212 for further discussion). They have often been used productively over the centuries, particularly in the creation of verbs referring to concepts known from the Bible or Jewish tradition (e.g., ‫זיין‬ ‫רגל‬ ‫עולה‬ oyle regl zayn 'to go on pilgrimage,' ‫זיין‬ ‫בעגל‬ ‫חוטא‬ khoyte b'eygl zayn 'to commit the sin of the Golden Calf,' etc). Yiddish periphrastics have parallels in some other Germanic languages, including so-called noun-incorporation verbs (also known as nominal separable complex verbs) in Dutch (Weggelaar 1986, Booij 1990), but such verbs tend to incorporate nouns rather than adjectives or participles, and are not used as a means to borrow lexical material from other languages. A closer parallel to Yiddish periphrastics with a loshn koydesh component is found in Judezmo (Ladino/Judeo-Spanish), which has a similar type of construction with an auxiliary verb plus a Hebrew-origin participle (see Bunis 2009).
From a descriptive point of view, the loshn koydesh elements of the periphrastic verbs are often based on the masculine singular participle form of a Hebrew verb in the qal (basic) stem, while others are based on the masculine singular participle form of a verb in the piel (so-called 'intensive') stem or, somewhat less commonly, the masculine singular participle of a verb in the hifil (causative) stem or hitpael (reflexive) stem, although these derivational relations do not necessarily hold in Yiddish speakers' mental grammars. In many cases, the periphrastic verb in question has a synonym or near-synonym from the Germanic lexical component of Yiddish. For example, the everyday Yiddish verb ‫זיין‬ ‫מוחל‬ moykhl zayn 'to forgive' has a near-synonym ‫פאַרגעבן‬ fargebn 'to forgive,' which derives from the Germanic component of the language and can typically be used in the same contexts as the periphrastic verb.

Adjectives Derived from Hebrew Passive Participles
An area of Yiddish morphosyntax related to the periphrastic verbs with a loshn koydesh element is the use of Hebrew-derived passive participles as attributive and predicative adjectives. The loshn koydesh element of a Yiddish periphrastic has an invariable form, which is often a Hebrew masculine singular active participle. In Hebrew, such participles are derivationally related to an associated participial form with a passive or resultative sense. This occurs in a variety of verbal stems, namely the qaṭul (the passive participle of the qal or G-stem, i.e., the basic Hebrew active and stative verbal pattern), the nifal (also known as the N-stem, i.e., the basic passive Hebrew verbal pattern), the pual (the passive equivalent of the piel or D-stem, i.e., the Hebrew stem with a doubled middle radical which sometimes has a factitive or distributive sense), and the hufal (the passive equivalent of the hifil or C-stem, i.e., the causative stem in Hebrew).
Historically, some of these adjectival forms existed in Yiddish. For example, the Yiddish periphrastic verb ‫זיין‬ ‫מכבד‬ mekhabed zayn 'to honor, to respect' is derivationally associated with the Yiddish adjective ‫מכובד‬ mekhubed 'respected, esteemed' (Niborski 2012:283). However, the derivation of adjectives from loshn koydesh participial forms does not appear to have been a productive part of Yiddish grammar: only a handful of examples exist.

Methods and Data
In the following sections we discuss the questionnaire used for this study, the participants and their selection process, and the individual tasks and findings.

Questionnaire
In order to investigate the domains of nominal gender, use of periphrastics, and derivational patterns in adjectives, as they relate to the loshn koydesh component of Yiddish, we developed a questionnaire to elicit written data and judgments from native Hasidic Yiddish speakers in a) Israel (primarily, Bnei Brak and Jerusalem), b) the New York area, and to a smaller extent c) London's Stamford Hill. The questionnaire was administered in Hasidic Yiddish. The questionnaires were administered in person on location (with a small number administered by Zoom call during lockdown) by either a native or non-native fluent speaker of Hasidic Yiddish. The questionnaire was primarily a written task, although the experimenter was present for the duration of its completion. Sessions were usually audio recorded, and where they were not, the experimenter made notes about aspects of the discussion with the participant. Our findings are primarily based on the written tasks, but where the audio data are informative, they are mentioned in our discussion below. The questionnaire was composed of three separate tasks, each of which focused on a different linguistic domain. These are described in detail below, together with the results we obtained for each. The full questionnaire is provided in Appendix 1.

Participants
The participants consisted of 26 native speakers of Hasidic Yiddish from Israel, the New York area, and London. This total can be broken down into 10 speakers who were born and raised in Israel (specifically Jerusalem, Bnei Brak, and Ashdod), of whom 6 were male and 4 female; 12 who were born and raised in the New York area9 (specifically the Brooklyn neighborhoods of Williamsburg, Flatbush, and Borough Park, as well as Hasidic communities in New York State and New Jersey, i.e., Lakewood, Monsey, Monroe, and New Square), of whom 7 were male and 5 female); and 4 who were born and raised in London (specifically Stamford Hill), of whom 2 were male and 2 female. The age range of the speakers was 18 to 59, and they were members of 9 different Hasidic groups (Belz, Biale, Karlin, Nadvorner, Pinsk, Pupa, Satmar, Skver, and Vizhnitz) in addition to some who reported their affiliation as "general Hasidic." Table 1 provides information about the Hasidic affiliation, age range, hometown and current place of residence, and gender of each participant.10 Participants were selected on the basis of being native Hasidic Yiddish speakers who were raised and/or educated in Yiddish, and for whom Yiddish was the main language into adulthood. We sought to recruit speakers from a variety of Hasidic affiliations and age ranges, and to have a balance of genders.11 9 One of these participants was born in Israel but moved to the New York area before the age of 5. We included this participant with the New York speakers as she has lived in the New York area since a very young age. 10 Note that age ranges instead of precise ages have been given for each speaker due to confidentiality concerns. For the same reason, precise geographical details for individual speakers (e.g., Williamsburg, Monsey, Borough Park, Bnei Brak, etc.) have been avoided in favour of more general information (i.e., UK, Israel, the New York area). The specific geographical details do not have a bearing on our results, so this precise and potentially compromising information is not germane for the study. 11 Our interview data reveal that boys in all Hasidic communities tend to be educated in Yiddish and loshn koydesh in parallel from the age of three. In higher grades (from age Journal of Jewish Languages 8 (2020) 39-89 With respect to dialect, all participants described their variety of Yiddish as "Hasidic Yiddish" or heymish yidish 'observant Yiddish.'12 All Israeli participants were fluent in Modern Hebrew, and all New York and Stamford Hill participants were fluent in English (although for many male speakers in New York, Yiddish and loshn koydesh are the preferred written medium). Most Israeli participants, especially women, had some familiarity with English, but all are more comfortable in Yiddish or Modern Hebrew. Very few female participants in New York or Stamford Hill were fluent or even comfortable in Modern Hebrew or loshn koydesh, but many male participants had spent extended periods of months or even years (e.g., during yeshiva studies) in Israel, and were consequently more confident in Modern Hebrew. However, no male participants from outside of Israel spoke a native-like Modern Hebrew, as judged by a native Modern Hebrew speaker. The male participants' education, being largely conducted in loshn koydesh, made them more familiar with the language than the female participants. M 13 and up) loshn koydesh is increasingly used as a medium of education alongside and even in preference to Yiddish. In Israel, girls are often educated exclusively in Yiddish in the early years, but in higher grades Modern Hebrew is increasingly intermingled. In Stamford Hill and in the New York area, girls tend to study religious subjects for half the day, when Yiddish is used as the primary medium, and secular subjects for the other half, when English is used as the primary medium. These findings are largely supported by those of Abraham (1999) and Glinert (1999) for Stamford Hill, Fader (2009, especially 126-138) and Nove (2018b:88) for New York, and Bogoch (1999)

Task 1: Nouns
We chose to investigate the question of morphological gender of loshn koydesh nouns in order to ascertain the extent to which contemporary Hasidic Yiddish speakers' usage in this regard is connected to their understanding of traditional textual Hebrew and of Modern Hebrew. Following on from earlier examinations of Israeli Hasidic Yiddish (Assouline 2017) and New York Hasidic Yiddish (Krogh 2012(Krogh , 2018; see also Nove 2018a), we have demonstrated in previous work that morphological gender is not a feature of the Yiddish of Hasidic Jews in London, New York, or Israel (Belk, Kahn, & Szendrői 2020;and Belk, Kahn, & Szendrői under review). In spoken use, speakers pronounce the definite article as an invariable form de with a realization varying between /ɛ/ and schwa, while in written usage the traditional gendered forms ‫דער‬ der, ‫די‬ di, and to a much lesser extent ‫דעם‬ dem and ‫דאס‬ dos are used interchangeably and inconsistently even by a single speaker. Speakers often resort to a default, the choice of which appears to be influenced by phonological factors (Belk, Kahn, & Szendrői under review). Israeli and New York speakers tend to prefer writing ‫די‬ di to represent the definite article in writing, while London speakers sometimes use ‫דער‬ der. This is because the non-rhotic nature of London English and consequently Yiddish pronunciation of these speakers brings this form phonetically close to the spoken de form. Some, especially young speakers, even Table 1 Biographical data for participants in our loshn koydesh grammatical survey (cont.) use the innovative form ‫דע‬ de in writing. In previous work, we have not tested loshn koydesh nouns systematically for gender assignment, but to the extent that they form an integral part of Yiddish vocabulary, we would expect that they too have lost their gender assignment in contemporary Hasidic usage. At the same time, Yiddish-speaking Hasidim in New York and London are often familiar with traditional loshn koydesh through study, recitation, and regular hearing of texts such as the Torah, haftaroth, and siddur, as well as, to a lesser extent, with Modern (Israeli) Hebrew. Speakers are often aware of the fact that Hebrew (whether loshn koydesh or Modern Hebrew), in contrast to their own variety of Yiddish, does have morphological gender. For example, one participant in our research commented that "Yiddish is not like Hebrew or French" in this respect. Yiddish-speaking Hasidim in Israel are frequently bilingual in Modern Hebrew and are typically exposed to this language on a daily basis, which likely further entrenches familiarity with the Hebrew morphological gender system and may in turn influence speakers' understanding of morphological gender vis-à-vis the Hebrew-origin nouns in their Yiddish. This Hebrew-Yiddish language contact and diglossia, both in Israel and in the US and UK, raises the question of whether Hebrew-origin nouns are treated differently with respect to morphological gender than Yiddish nouns not derived from Hebrew, which our previous research has demonstrated lack morphological gender completely. Thus, through this research we seek to ascertain whether Hebrew-derived nouns in contemporary Hasidic Yiddish are more likely to be associated with morphological gender than those not derived from Hebrew. Likewise, our study investigates whether there is a difference in this respect between Israel on the one hand, where Modern Hebrew is the main language of the public domain, and New York and London on the other, where Hasidic Yiddish speakers are not typically exposed to Modern Hebrew on a daily basis. Furthermore, we will examine the question of whether there is a difference between male and female speakers in each case, as male speakers in all communities have greater exposure to loshn koydesh than their female counterparts.
On the face of it, it would perhaps seem natural to expect that greater exposure to Modern Hebrew would facilitate the retention of Hebrew gender assignment for Hebrew-origin nouns in Yiddish. But this is in fact an illusion. If our previous work on the loss of morphological gender in contemporary Hasidic Yiddish is on the right track, then in fact our expectation should be that morphological gender in Modern Hebrew and loshn koydesh should not influence speakers' choice of gender morphology in Yiddish. This is because the Yiddish of Hasidic speakers does not mark gender morphologically, so the original gender of the loshn koydesh elements does not manifest itself in Yiddish in the grammar of Hasidic speakers. Our prediction is thus that morphological gender in Modern Hebrew and loshn koydesh should not influence speakers' choice of gender morphology in Yiddish, and that this effect should hold regardless of speaker gender or exposure to Modern Hebrew.
Description of Task 1 (Nouns) The noun task was designed to allow us to investigate realization of gender agreement morphology, specifically of loshn koydesh nouns. We tested 36 nouns, of which 29 were loshn koydesh nouns and 7 Germanic controls. The nouns were all commonly used items. We co-varied Hebrew gender (16 feminine, 20 masculine) and Standard Yiddish gender (12 feminine, 8 neuter, 16 masculine) in such a way that both nouns with matching and non-matching Hebrew-Yiddish gender were tested. A full list of the nouns is provided in the Appendix. The noun task comprised 36 sentences with two blanks in each: one in place of the subject's determiner, and the second in place of the main verb. For each blank, both a Modern Hebrew and an English word was given: the definite determiner for the first blank (i.e., the/‫)ה‬ and an infinitival verb for the second (e.g., smell/‫.)להריח‬ Participants were asked to translate (and conjugate in the present tense, in the case of the verb) each word in the blank. The subject of the sentence was a noun derived either from loshn koydesh or, as a control, from Germanic. The verbal conjugation task was included as a distractor. An example of the task is given in (1). Note that participants only saw the example in Hebrew letters and not the transliteration, which is provided here for the reader's convenience.
(1) ‫ערנסט.‬ ‫זייער‬ ________________________________________ ‫פנים‬ ________ ‫להיות/‪be‬‬ the/‫ה‬ ________ ponem _____________________ zeyer ernst. '_______ face________________________ very serious.' the be Results Israeli and New York participants behaved fairly uniformly in this task. Table 2 shows the percentage of participants' responses that matches pre-War and Standard Yiddish gender morphology. In both groups, participants produced ‫די‬ di as the definite article most of the time, irrespective of the noun's Standard Yiddish gender. In contrast, participants from Stamford Hill showed a more varied pattern of behavior, with 2 of the 4 participants matching the Israeli/ New York participants' behavior, one participant assigning the article ‫דע‬ de uniformly, and one participant assigning the article ‫דער‬ der in the vast majority of cases. Thus, all the participants, irrespective of geographical location, assigned a default form of the article, but there was some variation, at least in Stamford Hill, regarding which form is taken to be the default. Almost all participants therefore apply a default form of the article irrespective of the noun's gender. Note also that the default form is not uniformly applied 100% of the time by the speakers. There is some variation with speakers occasionally employing even ‫דאס‬ dos, ‫דעם‬ dem, and the innovative form ‫דע‬ de. In sum, our findings indicate a lack of awareness of morphological gender in contemporary Hasidic Yiddish by its speakers.13 This pattern meshes with our previous findings on article assignment to Germanic origin nouns by Hasidic Yiddish speakers as described (Belk, Kahn, & Szendrői 2020;Belk, Kahn, & Szendrői under review). The same pattern was observed with our Germanic control words. Here, for nouns that are feminine in Standard Yiddish, Israeli and New York speakers used an article that corresponded to the article used in Standard Yiddish 95% and 100% of the time, respectively. For neuter nouns, Standard-like usage was 0% by Israeli speakers and 2.7% for New York speakers. Masculine nouns received a Standard-like gender assignment 5% of the time by Israeli speakers and 0% of the time by New York speakers. As far as Stamford Hill speakers are concerned, the same individual default patterns carried over to Germanic nouns as observed for loshn koydesh nouns. 14 Stamford Hill speakers are presented individually due to their smaller number, which makes aggregate data unreliable. 15 Note that one participant from New York did not complete this task.
As Table 3 shows, participants did not show a particular sensitivity to the original Hebrew gender assignment of the nouns either, whether they are from Israel, New York, or London. Importantly, as we predicted, participants from Israel were not more likely to assign articles matching the original Hebrew gender of the nouns. For example, the noun ‫לשון‬ loshn 'language,' which is feminine in Hebrew and neuter in pre-War/Standard Yiddish, was nonetheless assigned the definite article ‫דער‬ der (which is traditionally masculine nominative) by five participants and ‫די‬ di by all the other participants. Conversely, the noun ‫כוח‬ koyekh, which is masculine in both Hebrew and Standard/pre-War Yiddish, is assigned the definite article ‫די‬ di by almost all participants. This is consistent with our hypothesis that they do not have the concept of grammatical gender in Yiddish. We were also interested in whether speakers' gender or a combination of gender and geographical origin was an important factor in noun gender assignment for loshn koydesh nouns. The breakdown of the findings is given in Tables 4 and 5 for Standard Yiddish and original Hebrew grammatical gender assignment, respectively. We do not report gender breakdowns for the London speakers, because there were not enough participants in this location to make such comparison informative.  The breakdown of results shows that Israeli men showed a slightly higher association between the article ‫דער‬ der and loshn koydesh nouns that have masculine gender in Standard Yiddish (40.6%) and in the original Hebrew (39.3%) compared to the other participants (25% and 25.7%, respectively), while Israeli women showed a slightly lower association. Note, however, that this performance is nowhere near what would be expected if they actually treated loshn koydesh nouns as having grammatical gender. Gender assignment errors for frequent nouns in languages with grammatical gender are vanishingly small among native speakers (Schmid 2002). Rather, our data reveal two general patterns or tendencies for occasional ‫דער‬ der assignment despite an overall general default tendency to assign ‫די‬ di to most loshn koydesh nouns. Some Israeli men employ both of these tendencies, and most Israeli women employ neither, while New York speakers seem to employ one or the other, mostly not both, and with no discernible speaker gender pattern. This gives rise to the differential result of Israeli men and women. We examine these two patterns of increased ‫דער‬ der assignment in turn.
One pattern can be observed whereby some speakers tend to select the form ‫דער‬ der (traditionally the masculine nominative definite article) for human male referents. Thus, in about 42% of cases speakers selected ‫דער‬ der for the nouns ‫חבר‬ khaver 'friend,' ‫שכן‬ shokhn 'neighbor,' ‫מלך‬ meylekh 'king,' ‫חתן‬ khosn 'bridegroom,' and ‫גנב‬ ganev 'thief.' However, this result does not suggest that speakers associate ‫דער‬ der with masculine gender. Rather, as almost 60% of the time speakers selected ‫די‬ di for male human nouns, it indicates that there is no grammatical rule associating words denoting human or animate male entities with ‫דער‬ der. The results suggest that ‫די‬ di is not thought of as a feminine article for speakers of contemporary Hasidic Yiddish, as that would render it incompatible with masculine, and especially animate masculine, nouns. ‫די‬ di can instead be understood as the default article, and its use may be disfavored due to other patterns or tendencies. The assignment of ‫דער‬ der to words denoting male humans is one such tendency. It is likely that this tendency has its origin in earlier stages of the language, where ‫דער‬ der was associated with masculine grammatical gender, which in turn was associated with semantic maleness. However, this pattern appears to be a fossilized relic from a historical variety rather than a productive part of the grammar of contemporary Hasidic Yiddish. Furthermore, it is almost never reflected in the speech of contemporary Hasidic Yiddish speakers, but only in the written language which medium is generally considered more conservative (Fromkin, Rodman, & Hyams 2010).
A further, perhaps more interesting and innovative, pattern can also be observed whereby a significant minority of speakers assign ‫דער‬ der to the nouns ‫שבת‬ shabes 'Sabbath,' ‫ים‬ yam 'sea, ocean,' ‫שלום‬ sholem 'peace,' ‫ספר‬ seyfer 'Jewish religious book,' ‫סידור‬ sider 'prayer book,' and ‫לשון‬ loshn 'language.' Metalinguistic discussion with participants suggests that they associate the form ‫דער‬ der with objects or people that are considered important.16 The concept of importance is clearly subjective but is associated with items or concepts of religious or cultural significance such as the Sabbath, prayer books 16 Another illustration of the pervasive extent of this phenomenon can be seen by the use of the traditionally masculine accusative article ‫דעם‬ dem for the Germanic-origin noun ‫שפראך‬ 'language' (which is feminine in Standard Yiddish) in the context of a song praising the significance and importance of the Yiddish language, by Levy Falkowitz and the Shira Choir: ‫אידיש‬ ‫שפראך‬ ‫דעם‬ ‫נאר‬ ‫מען‬ ‫רעדט‬ ‫אידן‬ ‫אונזערע‬ ‫ביי‬ bay undzere yidn redt men nor dem shprakh yidish 'our Jews only speak the Yiddish language.' See the recording online at: https://www.youtube.com/watch?v=3puJAZg2TEU. and other holy books, and peace, but not for example ‫מעשה‬ mayse 'story' or ‫פנים‬ ponem 'face.' It is not clear from our data whether this pattern has its roots in a stronger tendency for words denoting important or significant objects to retain their original gender assignment in a fossilized form simply because they are words that are perceived to be more important. In fact, some of our data suggest that this pattern is more far-reaching and more innovative: we find nouns denoting important or significant objects that were assigned feminine or neuter gender in pre-War/Standard Yiddish, as well as in Hebrew, which now fall into this pattern, such as ‫לשון‬ loshn 'language,' or ‫חלה‬ khale 'challah.' This tendency, like the previous one, is nearly absent from spoken contemporary Hasidic Yiddish.
Interim Summary Returning to our original predictions and research questions, we find that, in accordance with previous work, loshn koydesh nouns are not associated with morphological gender in Yiddish and do not preserve their pre-War or Standard Yiddish grammatical gender or the original Hebrew gender assignment. There is no difference between male or female speakers, despite the former group's higher exposure to loshn koydesh. There is also no advantage for Israeli speakers, who have greater exposure to Modern Hebrew (especially Israeli women), in retaining the original loshn koydesh gender assignment. We observe two further trends in the data that we did not anticipate: some (often male Israeli) speakers sometimes assign the article ‫דער‬ der to words denoting male human entities as well as to words whose meaning is perceived as important or significant.
The task demonstrates that there is no particular distinction made by Hasidic Yiddish speakers in either majority Hebrew-or English-speaking countries in terms of morphological gender between nouns from the loshn koydesh component and nouns from the Germanic component. Overall, our findings in the nominal domain are consistent with the findings of Belk, Kahn, & Szendrői (2020) in demonstrating a complete lack of morphological gender in contemporary Hasidic Yiddish. That this is true even where speakers know the gender of a given noun in the language from which it is borrowed underlines the fact that the concept of morphological gender is entirely absent from the mental grammars of contemporary Hasidic Yiddish speakers. There is a tendency (in writing, and typically not in speech), for words denoting male human entities to optionally make use of the article ‫דער‬ der, which was historically associated with masculine grammatical gender. There is also some indication that traditional masculine gender morphology may be associated (again, in writing) with the importance or significance of the concept denoted by certain nouns. These tendencies are strongest among Israeli males and weakest among Israeli female speakers.

Task 2: Periphrastic Verbs
Periphrastic verbs with a loshn koydesh element are a characteristic aspect of Standard Yiddish and traditional pre-War Yiddish dialects. By investigating the use of these periphrastics among contemporary Israeli, New York, and London Hasidic Yiddish speakers we can ascertain whether periphrastics are still a productive element of the grammar of contemporary Hasidic Yiddish, and whether any differences in their use can be detected between subgroups of speakers. Furthermore, we aim to explore speakers' attitudes to synonyms to periphrastics derived from the Germanic component of Yiddish as well as direct (i.e., phonologically and morphologically unintegrated) and integrated borrowings from English and Modern Hebrew.
This task was designed to test not just the acceptability of each lexical item, but the relative acceptability of a number of potential synonyms: traditional Yiddish periphrastics, synonymous verbs from the Germanic component of Yiddish, direct borrowings from English and Modern Hebrew, and grammatically integrated borrowings from English and Modern Hebrew. As English is a Germanic language, it is relatively straightforward to add Yiddish verbal morphology onto an English verbal root. However, the issue of integrating Modern Hebrew roots into Yiddish grammar deserves further comment.
Despite the fact that periphrastic verbs typically have a loshn koydesh element, in many cases this element is not actually commonly used in Modern Hebrew, which instead employs a different form for the same meaning. For example, the Yiddish periphrastic verb ‫זיין‬ ‫מוחל‬ moykhl zayn 'to forgive' does not have a frequently used equivalent of the same root in Modern Hebrew, in which the cognate form ‫למחול‬ limḥol 'to forgive' is reserved for high-register and religious contexts. Instead, the everyday verb meaning 'to forgive' is ‫לסלוח‬ lisloaḥ. Traditionally, one way to integrate loshn koydesh verbs into Yiddish grammar was by use of periphrastics, so we wanted to investigate whether speakers accepted novel periphrastics based on Modern Hebrew roots and how such novel periphrastics compared to other synonyms. In short, we are interested in investigating i) speakers' familiarity with a variety of loshn koydesh periphrastics, ii) speakers' familiarity with synonyms to these periphrastics, derived from the Germanic component of Yiddish, iii) patterns of borrowing from English and Modern Hebrew, and particularly iv) grammatical integration of borrowed items into Yiddish.
We predict that participants from Israel, New York, and London should all be familiar with the standard periphrastic and, where available, the Germanic equivalent. Furthermore, we predict that participants in Israel will more readily accept forms with Modern Hebrew cognates and direct borrowings from Modern Hebrew, while those in New York and London will more readily employ forms derived from English, either with or without the Yiddish infinitival marker. Finally, we predict that the Israeli Hasidic Yiddish speakers will be more comfortable with the novel Yiddish periphrastics in everyday contexts than their peers from English-speaking countries, where their exposure to Modern Hebrew verbs is much lower. Conversely, we predict that the Hasidic Yiddish speakers from New York and London may exhibit a greater preference for Germanic roots than the Israeli speakers, since they are often cognate with the English verbs to which they have regular exposure. For Germanic verbs without a transparent English cognate, we aim to ascertain whether our participants in majority English-speaking countries will nevertheless prefer them to the loshn koydesh forms and/or the forms specifically based on Modern Hebrew verbs.
We may also predict differences between men and women. In particular, men in Hasidic communities tend to use Yiddish more regularly than women, while women tend to use the language of the surrounding community more often than men. Thus, we predict that women in Israel should be more likely to borrow from Modern Hebrew than their male counterparts, while women in New York and London should be more likely to borrow from English than their male counterparts. Men in New York and London may prefer English borrowings with the Yiddish infinitival marker (a preference that may be less marked or entirely absent in women). Men in both groups may be more likely to use Germanic roots than their female counterparts, as their Yiddish vocabulary may be richer.
Description of Task 2 (Periphrastic Verbs) This task aims to investigate the acceptability of loshn koydesh periphrastics and their synonyms. Participants were presented with 21 Yiddish sentences, each of which included a missing infinitival verb. For the infinitive, between four and seven different options were provided in a randomized order. In each case, the options were minimally composed of a) a loshn koydesh periphrastic (selected on the basis of their common use in pre-War and Standard Yiddish, and to represent a range of auxiliaries and loshn koydesh components), b) the equivalent verb in Modern Hebrew, c) the equivalent verb in English (presented in the Yiddish alphabet) with the Yiddish infinitival marker ‫-ן‬ (-n), and d) the equivalent verb in English (presented in the Yiddish alphabet) without the infinitival marker.17 Additional options were provided where there is e) a synonymous word in Yiddish from the Germanic component, f) a synonymous loshn koydesh or Modern Hebrew periphrastic, or g) more than one English or Hebrew equivalent. An example is given in (2). Note that participants only saw the example in Hebrew letters and not the transliteration, which is provided here for the reader's convenience.
(2) The participant's task was to rank the naturalness of each option in descending order. In other words, the word that they would use most often, or that sounds most natural to them, would be ranked '1,' while a word they would use less often, or that sounds less natural to them, would be ranked lower. Where they would never use a particular option, they were asked to indicate this with an 'X' or leave the box blank.

Results
The results of this task are summarized in Table 6. Here we can see that all speakers everywhere accept loshn koydesh periphrastics. This feature continues to be an integral part of Yiddish in the contemporary Hasidic setting. Most speakers also find Germanic equivalents grammatical, again without any geographical differences. Speakers also uniformly reject direct loans from English. 17 The use of English lexical items in Yiddish script is an extremely commonly observed practice in Hasidic Yiddish-speaking communities; see Fader (2009:108).
Furthermore, although Table 6 indicates only the acceptability of each item type, our results show that, perhaps unsurprisingly, items that are accepted by a greater number of participants also tend to be ranked higher than items that are less widely accepted. Differences between the locations emerge. On the one hand, Israeli speakers are more accepting of Modern Hebrew-based innovative periphrastics and, to a lesser extent, of direct borrowings from Modern Hebrew than speakers from English-speaking countries (see also Assouline 2019). On the other hand, New York speakers and, to a lesser extent, London speakers are more accepting of English-origin verbs with Yiddish inflection compared to Israeli speakers. This result is as expected and is a straightforward effect of contact with Hebrew and English respectively. Somewhat more surprising is the fact that very few speakers indicated that they would use an English infinitive without the Yiddish infinitival marker ‫-ן‬ -n, and relatively few speakers would use a Hebrew infinitive (for Israeli speakers) or Yiddishized English infinitive (for New York and London speakers). Nevertheless, this pattern of borrowing is more informative than it might at first seem. The fact that New York and London speakers reject pure English borrowings in favor of borrowings that have been grammatically integrated into Yiddish (through addition of the Yiddish infinitival ending), may indicate that even when borrowing, these participants make use of their Yiddish grammar. By contrast, Israeli speakers borrow from Modern Hebrew without changing the form of the infinitive: they use the Hebrew infinitival marker ‫ל-‬ lrather than the Yiddish infinitival marker. Additionally, they pronounce the infinitive with a Modern Hebrew accent, rather than the Ashkenazi accent that would be used for the loshn koydesh component of periphrastics. These instances of borrowing from Modern Hebrew are thus less grammatically integrated than borrowings from English. At the same time, speakers from Israel appear to be more creative in their use of periphrastics derived from loshn koydesh, especially compared to New York speakers. While almost all speakers in all communities prefer the standard periphrastic to the alternatives in all carrier sentences, Hebrew speakers accept non-standard periphrastics, or periphrastics based on Modern Hebrew roots, more often than speakers from New York or London. This suggests a greater degree of integration of the Modern Hebrew forms into the periphrastic construction by Israeli speakers. So, Israeli speakers appear to be treating Modern Hebrew as a mere borrowing from one perspective, and as an integrated part of Yiddish from another.
Let us now turn to the breakdown of the results for speaker gender. As Table 7 reveals, there are some significant gender differences, especially among Israeli speakers. First, we can see that Israeli women are the least accepting of Germanic equivalent forms. They tend to reject innovative, integrated Modern Hebrew periphrastic forms, and instead have a much higher acceptance rate of straight borrowings of Modern Hebrew infinitives than any other group. In contrast, Israeli male speakers show the highest acceptance of Germanic equivalents among all the groups. They never accept direct borrowings of Modern Hebrew infinitives, and they are the most accepting of integrated, innovative Modern Hebrew periphrastics.
Thus, we find a difference in borrowing strategies between the two genders among Israeli speakers. The use of a loshn koydesh-derived periphrastic, rather than a Hebrew infinitive, indicates that the borrowing is more grammatically integrated into Yiddish. The finding may therefore be indicative of more productive use of Yiddish grammatical patterns among male speakers in Israel as compared to female speakers, which would be in line with the greater degree of exposure to Yiddish in daily life among Hasidic men compared to women and a greater degree of exposure to Modern Hebrew among Hasidic women in Israel. Among New York speakers, the gender differences were less pronounced. At the same time, the breakdown of the data by speaker gender clearly reveals that the trend to accept Yiddishized English words is driven by New York women more than by men. Again, this points to a greater degree of influence of English in the case of New York women compared to men, but the degree of integration of the borrowing into Yiddish grammar is higher than for Israeli women, who also accept direct borrowings without grammatical integration. New York female speakers are also the least accepting of innovative, integrated Modern Hebrew-based periphrastic forms, while New York men accept these at about the same rate as Israeli women. Again, this shows that the periphrastic construction is less flexible and less productive in the grammars of female speakers from New York compared to the grammars of men (in both New York and Israel), which is presumably explained by women's lesser exposure to loshn koydesh compared to men. Interim Summary Overall, the results of Task 2 support our predictions that a greater familiarity with Modern Hebrew and loshn koydesh will correlate with higher usage of and more productivity with periphrastics derived from loshn koydesh. Additionally, and perhaps unsurprisingly, speakers who are exposed to more Modern Hebrew will tend to borrow from Modern Hebrew, while those exposed to more English will tend to borrow from English. What is more unexpected is that borrowings from English are more grammatically integrated than those from Hebrew. All speakers accepted infinitives from the Germanic component of Yiddish. There was no advantage of an English-speaking background even where those infinitives are cognate with English. We found a strong gender split among Israeli speakers in that female speakers are driving direct borrowing and are least accepting of innovative, integrated periphrastic forms, while male speakers showed the opposite pattern. Taken together, these results suggest that knowledge of loshn koydesh and Modern Hebrew have a variety of effects on the selection of verbal lexis and strategies for integrating these verbs into the grammar.

Task 3: Adjectives Derived from Loshn Koydesh Passive Participles
As discussed in the historical overview, Yiddish periphrastics are often based on a loshn koydesh masculine singular active participle. These participles also have passive variants, a small number of which were used as adjectival forms in historical varieties of Yiddish. The derivational relations between the active forms and their passive counterparts within the grammar of loshn koydesh and Modern Hebrew are well known, but it is not clear the extent to which these derivational patterns are active in the mental grammar of Yiddish speakers (either pre-War or contemporary Hasidic) or whether they represent borrowings from Semitic. Because these adjectival forms have varying rates of frequency in everyday Yiddish speech, we predict that there may be a difference in responses between the speakers whose co-territorial language is Modern Hebrew and those whose co-territorial language is English. This is due to the fact that the passive participles of these roots are typically employed in Modern Hebrew as adjectival forms, e.g., ‫מסודר‬ mesudar 'organized,' ‫מצליח‬ maṣliaḥ 'successful.' The high frequency of these forms in Modern Hebrew may result in Israeli Yiddish speakers' increased ability to use the corresponding Yiddish adjectives productively, whereas the Hasidic Yiddish speakers from English-speaking countries may be less comfortable producing such adjectival forms. There may also be a difference between male and female speakers in both countries due to their varying levels of familiarity with loshn koydesh and Modern Hebrew (see Introduction).
An additional question concerns the use of Yiddish adjectival morphology in the formation of the adjective associated with these periphrastic forms. In pre-War and Standard Yiddish, the Germanic suffix ‫-דיג‬ -dig (spelt ‫-דיק‬ -dik in the YIVO orthography) can be used to form adjectives from nouns and other non-adjectival parts of speech, including those from the loshn koydesh component. For example, the noun ‫שבת‬ shabes 'Sabbath' has the adjectival form ‫שבתדיג‬ shabesdig 'pertaining to the Sabbath, suitable for the Sabbath,' and the noun ‫חן‬ kheyn 'charm' has the corresponding adjective ‫חנעוודיג‬ kheynevdig 'charming.' Similarly, the noun ‫וואך‬ vokh 'week,' from the Germanic component of Yiddish, has the adjectival form ‫וואכעדיג‬ vokhedig 'pertaining to the weekdays.' Our study required participants to derive adjectives from a selection of the periphrastic verbs tested in the same research questionnaire, allowing us to gauge the productivity of the suffix ‫-דיג‬ -dig in their language. We wanted to examine the extent to which the predictable formation of passive participles based on these verbal roots in Modern Hebrew has led to an increased use of parallel adjectival forms in the Yiddish of Israeli speakers, and whether a similar pattern is found in the Hasidic Yiddish of English-speaking countries. Belk, Kahn, & Szendrői (2020) show that former gender agreement morphology has been reanalyzed as attributive marking in contemporary Hasidic Yiddish. Thus, we also predict that such morphology should appear on all and only attributive adjectives in contemporary Hasidic Yiddish, regardless of the component of Yiddish they are derived from. Such marking would indicate that the adjectives in question are integrated into the grammar of contemporary Hasidic Yiddish. On the other hand, direct borrowings from another language may be expected to be less well integrated.
Thus our aim was to investigate morphological patterns in the derivation of adjectives from loshn koydesh past participles. We predict that women in Israel, who are more familiar with Modern Hebrew, will show similar patterns of behavior to men in both New York/London and Israel in preferring Semitic morphological patterns, as men in both communities use pre-Modern Hebrew and Aramaic extensively in their daily lives. Women in New York and London are exposed to less Modern and pre-Modern Hebrew, and very little Aramaic, and so may either prefer Germanic patterns or struggle to derive adjectives from loshn koydesh past participles at all. Equally, we predict that speakers from New York and London preferentially use the adjectivalizing suffix ‫-דיג‬ -dig, while those in Israel prefer Hebrew morphological changes. Finally, we predict that attributive adjectives, or at least those that are integrated into Yiddish grammar, will appear with attributive adjectival morphology.

Description of Task 3 (Adjectives)
Task 3 focuses on adjectives derived from loshn koydesh passive participles. Participants were presented with a sentence in the past tense, which was followed by two related sentences with gaps in the position of a predicative and an attributive adjective, respectively. For example, the sentence "Zisi has cleaned her room" might be followed by "Now her room is ________" and "It is a ________ room". Participants were asked to fill in the blanks with an appropriate word, and were provided with a model item demonstrating that the appropriate word is derived from the past participle. The entire questionnaire is provided in the appendix to this article, but an example is provided in (3) for ease of reference. Again, as before, the transliteration was not included in the original questionnaire and is provided for the reader's benefit.
(3 Task 3 included 14 sets of sentences. Two sets made use of Germanic roots as controls, two sets used periphrastics involving loshn koydesh nouns, and the remaining 10 sets used loshn koydesh periphrastic verbs from Task 2. The verbs were selected in order to ensure a range of light verbs and of loshn koydesh participle types. Furthermore, we included two verbs that were likely to produce semantically marked adjectives (e.g., ‫זיין‬ ‫מסכים‬ maskem zayn 'agree') even for Modern Hebrew speakers, with the intention of testing the boundaries of participants' creativity with respect to Hebrew-origin deverbal adjectives. The design of the task was open-ended enough that speakers could (and sometimes did) choose a completely different Yiddish lexical item (rather than a Modern Hebrew or English loanword) if they preferred this option.

Results
The results of this task showed a similar overall pattern across all participants, but participants from London displayed the pattern to a smaller degree. Let us look at the pattern of results for speakers from each location.
Israeli participants were highly reluctant to use the original form of the loshn koydesh element (i.e., a masculine singular active participle) in the predicative and even more so in the attributive position. In about half of cases they applied a stem change using their knowledge of Hebrew grammar (e.g., ‫מסדר‬ mesader → ‫מסודר‬ mesuder) in order to derive an adjective (identical to a Hebrew passive participle) from the provided active participle. Thus they selected adjectival forms based on Hebrew passive participles as opposed to active ones (e.g., the pual [passive factitive stem], forms ‫מחונך‬ mekhunekh 'educated,' ‫מבולבל‬ mevulbl 'confused,' ‫מכובד‬ mekhubed 'honorable/honored,' and ‫מסודר‬ mesuder, and the hufal [passive causative stem] form ‫מוצלח‬ mutslekh 'successful'). They produced these forms based on a prompt containing a periphrastic verb in the corresponding active stem, namely the Hebrew piel (factitive stem) forms ‫מחנך‬ ‫זיין‬ mekhanekh zayn 'to educate,' ‫זיין‬ ‫מבלבל‬ mevalbl zayn 'to confuse,' ‫זיין‬ ‫מכבד‬ mekhabed zayn 'to honor,' and ‫זיין‬ ‫מסדר‬ mesader zayn 'to organize,' and the hifil (causative stem) form ‫זיין‬ ‫מצליח‬ matsliyekh zayn 'to succeed.' This shows that their grammar has a strong link between these active and passive loshn koydesh forms. It is not clear the extent to which such a link existed in pre-War and Standard Yiddish as the adjectival forms are relatively rarely attested, but these results are consistent with the link having been introduced or reinforced through familiarity with Modern Hebrew, in which the same active and corresponding passive forms exist and are in regular usage. In about 40% of instances Israeli speakers applied an alternative strategy, such as replacing the original loshn koydesh participle with a Germanic-origin adjective or past participle or chose not to provide an adjective. Thus, overall, the Israeli participants demonstrated an ability to create adjectives derived from the periphrastic verbs given using loshn koydesh adjectivalizing morphology with a high degree of productivity (Table 8). Participants from New York also applied the loshn koydesh stem change adjectivalizing morphology to both predicative and attributive adjectives about 44% of the time. This is an interesting finding, given that it shows the productivity of Hebrew morphology in Yiddish carrier sentences. New York speakers showed somewhat less reluctance to leave the stem unchanged both attributively and predicatively compared to Israeli speakers, although their reluctance was more pronounced in the attributive position. Nevertheless, their overall pattern of response is similar to the Israeli participants, showing an ability to create adjectives derived from the periphrastic verbs given using loshn koydesh adjectivalizing morphology with a high degree of productivity. Participants from London were the most tolerant towards using the stem in its unchanged form, although again this tendency is less pronounced in the attributive position. Nevertheless, they were able to use stem change morphology productively a little over 20% of the time, showing a limited degree of productivity with Hebrew morphology. This difference between London speakers and speakers from New York and Israel can perhaps be attributed to the smaller sample. 18 One participant from New York did not finish this task. Table 8 Proportion of each response type in the adjective task (cont.) As Table 9 illustrates, loshn koydesh adjectivalizing stem change morphology is productively used by all groups except London female participants, with the highest level of productivity apparent in Israeli women. Nevertheless, men from all communities and female participants from New York also show an ability to perform stem change in both the attributive and predicative position. Gender and location differences are most pronounced in the attributive, where speakers are most reluctant to use the stem in its unchanged form. Israeli men and women were most reluctant to do so, followed by men from New York and London. Women from New York and especially women from London showed the highest tolerance for keeping the unchanged stem. This gender-location breakdown of the data is consistent with the original prediction in the sense that the language use correlates with the degree to which participants are exposed to Modern Hebrew (high: Israeli men and women; low: New York and London men and women) and loshn koydesh (high: all men; low: all women). It is nevertheless interesting that the influence of exposure to Modern Hebrew and loshn koydesh did not manifest itself in an increased ability to apply stem changes productively, but rather in an increased reluctance to use the unchanged stem. In such cases, participants typically either produce an equally appropriate (often Germanic) adjective or refuse to provide a response. Altogether, these findings suggest that for some items the stem variants are readily available to the speakers and form part of their productive morphology (e.g., that there is a link between the forms ‫מסדר‬ mesader and ‫מסודר‬ mesuder). For items where speakers are unable to produce the stem change variant, either because they do not know it or because the form pair simply does not exist in Hebrew, the more exposure to loshn koydesh or Modern Hebrew the speaker has, the more reluctant they are to use the unchanged variant. These results suggest that a high degree of exposure to Modern Hebrew or loshn koydesh makes people aware of the inappropriateness of the unchanged stem in an adjectival position, especially in the attributive, and this pushes them to instead use an alternative strategy or give a zero response.
If participants are relatively comfortable with loshn koydesh adjectivalizing morphology, one might wonder to what extent they make use of the Germanic adjectivalizing suffix ‫-דיג‬ -dig. Our results demonstrate that there was a marked difference in form between the predicative and attributive forms of the adjective that participants produced, a finding that was entirely unexpected.19 In the predicative position, participants typically used the masculine singular participle of the periphrastic verbs in question without adding the adjectival suffix ‫-דיג‬ -dig, providing forms such as ‫מסודר‬ mesuder 'organized,' ‫מכובד‬ mekhubed 'honorable/honored,' ‫מחונך‬ mekhunekh 'educated,' ‫מתפעל‬ mispoel 'amazed.' By contrast, when they formed the attributive adjective from the same periphrastic form, they often added the adjectival suffix ‫-דיג‬ -dig, followed by the attributive suffix ‫-ע)ר(‬ -e(r) (see Belk, Kahn, & Szendrői 2020;and Belk, Kahn, & Szendrői under review) for discussion of the attributive suffix in contemporary Hasidic Yiddish). Most unexpectedly, they form the attributive adjective by adding the Germanic adjectivalizing suffix onto a form that predicative positions; see, e.g., the entries for ‫מסוכּן‬ mesukn and ‫מסוכּנדיק‬ mesukndik 'dangerous' and others in Niborski (2012:296). has already undergone a loshn koydesh adjectivalizing stem change. In other words, participants added a suffix that, in other areas of Yiddish grammar, is only used to derive adjectives from non-adjectival categories to a form that they recognize and produce as a predicative adjective. This highly productive strategy resulted in attributive forms such as ‫מסודרדיגע‬ mesuderdige 'organized,' ‫מכובדיגע‬ mekhubedige 'honorable,' ‫מחונכדיגע‬ mekhunekhdige 'educated,' and ‫מתפעלדיגע‬ mispoeldige 'amazed.' The pattern is illustrated in Table 10. This difference is noteworthy because it indicates that the loshn koydesh participial forms that can function independently in the predicative position must nevertheless undergo further derivation with the addition of the adjectivalizing suffix ‫-דיג‬ -dig for the attributive marking inflection to apply. This contrasts with other Yiddish adjectives derived from non-adjectival categories, including those derived from loshn koydesh nouns such as ‫חנעוודיג‬ khyenevdig 'charming' or ‫שבתדיג‬ shabesdig 'pertaining to or suitable for the Sabbath,' which take the adjectival suffix ‫-דיג‬ -dig in both predicative and attributive contexts. Indeed, in the two control items we included that involved loshn koydesh nouns, ‫-דיג‬ -dig was applied in all the derived adjectival forms. Assouline (2017:148-150) has documented a similar morphological difference between predicative and attributive participial adjectives borrowed from Modern Hebrew (e.g., metuskal 'frustrated') in the Yiddish of Hasidic speakers in Israel. Her findings are consistent with an analysis where borrowed adjectives are marked as lexically distinct and therefore not morphologically integrated into Yiddish grammar in the predicative position. By contrast, our findings are not based on borrowed adjectives, but rather on adjectives resulting from an integrative and productive Yiddish periphrastic model. As such, we analyse our results as deriving from an incompatibility between Semitic participial morphology and the Germanic attributive suffix, which is overcome by means of the adjectivalizing suffix ‫-דיג‬ -dig.
As expected, participants applied the attributive marker ‫-ע)ר(‬ -e(r) to the adjectives overwhelmingly in the attributive position and never in the predicative position. This confirms our earlier findings that this suffix is used as an attributive marker rather than a case or gender marker. As Table 11 shows, this pattern holds of all participants and there were no location or gender differences in this regard. This pattern also suggests that the adjectives were highly integrated into the Yiddish sentence and not used as direct borrowings. The few instances where the attributive marker is not applied in attributive contexts typically lack the ‫-דיג‬ -dig suffix and instead make use of Hebrew gender agreement morphology; e.g., one speaker provided ‫מסכימה‬ maskima (lit: 'agreeing;' intended: 'agreeable,' fem.). This suggests that these specific forms are indeed direct borrowings from Modern Hebrew or loshn koydesh and are not evidence of Yiddish grammatical proficiency. In fact, this pattern appears to be restricted to a very small number of lexical items, e.g., ‫מקיים‬ mekayem (lit. 'fulfillable;' intended: 'fulfilled,' masc.).20 Table 11 Proportion of attributive marking in the adjective task for Israeli, New York, and London participants Additionally, we propose that the fact that these attributive adjectives, but not their predicative counterparts, appear to require the adjectivalizing suffix ‫-דיג‬ -dig is not due to the fact that speakers do not recognize the predicative versions as Yiddish adjectives, but rather due to a grammatical incompatibility between the Germanic attributive marker ‫-ע)ר(‬ -e(r) and a Semitic morphological stem. The attributive position requires the presence of the attributive 20 We believe that at least some instances of this pattern are due to a misanalysis of the prompt as requiring a verb rather than an adjective. marker ‫-ע)ר(‬ -e(r), and this itself requires the presence of ‫-דיג‬ -dig to provide a non-Semitic morphological stem. In other words, the application of ‫-דיג‬ -dig is a tactic used to derive a Germanic stem rather than performing its usual adjectivalizing function. The incompatibility between ‫-ע)ר(‬ -e(r) and a Semitic morphological stem appears to be a new development in contemporary Hasidic Yiddish: the proposed requirement that ‫-ע)ר(‬ -e(r) not attach to a Semitic morphological stem does not appear to have existed in pre-War Yiddish, in which ‫-ע)ר(‬ -e(r) could attach freely to Semitic participials and conversely ‫-דיג‬ -dig could appear in the predicative position, where it was not needed in order to provide a stem for an attributive marker (see footnote 19).
Interim Summary All participants except for London female participants made use of Hebrew derivational patterns in deriving adjectives from periphrastic verbs and applied these productively. At the same time, the more exposure a speaker had to Modern Hebrew or loshn koydesh, the more reluctant they were to leave the verbal stem unchanged when deriving an adjective if they were unable to supply the stem change variant, especially in the attributive position. Thus overall, our findings confirm our prediction that being exposed to Modern Hebrew or loshn koydesh makes a difference in how participants form predicative and attributive adjectives from periphrastics. Speakers in all communities recognize Hebrew participial forms as adjectives, as indicated by the fact that they do not apply the adjectivalizing suffix in predicative contexts, but nonetheless add it in attributive contexts (in addition to the standard attributive marker ‫-ע)ר(‬ -e(r)). This finding is particularly surprising, as we are not aware of any other cases where the ‫-דיג‬ -dig suffix is applied to a stem that is already adjectival, nor to any cases of attributive adjectives requiring further derivational morphology compared to their predicative counterparts. We take this pattern to indicate that although Hebrew participial forms are treated as adjectives, they are not fully integrated into Yiddish grammar (cf. Assouline 2017:148-150). In order for Yiddish attributive morphology to be applicable to such adjectival stems, a Yiddish derivational morpheme, the ‫-דיג‬ -dig suffix, must also be present. This reveals an intricate pattern of language mixing, which applied to all of our speakers, irrespective of their gender and geographical origin. It also reflects an innovation in the grammar of contemporary Hasidic Yiddish which affects all geographical communities of speakers.

Discussion
The present study showed us that the loshn koydesh component continues to be a substantial, integrated, and productive element of 21st-century Hasidic Yiddish. We found that participants were familiar with the nouns, periphrastic verbs, and derived adjectives that we provided in the questionnaires. This was true of all the speakers regardless of their geographical origin or their gender. They overwhelmingly accepted loshn koydesh periphrastics as grammatical, and they were mostly able to apply Hebrew morphology when a stem change was required to create an adjectival form based on loshn koydesh periphrastics.
The present study suggests that the loshn koydesh component in the language of Yiddish-speaking Hasidim exhibits a number of developments when compared to the pre-War and Standard varieties of Yiddish. Some of these effects affected all the speakers universally, such as the loss of grammatical gender. We saw that just like with Germanic-origin words (see Belk, Kahn, & Szendrői 2020;and Belk, Kahn, & Szendrői under review), contemporary Hasidic Yiddish speakers do not assign grammatical gender to loshn koydesh nouns. Rather, they overwhelmingly use a default form of the article, with two additional tendencies optionally overriding the default. These tendencies are that words denoting male humans sometimes receive the article ‫דער‬ der in writing and that words denoting important or significant entities or concepts also sometimes receive ‫דער‬ der. The former seems to be a relic carried over from historical dialects, a fossilized form of an earlier grammatical rule. The latter, however, seems to be a truly innovative feature of contemporary Hasidic Yiddish language use.
Another development that seems to have affected Hasidic Yiddish speakers in all communities studied is the finding that adjectival forms derived from loshn koydesh participles are unacceptable in the attributive position unless a Yiddish derivational affix ‫-דיג‬ -dig and an attributive marker are applied. This is true despite the fact that the stems to which ‫דיג‬ -dig is applied are produced as predicative adjectives -a surprising result given that in other domains ‫-דיג‬ -dig is applied only to non-adjectival categories. Furthermore, speakers in all communities have reinterpreted historical case and gender morphology as attributive marking, requiring such marking on attributive adjectives and disallowing it on predicative adjectives. This result supports earlier work on this issue (Belk, Kahn, & Szendrői 2020;Belk, Kahn, & Szendrői under review). Given that all participants pattern together in a way that is different from the pre-War and Standard varieties of Yiddish, these findings suggest that a development has taken place in the language that has been widespread enough to affect all Hasidic Yiddish speakers.
We also found developments that affected the various speech communities differently. These effects can be teased apart by examining the language of Yiddish-speaking men and women in Israel (with women being exposed to more Modern Hebrew than men), of Yiddish-speaking men and women in New York and London (with women being exposed to more English than men), and of men in Israel, New York, and London in comparison to women in those communities (with men being exposed to more loshn koydesh than women and mixing more with Yiddish speakers in other countries than their female counterparts, leading to a convergence effect). One aspect in which speakers from Israel and New York on the one hand differed from speakers in London was which article was selected as the default form. This appears to be phonologically conditioned: London is different from the other communities due to the non-rhotic nature of London English.21 Another diverging development concerned the effects of English and Modern Hebrew on acceptability of alternatives to periphrastic verb forms. New York speakers, especially women, had a much higher acceptance of Yiddishized English verbs, but not direct English borrowings. At the same time, Israeli women, but not men, had a high acceptance of direct borrowings of Modern Hebrew infinitivals. In contrast, men, especially Israeli men, had a high acceptance of innovative Modern Hebrew-based periphrastic forms, indicating a high level of grammatical productivity of these forms.
Finally, we also found that the level of exposure to Modern Hebrew or loshn koydesh seems to correlate with a speaker's reluctance to produce the original unchanged form of loshn koydesh participles as adjectives, especially in the attributive position, with Israeli men and women having the highest degree of reluctance, and London female speakers the lowest. We propose that this is attributable to familiarity with the customary Hebrew stem variations and thus ultimately can be tied to level of exposure. In sum, we found that the loshn koydesh component in contemporary Hasidic Yiddish language use has been affected by a number of grammatical changes. Some of these changes affected all the speakers, while some reflect the sociolinguistic landscape of the speakers. 21 This is in fact a pervasive change that affected not only the loshn koydesh component but all nouns.

Conclusions
We hope that this study has provided readers insight into the contemporary language practices surrounding the loshn koydesh component of Yiddish as spoken by Hasidim in Israel, New York, and London. Our findings are largely consistent with those of Assouline (2010Assouline ( , 2017Assouline ( , 2019, although our study covers speakers in a wider range of geographical communities and investigates the issue of speaker gender on language practices. Our use of negative and metalinguistic data provides further insight into these questions. The results of our investigation into morphological gender of loshn koydesh-component nouns in Yiddish represent novel findings, although they are consistent with Belk, Kahn, & Szendrői 2020 and Belk, Kahn, & Szendrői under review. We also found that speaker gender plays a significant role in the use of periphrastic verbs and their synonyms as well as patterns of adjectival derivation from loshn koydesh. Our interpretation of our findings, especially as they relate to adjectives derived from loshn koydesh, provides a different perspective than previous literature. The methodology employed for this research (a three-part questionnaire with two open questions and one question consisting of multiple ranked options, accessible to speakers who spoke either English or Modern Hebrew as an additional language) had certain advantages and disadvantages. As the results show, the questionnaire provided us with comprehensive information on the exact issues that we wanted to investigate in a way which would be difficult to achieve via a corpus of spontaneous speech. Moreover, giving the same questionnaire to speakers from different geographical backgrounds, age ranges, Hasidic affiliations, and genders allowed us to make a clear comparison across all of these axes. It also facilitated a statistical analysis of specific sociolinguistic parameters that might have been more challenging to analyse meaningfully without such a controlled dataset. For example, we were able to investigate the possibility of grammatical differences by Hasidic affiliation and to confirm that this axis did not yield any meaningful results. Conversely, it must be acknowledged that a questionnaire composed of artificial questions is not the same as a spoken or written corpus, so the latter type of research is also desirable as a complement to the present study. The use of a questionnaire meant that some of our tasks contained items which were designed to push the boundaries of grammatical acceptability and thus may have resulted in speakers producing forms which they would not employ in spontaneous speech or writing. However, the inclusion of these items in the questionnaire was necessary in order to test the limits of what speakers judged to be permissible, and the data gleaned from this aspect of the questionnaire is therefore useful even though it could be complemented in future research by corpus analysis along the lines of Assouline 2019. Additionally, the open-ended nature of two of the tasks meant that speakers did not always produce the intended forms, leading to a dataset that was not as complete as one produced by, for example, a forced-choice task. This methodology allows for a different, complementary perspective on the phenomena in question than studies focusing on corpus data.
While it is hoped that our study has helped to provide an overview of the role of loshn koydesh among contemporary Hasidic Yiddish speakers internationally, further research needs to be conducted into various aspects of this component of the language. We concentrated in this article on geographical origin, contact languages, and speaker gender, but we also examined other factors, such as a speaker's Hasidic dynasty. With respect to the latter variable, we did not find any emergent patterns. This could be because there are no such patterns, or that some aspect of our data collection method was inadequate to show it. First, given that there are dozens of Hasidic dynasties, a more extensive dataset might provide further insight into this issue. Moreover, we only included speakers who acquired Yiddish in infancy or early childhood or who went to Yiddish-medium educational institutions. This means that speakers from certain Hasidic dynasties would not be represented in our sample. Additionally, our sociolinguistic discussions with speakers suggest that there is a relatively high degree of contact between speakers from different Hasidic dynasties (with some exceptions, where particular dynasties are more closed to outsiders). For example, many Hasidic families (especially those from smaller dynasties or smaller communities) consider factors other than Hasidic dynasty when choosing schools for their children, such as language of education, educational reputation, or geographic factors. Thus, many children are educated outside of their Hasidic dynasty. Overall, the question of the effect of Hasidic dynasty on the loshn koydesh component of Yiddish is still very much an open one. A second topic to investigate in the future is the distribution of the use of ‫דער‬ der in writing, in light of metalinguistic data suggesting that speakers tend to select it in conjunction with loshn koydesh (and other) nouns which they perceive as semantically important. A third topic is the contemporary Hasidic use of loshn koydesh-derived verbs with an infinitive in ‫-ן‬ -n, such as ‫פטרן‬ patern 'to exempt,' ‫חתמנען‬ khasmenen 'to sign,' etc., and in particular how productive such a pattern is in contemporary Hasidic Yiddish in comparison with periphrastics. Finally, as mentioned above, corpus analysis would enable us to gain a different perspective on the use of Hebrew-origin elements in Hasidic Yiddish from the one presented in this article. Such future research will help to build up a more comprehensive picture of the loshn koydesh component of the language.
The results presented in this article show clearly that the grammatical properties associated with the Hebrew-origin component of Yiddish (with respect to morphological gender of loshn koydesh nouns, replacement of loshn koydesh periphrastic verbs with Germanic, Modern Hebrew, and English equivalents, and morphosyntax of deverbal loshn koydesh adjectives) differ in contemporary Hasidic Yiddish from the pre-War and Standard varieties of the language. These differences support our previous suggestion (Belk, Kahn, & Szendrői 2020) that contemporary Hasidic Yiddish should be considered an independent variety of Yiddish distinct from any of the pre-War dialects of the language. While today's Hasidic Yiddish speakers are of course in large measure descendants of speakers of the historical Eastern European varieties of Yiddish, their language has evolved to such an extent that it cannot be classified within this traditional framework. This new variety of the language can be called Contemporary Hasidic Yiddish.
The results of the present study also support our broader argument that Contemporary Hasidic Yiddish represents not only a distinct variety from pre-War and Standard Yiddish, but also a relatively cohesive one. There are differences between speakers with respect to the loshn koydesh component along sociolinguistic (i.e., gender-and contact-based) dimensions. However, these appear to be quantitative rather than categorical differences, and they predominantly affect the lexis, in contrast with the grammatical differences between historical Mideastern, Southeastern, and Northeastern Yiddish as spoken by contemporary speakers' Eastern European ancestors. As such, we posit that it would be beneficial to move away from attempts to classify present-day Hasidic Yiddish within the pre-existing framework of older dialects and instead regard it as a newly evolved form of the language with its own grammatical conventions, including but not limited to its relationship with the loshn koydesh component.