Malin 2025

AAWM Journal 13/No. 1 (2025)

Timing in Klezmer Performance

Yonatan Malin

klezmer, performance practice, timing, non-isochronous meters, transcription, world music pedagogy

This study explores aspects of timing in the performance of Jewish instrumental music from Eastern Europe, known as klezmer. Performance style is an essential aspect of the tradition; musicians today study historical recordings and teach the style in lessons and workshops. The paper thus dovetails with concerns of contemporary musicians who look back to historical sources as they continue to innovate.
Three common timing features are documented: (1) variable timing within and between performances, (2) non-isochronous (unequal) beats in a genre known as the zhok or hora, and (3) a distinctive move-ahead-and-wait or accelerate-and-delay gesture that is a signature of the style more broadly and an inverse of a common timing profile in jazz (see Ashley 2002). I analyze historical recordings of a tune known as “Gasn Nign” by Jacob Hoffman with Kandel’s Orchestra (1923), Abe and Sylvia Schwartz (1920), and the Naftule Brandwein (1923). I then analyze a more recent recording of “Gasn Nign” by Alicia Svigals with Steven Greenman (1996) and a piece called “I Flow with Water Under the Ice” composed and performed by Cookie Segelstein with Joshua Horowitz (2019). Methodologies include close listening, transcription, score annotation, timing measurements using Sonic Visualiser, and customized forms of data visualization. Pedagogical materials, discussions with leading musicians, and information from workshops and lessons show how contemporary musicians understand and teach the timing features documented here.
This study contributes to a growing field of timing studies in world music traditions, including the Southeastern European traditions that are closely aligned with klezmer. It contributes to the literature on aksak meters by documenting a long-short metric cycle with beats in a 3:2 ratio and by investigating the timing of rhythms that work with this cycle. It builds on London’s many meters hypothesis (London 2012), which suggests that metric and rhythmic cognition are deeply enculturated. This study further demonstrates how variable timing merges seamlessly into variable rhythm in oral traditions. Timing measurements are then valuable alongside transcription and annotation for the representation and study of performance practice.

Yonatan Malin is Associate Professor in the College of Music at the University of Colorado Boulder.

---

Click for DOI, citation, PDF version.

---

 

Introduction

 

This study is about timing in the performance of Jewish instrumental music from Eastern Europe, known as klezmer. Performance style is an essential aspect of the tradition; musicians today study historical recordings and teach the style in lessons and workshops.[1] The article thus dovetails with concerns of contemporary musicians who look back to historical sources as they continue to innovate.

[2] Klezmer performance is known for its rapid and sometimes virtuosic embellishments, which are linked on the one hand to other Eastern and Southeastern European styles, and on the other hand to Ashkenazi (Eastern European) Jewish vocal performance (Slobin 2000, 99 and 109; Rubin 2020, 206). But while Joel Rubin has provided a rigorous and detailed study of klezmer embellishments (Rubin 2020, chap. 6), there hasn’t been an equivalent study of timing.[2]

[3] Klezmer musicians use expressive timing along with embellishments to shape phrases and sections, to convey emotion, to tailor the music for dance, and to maintain creative continuity with Jewish culture in Eastern Europe before World War II. In this paper, I identify and explore three aspects of klezmer timing: (1) variable timing within and between performances, (2) non-isochronous beats in a genre known as the zhok or hora, and (3) a distinctive move-ahead-and-wait or accelerate-and-delay gesture that is a signature of the style more broadly. I also take note of asynchronies between the melody and harmonic/rhythmic accompaniment, with the melody typically ahead. Timing variability is especially evident in slower dance tunes as Rubin (2020, 225­–26) has observed, and that is the focus here. I set aside klezmer genres without a regular beat such as the doina and kale bazetsn; these will require separate study.

[4] My work is similar in some ways to the study of expressive timing in Western art music, which up until recently has focused on late eighteenth and early nineteenth-century piano music.[3] But whereas in Western art music, we have scores to work from, the same cannot be said for most klezmer recordings. Historically, tunes were learned orally in family lineages, in apprenticeships, and on the job at weddings and other events.[4] In the contemporary context, tunes may be learned with charts, but they are also learned orally in workshops, lessons, or jam sessions, or from recordings. Scores, when they exist, are more akin to lead sheets, which the performer may depart from and will certainly embellish. Thus, for the most part, we cannot study timing in klezmer vis-à-vis the “ground truth” of notated scores.

[5] This study contributes to a growing field of timing studies in world music traditions including Southeastern European traditions that are closely aligned with klezmer. The klezmer genre known as the zhok or hora, which I focus on here, comes from and was developed together with the stately Moldavian dances joc and hora (Feldman 2022, 147–48). It belongs to a repertoire that Zev Walter Feldman has identified as “transitional,” i.e., genres that “were adopted into Jewish musical and dance culture and were played predominantly (although not always exclusively) for Jews” (Feldman 2016, 211–12; see also Feldman 2002, 93–94). In more recent studies Feldman observes that this transitional repertoire resulted from “the cultural and professional symbiosis” of Jewish klezmer and Roma lăutar musicians (Feldman 2022, 148), and that there was a corresponding Moldavian repertoire of Jewish origin that was performed for Moldavian Christians (Feldman 2020). The direct comparison of these repertoires lies outside the scope of the present paper, but the methodologies and observations offered here may serve as the basis for future comparative work. It should also be noted that the tune studied below has the rhythm of the Moldavian joc, but melodic elements are characteristic of “core” Jewish repertoires (see Feldman 2016, 209–11).[5] The tune is known as “Gasn Nign” (Street Tune), and it would have been used for procession rather than dance. As Beregovski observes, “Klezmorim played the zhok as a street tune, less often as a dance tune, at first doubtless adopting it from the Moldavan or Romanian folk repertoire” (Beregovski 2015, I-11; see also Stutschewsky 1959, 216).

[6] This study contributes more specifically to the study of non-isochronous or aksak meters in Southeastern European traditions, building on prior studies (During 1997; Moelants 2006; London 2012; Bonini-Baraldi et al. 2015; Cler 2015; Goldberg 2015 and 2019; Leong 2020 and others). I contribute to this literature by documenting beat ratios in the klezmer zhok and by exploring the variable timing of rhythmic figures that go with the non-isochronous meter. These variable rhythms in turn suggest that the unequal beats are not dependent on a regular (isochronous) underlying pulse. Bonini-Baraldi et al. similarly observe that “the Transylvanian slow aksak is not ‘bodily’ decomposed in smaller pulses” (Bonini-Baraldi et al. 2015, 275). Cler suggests that there are two kinds of aksak, “one with underlying explication of the isochronous unit, the other one without, depending on tempo” (Cler 2015, 303). And During finds that beats in Baluch and Tajik-Uzbek traditions slide between different ratios, so lower-level divisions “lose all relevance” (During 1997, abstract).[6]

[7] More broadly, this paper builds on Justin London’s many meters hypothesis, which indicates that metric and rhythmic cognition are deeply enculturated. As London puts it, “A listener’s metric competence resides in her or his knowledge of a very large number of context-specific metrical timing patterns. The number and degree of individuation among these patterns increases with age, training, and degree of musical enculturation” (London 2012, 182). In the present context, we can say that listeners, performers, and dancers who are immersed in klezmer music over a period of time become attuned not only to its melodies and rhythms, but also to its varied and distinctive timings. Relations between musical exposure and metric entrainment have been explored and tested in Jakubowski et al. (2022), Polak et al. (2018), and Danielsen et al. (2022).[7]

[8] In the next part of the paper, I explore three historical recordings of the tune known as “Gasn Nign.”[8] All three recordings were made in New York in the 1920s. The term gas-nign (generally spelled as such) also refers more broadly to a genre of tunes used to accompany wedding parties through the streets in community-wide celebrations often spanning several days (Slobin 2000, 123).[9] The first historical recording, by Jacob Hoffman on xylophone with Kandel’s Orchestra (1923), will give us a baseline from which to work. The second recording, by Abe Schwartz on violin with his daughter Sylvia Schwartz on piano (1920a), will introduce the three characteristic timing elements: variability, the aksak meter with unequal beats, and the accelerate-and-delay profile. The third recording, by the clarinetist Naftule Brandwein with the Naftule Brandwein Orchestra (1923), demonstrates yet one more timing option from a historical source. In part III, I discuss a more recent recording of the Gasn Nign by the acclaimed klezmer violinist Alicia Svigals (1996); this will demonstrate the continuing vitality of the tradition with characteristic timing profiles and new levels of variability. In part IV, I move beyond the Gasn Nign; I test the methodologies and concepts with an original tune composed and performed by the violinist Cookie Segelstein with Joshua Horowitz on tsimbl (track 29 on The Magid Chronicles; Veretski Pass with Joel Rubin 2019). I conclude in part V with reflections on the study and directions for further research.

 

Three Historical Recordings

 

[9] Harry Kandel, leader of Kandel’s Orchestra, was born in either Kraków (Netsky 2015, 40) or Lviv (Sapoznik 1999, 94–95) and classically trained at the Odessa Conservatory.[10] He emigrated to New York in 1904 and worked as an assistant for John Philip Sousa and as a conductor in the Yiddish Theatre in Philadelphia (Netsky 2015, 40). Jacob Hoffman, the xylophone player, belonged to a klezmer family from Kriovozer, a Jewish town in Ukraine.[11] Hoffman performed for the Keith Vaudeville Circuit and with the Philadelphia Orchestra (Netsky 2002, 55–56; 2015, 43).

[10] Hoffman’s recording of the Gasn Nign with Kandel’s orchestra is straight, without much variation in timing. The tune is in two sections; Example 1 provides a transcription of the first section along with an excerpt from the recording (Audio Example 1). The transcription focuses on the main melodic pitches, leaving out xylophone embellishments.[12] Brackets in Example 1 show a characteristic rhythmic figure for the zhok, the dotted eighth with three sixteenths. (The figure with triplet sixteenths in bars 12 and 14 is a common variant.) On the recording, we also hear the common backing rhythm with events on beats 1 and 3 of the triple meter.[13]

Example 1. Jacob Hoffman with Kandel’s Orchestra, Gasn Nign (1923).

Audio Example 1. Hoffman, Gasn Nign section A.

 

[11] Abe Schwartz recorded this same tune in 1920 with his daughter Sylvia on piano. Abe was born in 1881 in Romania, near Bucharest, emigrated to the United States around 1900, and began recording for Columbia records in New York in 1917 (Sapoznik 1999, 90–91).[14] Example 2 provides a transcription of the A section from the Schwartz recording (melody only) with an excerpt from the recording (Audio Example 2a). (I will return to Audio Example 2b below.) Overall, this recording has more lilt to it, with pitch bends, anticipations, and embellishments. There are also new rhythmic ideas. Hoffman and Kandel’s orchestra begin with a rhythm of three eighths (Example 1 above); Abe Schwartz transforms this into something more etched, with a stronger impulse: a <sixteenth, dotted eighth, and eighth> rhythm. Schwartz continues that rhythm in bars 2–3 and adapts it for the rising and falling scalar figure in bars 10–12.

Example 2. Abe and Sylvia Schwartz, Gasn Nign (1920a), section A.

Audio Example 2a. Abe and Sylvia Schwartz, Gasn Nign section A.

Audio Example 2b. Abe and Sylvia Schwartz, Gasn Nign bars 1­–9 at 80% of the original speed.

 

[12] These changes point broadly to rhythmic variability, which Mark Slobin (2000, 127–29) has also noted in his study of this tune. A given tune may be played with distinct rhythms, which can come from an individual performer’s style, from how and where they learned the tune, and from expressive decisions in the moment.[15] The variability also may be greater in solo or small ensemble settings, as here, than in larger ensembles, as in the recording by Hoffman with Kandel’s Orchestra.

[13] But there are also aspects of timing in the Schwartz recording that are characteristic of klezmer performance more broadly. There is a tendency to move ahead and wait, to accelerate and then delay. The transformation of equal eights (<eighth, eighth, eighth>) into the more etched rhythm (<sixteenth, dotted eighth, eighth>) is one example of this; the first duration is shorter and then there is time. Breath marks in the transcription just before bars 7 and 9 show further examples; Schwartz rushes ahead and then waits for the downbeats. Audio Example 2b above slows the recording to 80% so that readers can focus on the timing nuances.

[14] Acceleration and delay leading to a downbeat is a broad stylistic feature of klezmer performance.[16] The accordionist and composer Joshua Horowitz described this to me in a lesson, “In general, we can make a rule, that before any long note, the long note just in relation to the previous notes, we are going to leave space there, and kind of do the one before it a little sooner” (private lesson, October 16, 2018). Joel Rubin observes that one characteristic timing profile in the early 20^th century recordings of Naftule Brandwein and Dave Tarras involves “a slight rushing of the melodic line toward the next bar, followed by a slight pause, so that the downbeat of both melody and accompaniment coincide” (2020, 224). And in a pedagogical video, the klezmer fiddler Alicia Svigals notes that the hora or zhok “should be played with rushed sixteenth notes at the end of the measure,” and a “little silence after the sixteenth notes” (Svigals 2013a).

[15] Abe and Sylvia Schwartz’s recording of the Gasn Nign also features a non-isochronous beat. The basic accompaniment rhythm, which would typically be notated on beats 1 and 3 of a 3/8 meter (Example 3), does not in fact align with a pure triple division. The third beat comes early, as it were. This can be experienced by tapping in three with the recording (Example 2a or 2b above) and feeling the extra time for the third beat in each measure.

Example 3. Accompaniment rhythm notated in 3/8.

[16] I measured the inter-onset durations for the basic piano rhythm throughout this track with the software Sonic Visualiser. [17] The track has ninety-nine bars, and the average ratio was 3:2. In other words, the second event occurs 60% of the way through the bar, on average. The timing is relatively consistent: the standard deviation is 3%. And this suggests an aksak rhythm in 5/8, shown in Example 4. This 3:2 aksak ratio is characteristic of some klezmer performances of this particular genre, but not all. It is no doubt significant that Abe Schwartz grew up near Bucharest, in Romania, where the co-territorial music included these kinds of rhythms.

Example 4. Accompaniment rhythm notated in 5/8.

[17] A thread in the research on non-isochronous meters has asked whether the ratios vary with tempo changes or phrasing (Polak and London 2014 and 2022; Goldberg 2015). In Abe and Sylvia Schwartz’s Gasn Nign, the tempo is relatively consistent. The average duration of the bar is 1.22 seconds with a standard deviation of 0.06. In the larger ABA form of this performance, the average for measure durations for each section were 1.25, 1.21, and 1.21 seconds. There is some tempo variability at the phrase level: for example, there is a slight tendency to shorten the bars just before the harmonic change at bar 7 of the A section (the downbeat with the new harmony comes early). These are small differences, however, which are barely perceptible. And given the uncertainty of the measurements with the historical recording, I did not attempt to correlate the beat ratios with tempo or phrasing.

[18] The notion of a five-part division for the zhok has been in circulation among klezmer musicians in recent decades. Mark Slobin reported that the clarinetist Kurt Bjorling and the band Brave Old World taught this tune in workshops in the late 1990s, and that Bjorling thinks of the rhythm “as forming a five-beat unit” (Slobin 2000, 123). Segelstein and Horowitz (2005) provide “beat placement variations” for the hora with notation in 3/4, 5/8, and 7/16. But the guitarist Jeff Warshauer described the rhythm in a workshop at Yiddish New York (2022) more broadly as moving towards a duple division with flexible placement of the second beat. And Joel Rubin describes the zhok as being “in an aksak meter that approximates 3/8 with an accentuation on beats 1 and 3” (2020, 353n25). The overall metric feel might thus remain one of <long, short, long, short> without subdivisions, and with an upbeat that is slightly longer than it would be in a triple meter.

[19] We can extend the study to consider the interaction of Abe Schwarz’s violin melody with the long-short aksak meter. I took inter-onset measurements of the basic rhythms shown in Example 5.[18] For the second rhythm (Example 5b), I measure the two sixteenths on beat three as a single duration to match the first rhythm (Example 5a). There are forty-three instances of these rhythms in the track, enough to get some initial data. The median timing ratio for the three events is 18% / 41% / 40% of the full measure duration. So, the first event tends to be slightly longer than a sixteenth (which, rounded to the nearest full percentage point, would be 17% of the measure). The second and third events tend to be about equal. Example 6 provides an alternative transcription of section A in 5/8 along with the recording excerpt (Audio Example 3).

Example 5. Recurring rhythms in Abe Schwartz’s performance (1920a).

Example 6. Abe and Sylvia Schwartz, Gasn Nign (1920a), section A transcribed in 5/8.

Audio Example 3. Abe and Sylvia Schwartz, Gasn Nign section A.

[20] Example 6 brings us closer, experientially, to some aspects of Schwartz’s recording, and it may help performers get a feel for the timing. But I do not advocate for the 5/8 notation as a “better” or “more correct” transcription. Tunes in this genre are typically notated in 3/8, and making use of notation that is emic to a community is of value. Also, the variability in Abe Schwartz’s performance goes against the notion of any one correct form of notation.

[21] Example 7 illustrates this variability; it graphs event onsets of the basic rhythms (Example 5 above) on a timeline that represents a single measure duration. X’s show the median location for the second and third events, relative to the full duration of the measure; dots immediately surrounding each X show one quartile of events on either side; and the dots further out show the min and max. Thus, while the median location for the second event is at 18% of the full measure duration, it varies from 11% to 25%. The median location for the third event is at 60% of the measure, but it varies between 47% and 66%.

Example 7. Variability of timing in basic three-note rhythms.

[22] This variability is especially apparent in the B section; see Example 8 and Audio Example 4 below. Ossia measures indicate variations in the repeat. I have labeled two segments as “impulsive” and one as “evenish”. Median timing ratios for the “impulsive” measures are 15% / 47% / 37%, which is close to the ratios implied by the notation in 3/8.[19] Median ratios for “evenish” measures are 20% / 39% / 41%, which is very close to the ratios of the 5/8 notation that we looked at earlier.[20] Listening, you may notice that Sylvia Schwartz doubles her father’s melodic line on the evenish measures, this is part of the effect. In other words, timing variations are used for expressive effect within the larger phrase and duo arrangement. Measure 10 in the repeat includes a distinctive rhythm that divides the measure in 4.

Example 8. Abe and Sylvia Schwartz, Gasn Nign (1920a), section B.

Audio Example 4. Abe and Sylvia Schwartz, Gasn Nign section B.

[23] We can also represent the timing variability with the rhythmic notation in Example 9. Rhythm a has even eighths, as in bars 1 and 5 of the recording by Hoffman with Kandel’s orchestra (Example 1 above). Rhythms b and c correspond with the “evenish” and “impulsive” rhythms in Example 8. Rhythm d moves the third event even earlier, it is “extra impulsive,” as it were. Timings that approximate this rhythm can also be heard in section A of Abe Schwartz’s performance (see Example 2 and Audio Example 2a above). Bar two the first time divides the measure 15% / 33% / 52%, bar one the second time divides the measure 16% / 37% / 47%, and bar two the second time divides the measure 14% / 33% / 53%. Listening to these measures again, one may experience the third event as an anticipation, coming in early in relation to the typical placement of the third beat. Rhythm e is “in four,” and it occurs in the repeat of section B (Example 8, bar 10 in the repeat).[21] Rhythms d and e both divide the measure in half; the vertical alignment in Example 9 reinforces this affinity.

Example 9. Abe and Sylvia Schwartz, Gasn Nign (1920a), three-note timings in rhythmic notation.

[24] These kinds of rhythms are part of klezmer pedagogy today. A set of “Rocky Rhythm Exercises” from Cookie Segelstein and Joshua Horowitz (2005) includes an exercise turning the zhok’s dotted eighth and three sixteenth rhythm into the “impulsive 3/8” rhythm (as I have labeled it in Example 9). The exercises also include duple/quadruple divisions of the 3/8 bar and anticipations and delays. Segelstein and Horowitz illustrate the variable rhythms with the tune “Di Groyse Hora” (Budowitz 2007, track 14).[22]

[25] To review, Example 9 uses rhythmic notation to capture the timing variability in the Hoffman and Schwartz recordings. The rhythmic notation complements more precise timing measurements, it is more musical (i.e., easier to perform or audiate), it suggests a set of timing categories that correspond with these rhythms, and it demonstrates the high degree of timing variability.

[26] Example 9, however, does not capture all the timing variability for this tune in the historical recordings. The clarinetist Naftule Brandwein recorded the A section of the tune on its own, with a repeat, in the middle of a longer track called “Doyne, ershter teyl” (1923); see Audio Example 5. Brandwein plays the tune at a slow tempo: the average duration of the bar is 1.64 seconds compared to 1.22 seconds in the recording by Abe and Sylvia Schwartz and 1.28 seconds in the recording by Hoffman with Kandel’s Orchestra. The second event of the orchestra’s <1 – 3> rhythm comes in at 62% of the measure on average, close to the 60% of Sylvia Schwartz’s piano accompaniment. But the relative durations of the three events in the recurring melodic rhythm are distinctive. Brandwein’s first event is relaxed, it gestures towards an even division of the bar. The median ratios for three events are 24% / 37% / 39%. (The sample size is small: there are ten instances of the rhythm under consideration.) This is different from the median in Abe Schwartz’s performance, which was 18% / 41% / 40%. It is also different from the ratios for the rhythms I identified as “impulsive” or “extra impulsive” or “in four” in Schwartz’s performance. Brandwein’s performance presents its own timing category, at least within the sample studied thus far, and points further to the diversity of timing profiles in klezmer performance.[23]

Audio Example 5. (Naftule Brandwein, Gasn Nign).

 

Recent Recording: Alicia Svigals

 

[27] I turn now to a more recent recording, produced in the context of the klezmer revival—a movement since the late 1970s that has revitalized the tradition with studies of historical recordings and new creative endeavors. We will find elements of continuity between the historical recording and the more recent one—continuity which is important to a culture that was almost entirely decimated in the Holocaust and in the Soviet Union under Stalin. We will also find discontinuity, elements that are not sustained, and innovation, elements that are added or developed—which itself signifies the vitality of the tradition.

[28] The recording is by the internationally acclaimed klezmer fiddler Alicia Svigals with Steven Greenman playing accompaniment harmony and rhythm, also on violin. It was released in 1996 on the compilation album Klezmer Music: A Marriage of Heaven and Earth. Among her many accolades, Svigals was a founder and co-leader of the Grammy winning band The Klezmatics.[24] Svigals herself draws attention to many of the timing features that I will discuss in a pedagogical video, produced by Stradmagazine (Svigals 2013a). Here we will see exactly how the features emerge in her recording of the Gasn Nign.[25]

[29] I use transcription, annotation, and verbal description for this analysis, setting aside the timing measurements. One could measure timing profiles here as well, but in this case, I found that music notation and annotation on their own effectively represent the timing variability. Example 10 provides a transcription of the A section, first time around, with an excerpt from the recording (Audio Example 6). The timing of the sixteenths is generally short-long, as shown in the ossia measure at the beginning. Short-long is the default timing here, so I marked instances of even sixteenths where they occur, in the bottom system.[26] Breath marks leading into bars 7, 9, and 13 show instances of acceleration and delay. These occur at structural moments in the melody; they lead into the cadence at bar 7, into the second half of the tune at bar 9, and into a cadential passage (the last four bars) in bar 13. The backwards arrow in bar 10 marks a full anticipation; Svigals accelerates and comes in before the beat. Svigals is often just a tad ahead, and, at this moment, even more so.

Example 10. Alicia Svigals with Steve Greenman, Gasn Nign (1996), section A first time

Audio Example 6. Alicia Svigals with Steve Greenman, Gasn Nign section A first time.

[30] The timing variability in Svigals’ recording is especially evident in a four-note rhythm—or rather the variable rhythms that involve four events per measure. Example 11 provides five of these four-note rhythms: (a) the classic zhok rhythm, (b) a variant of the “etched” rhythm discussed above, (c) an even rhythm with sixteenths at the end, (d) a syncopated rhythm, and (e) a rhythm in four, i.e., with a quadruplet. We have already seen that Jacob Hoffman and Kandel’s Orchestra primarily use rhythm a (see Example 1 above), whereas Abe Schwartz uses a and b (Example 2 above). Svigals also uses rhythms a and b in the first iteration of the A section (Example 10): she uses rhythm a throughout (bars 4, 6, 9, 12, 15) and rhythm b in the rising and falling scalar figure (bars 10­–11). Svigals in fact demonstrates the transformation of rhythm a into rhythm b in her pedagogical video (2013a); she plays the rising scalar figure from this tune (bars 9­–10) with classic zhok rhythm (a) and then observes that it “can become” rhythm (b). She also demonstrates the transformation of these rhythms into even quadruplets (e).

Example 11. Two four-note rhythms.

[31] Svigals continues to vary the rhythms as the performance continues. Example 12 transcribes the repeat of the A section; Audio Example 7 provides the recording excerpt. Svigals uses three different four-note rhythms in succession in the scalar rise and fall (bars 9–12); letters in the annotations correspond with the rhythms in Example 11. Arrows in Example 12 show anticipations and breath marks indicate further instances of the accelerate-and-delay gesture.

Example 12. Alicia Svigals with Steve Greenman, Gasn Nign (1996), section A second time.

Audio Example 7. Alicia Svigals with Steve Greenman, Gasn Nign, section A second time.

[32] The rhythms then evolve further in section B (Example 13, Audio Example 8). Annotations show selected instances of rhythms a, b, d, and e.[27] The quadruplets of rhythm e are also offset, beginning on beat three of the measure. The syncopated rhythm d becomes a recurring motive, each time in bars 5–6 of the eight-bar phrase with a plaintive descending figure. This syncopated figure may be understood as dividing the bar in two.[28]

Example 13. Gasn Nign, Alicia Svigals with Steve Greenman (1996), section B with repeat.

Audio Example 8. Alicia Svigals with Steve Greenman, Gasn Nign, section B with repeat.

[33] We can think of these rhythms as variable options, or we can model them with transformation pathways. Example 14 illustrates a set of transformations with multiple performances and iterations of a passage from section B. The top line shows this passage in the Schwartz recording with rhythm b (the “etched” rhythm); the middle line shows the same passage in Svigals’ performance the first time with rhythm d (syncopated); and the third line shows the same passage again in Svigals’ performance the second time with rhythm e (the quadruplets, offset from the bar). The notion of “transformation” here can be understood historically, with similarity and change from Schwartz’s recording to Svigals’, and compositionally, with musical development in the repetitions of Svigals’ own performance. It also echoes Svigals’ demonstration, showing how the classic zhok rhythm “can become” something else (Svigals 2013a). Finally, we can consider the transformational processes themselves. The transformation from rhythm b to d (first to second line in Example 14) involves anticipation, it is another example of acceleration and delay. The transformation from rhythm d to e (second to third line in Example 14) involves equalization and anticipation.

Example 14. Rhythmic transformations in the Gasn Nign.

 

Beyond the Gasn Nign

 

[34] An excerpt from another tune, composed and performed by contemporary musicians, will show how we might continue to apply the analytical techniques and ideas of this paper beyond the Gasn Nign and the genre of the zhok. The tune is “I Flow with Water Under the Ice,” composed by Cookie Segelstein and performed by Segelstein with Joshua Horowitz on tsimbl (track 29 on The Magid Chronicles; Veretski Pass with Joel Rubin 2019). In discussion, Segelstein indicated that she conceived of the tune in a khosidl tempo (phone conversation, May 5, 2023). The khosidl is an instrumental genre in a moderate tempo, related to Hassidic dancing (Feldman 2016, 315). The goal with this analysis is not to link timing features of the zhok and khosidl genres as such (they come from different sources and have distinct gestural qualities), nor to document a direct lineage of performance practice (Segelstein’s influences include the music of her family from the Carpathian Mountains region, now in Western Ukraine, as well as classical training and performance experience). Rather, the goal is to explore the continuing relevance of a few basic timing elements across genres and time periods.[29] I also continue to adapt and develop forms of representation for timing in klezmer music with this analysis.

[35] We observed earlier how aspects of timing in Svigals’ performance of the Gasn Nign coordinate with phrase structure. She uses the accelerate-and-delay gesture at structural moments, leading into cadences and new phrases (see Example 10 above). There is a similar coordination of timing elements and melodic phrase structure in Segelstein’s “I Flow with Water Under the Ice.” Example 15 provides a transcription of section A with the repeat in 2/4, following notation that Horowitz shared with me. Within that framework, I transcribed pitches and rhythms from the recording with annotations for timing.[30] Audio Example 9 provides an excerpt from the recording.

Example 15. Cookie Segelstein and Joshua Horowitz, “I Flow with Water Under the Ice” (2019), section A.

Audio Example 9. Segelstein and Horowitz, “I Flow with Water Under the Ice,” section A.

[36] Segelstein anticipates the beat within melodic groups (see the backward arrows) and synchronizes with Horowitz at cadences or group boundaries. Segelstein and Horowitz often use this technique of non-alignment mid-phrase and alignment at the end. When I shared this analysis with Horowitz, he noted that they jokingly use the expression “meet you at the cadence” to affirm their intentions (phone conversation, May 5, 2023). The technique is analogous to the “cadential anchoring” that Ashley (2002, 320) identifies in the performance of jazz ballads. It is also a form of the accelerate-and-delay strategy, but with more frequent anticipations of the beat (mostly the second beat, but sometimes also the downbeat; see mm. 4 and 12 in the transcription).

[37] Breath marks in mm. 9 and 10 of Example 15 show classic instances of the move-ahead-and-wait gesture; the sixteenths are played fast and there is a break before the next beat. It is a strategic use of the classic timing technique in combination with other effects. The gesture creates a form of emphasis, as though to say, “you heard this before, but listen now again, and it is a little different.” A comparison of mm. 2 and 10 in the transcription shows the melodic variation that goes along with this timing.[31]

[38] We hear the classic short-long timing for pairs of sixteenths in mm. 6, 7, and 15. Tenuto marks show the frequent lengthening of the first sixteenths in sets of four. There is also beautiful variability in the timing of the rising sixteenth arpeggios (mm. 1, 5, 9, and 13), which is difficult to capture in notation. Segelstein dwells and luxuriates ever so slightly on the tenutos in mm. 1 and 9. When the arpeggios come back in mm. 5 and 13, they are faster and more even, as though to say, “we know this already, let’s get on with it.”

[39] Example 16 provides timing profiles for the sixteenths in the rising arpeggios of mm. 1, 5, 9, and 13. The graphs on the left side show the beginning of the section (mm. 1 and 9); the graphs on the right show the analogous arpeggiated ascent mid-section (mm. 5 and 13). The contrast between beat 1.1 and 5.1 is especially clear. In beat 1.1, the first sixteenth is elongated to close to 0.24 seconds while the others vary between 0.15 and 0.18 seconds; in beat 5.1, all four sixteenths are in a narrow range between 0.14 and 0.17 seconds. The difference between beat 9.1 and 13.1 is not as stark. But beat 9.1 has a similar timing profile as beat 1.1, with the first sixteenth as the longest and the third as the shortest.

Example 16. “I Flow with Water”: timing for groupings of four sixteenths (rising arpeggios).

[40] The continuous sixteenths in mm. 3–4 and 11–12 of the transcription present yet another timing profile: on average, the first sixteenths are even longer and the others are shorter. Example 17 shows the average durations of sixteenth notes in mm. 3.1–4.1 and 11.1–12.1 (six sets of four sixteenths). The first sixteenth is 0.31 seconds on average, whereas the second, third, and fourth are less than half of that (0.13, 0.14, and 0.14 seconds). Example 18 shows the starkest contrast in duration, which is on beat 3.2: the first sixteenth (0.38 seconds) is three times longer than the others (0.12, 0.12, and 0.13 seconds). Once again, the measurements show a high degree of timing variability.

Example 17. “I Flow with Water”: average timing for mm. 3.1­–4.1 and 11.1­–12.1.

Example 18. “I Flow with Water”: timing for m. 3.2.

[41] And so, how might we best represent timing and rhythmic variability in klezmer music? We can use descriptive language; recall my comments above about relative durations and the effects of “dwelling,” “luxuriating,” and then “getting on with it.” We can add annotation to a transcription as in Example 15 above. And we can document the timing with measurements as in Examples 17­–19. I chose a single characteristic figure for each case study in this paper. Here in Segelstein’s recording, it was the sets of four sixteenths; in the Schwartz and Brandwein recordings it was the “etched” figure, which is typically notated with a sixteenth, dotted eighth, and eighth; and in Svigals’ recording it was the four-note figure in multiple rhythmic configurations. These combined methods give us a well-rounded picture of performance style relevant for listening, performance, and comparative study. They allow for the discovery of common trends, such as the accelerate-and-delay gesture, and they document the diversity of performance practice in historical and contemporary recordings.

[42] The representation of measurements can vary depending on the context. I introduced bar graphs and durations in seconds for “I Flow with Water Under the Ice” for two reasons. We are dealing with nominally equal values, at least based on notation from Segelstein, and so the bar graphs show the differences clearly. In the Gasn Nign, on the other hand, the notational values—when given in sources—are not equal, and so a bar graph is less informative.[32] And since Segelstein anticipates the beat often, it is less relevant to provide percentages of the beat for the durations, as I did above for the Gasn Nign.

[43] There is one more form of representation that can be relevant and helpful: a transcription that documents the durations in musical values more precisely. Example 19 provides the first four bars of “I Flow with Water Under the Ice” with the sixteenths, tenuto marks, and arrows below and more precise rhythmic notation above. An audio clip accompanies the example so that readers can listen again (Audio Example 10). The triplet sixteenths are relevant not only as more accurate durational representations, but also because they match the underlying rhythm of Horowitz’s accompaniment on the Tsimbl. Horowitz relates the triplet accompaniment style to Romanian sirbas (phone conversation, May 5, 2023). And the effect of this layering, with Segelstein playing in and out of phase at multiple levels is indeed like water flowing under the ice.

Example 19. “I Flow with Water” bars 1­–4: alternative rhythmic notation.

Audio Example 10. Cookie Segelstein and Joshua Horowitz, “I Flow with Water Under the Ice,” bars 1-4.

 

Conclusion

 

[44] The three elements that I have outlined today—variability of timing and rhythm, non-isochronous beats in the zhok/hora, and the move-ahead-and-wait or accelerate-and-delay gesture—can be used as a framework for understanding and analyzing performance practice in a broad range of klezmer recordings, both historical and contemporary. I have also tracked asynchronies between melody and accompaniment, with the melody ahead of the rhythm section at strategic moments and synchronized at others. And I have related timing to melody and phrase structure in several recordings.

[45] The elements that I have documented do not exhaust timing in klezmer, by any means. There are performances where the tempo accelerates gradually throughout a track; this is especially common in recordings from an early 20^th century group known as Belf’s Romanian Orchestra. There are heterophonic styles where lead instruments double the melody in a non-aligned fashion, which is also prevalent in Jewish prayer, the music of Moldavian lăutari, and other regional styles. The timing of so-called free rhythm klezmer genres like the doina or the kale bazetsn would be an area for further study. And for dance genres, we can document and explore the relations between timing, ornamentation, dynamic emphasis, and bodily movement.[33]

[46] My approach here has been to identify some common elements of klezmer performance practice and how they have carried through from historical recordings made in New York in the 1920s to current practice. With this initial framework place, we might go on in future studies to compare timing profiles in recordings from various geographic centers (e.g., regions of Eastern Europe, New York, Philadelphia), periods (before and after the 1920s), and performers. The historical discography is not large so the opportunities for broader corpus studies are limited. But we can delineate and study particular corpora such as the set of Belf recordings made for the Warsaw-based Syrena Record company before World War I (see Wollock 1997). Feldman (2016, 109–10) observes that rhythms with internally shifting pulses, which he refers to as “ovoid” following Jean During (1997), are common in early twentieth-century recordings from the historical region of Galicia (now southeastern Poland and western Ukraine); as such, these would form a valuable earlier repertoire for study.[34] There is a good example in the 1909 recording “Oriental Themes” from the violinist Zehngut (reissued on Aylward and Rubin 2015, track 18). At 2:58 on the track, the tsimbl accompaniment seems to be duple, but the timing of the middle beat is variable and leans slightly towards the non-isochronous timings observed above.[35] With genres or tunes that are shared by different ethnic groups, we can also compare the timings from Jewish musicians with those of other performers. And a further study of timing in recordings from the revival period would deepen our understanding of recent and current trends.

[47] Finally, in a broader context, this study contributes to our understanding of aksak meters from Southeastern Europe. I believe it is the first study to measure timing in a long-short metric cycle, here associated with the klezmer zhok.[36] And in oral traditions, in which scores are not definitive, variable timing merges seamlessly into variable rhythm. Timing measurements are then valuable alongside transcription and annotation for the representation and study of performance practice.

 

Acknowledgments

 

Earlier versions of this article were presented at the AAWM Special Topics Symposium on Theoretical, Analytical, and Cognitive Approaches to Rhythm and Meter in World Musics (June 2023) and a meeting of the Jewish Musics Analysis Group (November 2023). I am grateful for valuable feedback on earlier versions from Rosa Abrahams, Christina Crowder, Walter Zev Feldman, Daniel Goldberg, Joshua Horowitz, Daniel Katz, Daphne Leong, Joel Rubin, Uri Schreter, Cookie Segelstein, Mark Slobin, Alicia Svigals, Gabriel Zuckerberg, and the anonymous reviewers. I thank Avivah Malin for assistance with the rhythmic variability graph.

 

References

 

Ashley, Richard. 2002. “Do[n’t] Change a Hair for Me: The Art of Jazz Rubato.” Music Perception 19 (3): 311–32. https://doi.org/10.1525/mp.2002.19.3.311.

Beregovski, Moshe. 2015. Jewish Instrumental Folk Music: The Collections and Writings of Moshe Beregovski. Edited by Mark Slobin, Michael Alpert, and Robert Rothstein. Second edition, revised by Kurt Björling. Draft 3.4. Self-published.

Blacksberg, Dan. 2017. “Veretski Pass.” November 3, 2017, on Radiant Others, podcast, https://radiant-others-a.blubrry.net/veretski-pass/.

Bonini-Baraldi, Filippo, Emmanuel Bigand, and Thierry Pozzo. 2015. “Measuring Aksak Rhythm and Synchronization in Transylvanian Village Music by Using Motion Capture.” Empirical Musicology Review 10 (4): 265–91. https://doi.org/10.18061/emr.v10i4.4891.

Cler, Jerome. 2015. “Commentary on ‘Measuring Aksak Rhythm and Synchronization in Transylvanian Village Music by Using Motion Capture.’” Empirical Musicology Review 10 (4): 302–4. https://doi.org/10.18061/emr.v10i4.4960.

Csaranko, Szilvia, and Susi Evans. 2020. The Klezmer Playbook: 50 Traditional Klezmer Tunes for All Instruments. Self-published. https://www.shades-of-folk.com/en/klezmer-playbook/.

Danielsen, Anne, Kristian Nymoen, Evan Anderson, Guilherme Schmidt Câmara, Martin Torvik Langerød, Marc R. Thompson, and Justin London. 2019. “Where Is the Beat in That Note? Effects of Attack, Duration, and Frequency on the Perceived Timing of Musical and Quasi-Musical Sounds.” Journal of Experimental Psychology: Human Perception and Performance 45: 402–18. https://doi.org/10.1037/xhp0000611.

Danielsen, Anne, Kristian Nymoen, Martin Torvik Langerød, Eirik Jacobsen, Mats Johansson, and Justin London. 2022. “Sounds Familiar(?): Expertise with Specific Musical Genres Modulates Timing Perception and Micro-Level Synchronization to Auditory Stimuli.” Attention, Perception, & Psychophysics 84 (2): 599–615. https://doi.org/10.3758/s13414-021-02393-z.

During, Jean. 1997. “Rhythmes ovoïdes et quadrature du cycle.” Cahiers d’ethnomusicologie 10: 17–36.

Feldman, Walter Zev. 2002. “Bulgărească/Blugarish/Bulgar: The Transformation of a Klezmer Dance Genre.” In American Klezmer: Its Roots and Offshoots, edited by Mark Slobin, 84–124. Berkeley: University of California Press.

———. 2003. “Remembrance of Things Past: Klezmer Musicians of Galicia.” Polin 16: 29–57.

———. 2016. Klezmer: Music, History, and Memory. New York: Oxford University Press.

———. 2020. “Klezmer Tunes for the Christian Bride: The Interface of Jewish and Romanian Expressive Cultures in the Wedding Table Repertoire from Northern Bessarabia.” Revista de Etnografii Și Folclor 1–2: 5–35.

———. 2022. “Musical Fusion and Allusion in the Core and the Transitional Klezmer Repertoires.” Shofar: An Interdisciplinary Journal of Jewish Studies 40 (2): 1–25. https://doi.org/10.1353/sho.2022.0026.

Goldberg, Daniel. 2015. “Timing Variations in Two Balkan Percussion Performances.” Empirical Musicology Review 10 (4): 305–28. https://doi.org/10.18061/emr.v10i4.4884.

———. 2019. “What’s the Meter of Elenino Horo? Rhythm and Timing in Drumming for a Bulgarian Folk Dance.” Analytical Approaches to World Music 7 (2). http://iftawm.org/journal/oldsite/articles/2019b/Goldberg_AAWM_Vol_7_2.html.

Jakubowski, Kelly, Rainer Polak, Martín Rocamora, Luis Jure, and Nori Jacoby. 2022. “Aesthetics of Musical Timing: Culture and Expertise Affect Preferences for Isochrony but Not Synchrony.” Cognition 227 (October): 105205. https://doi.org/10.1016/j.cognition.2022.105205.

Kirshenblatt-Gimblett, Barbara. 1995. “Introduction.” In Life Is with People: The Culture of the Shtetl, by Mark Zborowski and Elizabeth Herzog, ix–xlviii. New York: Schocken Books.

Leong, Daphne, with the Takács Quartet. 2020. “Understanding of Structure: Performed Rhythm at the Center of Bartók’s Fifth String Quartet.” In Performing Knowledge: Twentieth Century Music in Analysis and Performance, 97–132. New York: Oxford University Press. https://doi.org/10.1093/oso/9780190653545.003.0005

London, Justin. 2012. Hearing in Time: Psychological Aspects of Musical Meter. Second edition. New York: Oxford University Press.

London, Justin, Kristian Nymoen, Martin Torvik Langerød, Marc Richard Thompson, David Loberg Code, and Anne Danielsen. 2019. “A Comparison of Methods for Investigating the Perceptual Center of Musical Sounds.” Attention, Perception, & Psychophysics 81 (6): 2088–2101. https://doi.org/10.3758/s13414-019-01747-y.

Lutins, Allen. n.d. “KlezmerGuide: Comprehensive Cross-Reference of Klezmer Tune Names, Recordings & Sheet Music Sources.” Accessed August 8, 2023. https://klezmerguide.com/.

Malin, Yonatan. Forthcoming. “Listening to Klezmer Music with Schema Theory.” Journal of Music Theory.

Malin, Yonatan and Daniel Shanahan. Forthcoming. “Modes in Klezmer Music: A Corpus Study Based on Beregovski’s Jewish Instrumental Folk Music.” Music Theory Online.

Moelants, Dirk. 2006. “Perception and Performance of Aksak Metres.” Musicae Scientiae 10 (2): 147–72. https://doi.org/10.1177/102986490601000201.

Netsky, Hankus. 2002. “The Klezmer in Jewish Philadelphia, 1915-70.” In American Klezmer: Its Roots and Offshoots, edited by Mark Slobin, 52–72. Berkeley: University of California Press.

———. 2015. Klezmer: Music and Community in Twentieth-Century Jewish Philadelphia. Philadelphia: Temple University Press.

Ohriner, Mitchell. 2018. “Expressive Timing.” In The Oxford Handbook of Critical Concepts in Music Theory, edited by Alexander Rehding and Steven Rings. New York: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780190454746.013.33.

Polak, Rainer, and Justin London. 2014. “Timing and Meter in Mande Drumming from Mali.” Music Theory Online 20 (1). https://mtosmt.org/issues/mto.14.20.1/mto.14.20.1.polak-london.html.

———. 2022. “Tempo, Meter, and Form: An Analysis of ‘Dansa’ from Mali.” In Trends in World Music Analysis, edited by Lawrence Beaumont Shuster, Somangshu Mukherji, and Noé Dinnerstein, 143–58. London: Routledge.

Polak, Rainer, Nori Jacoby, Timo Fischinger, Daniel Goldberg, Andre Holzapfel, and Justin London. 2018. “Rhythmic Prototypes Across Cultures: A Comparative Study of Tapping Synchronization.” Music Perception: An Interdisciplinary Journal 36 (1): 1–23. https://doi.org/10.1525/MP.2018.36.1.1.

Rubin, Joel E. 2020. New York Klezmer in the Early Twentieth Century: The Music of Naftule Brandwein and Dave Tarras. Rochester: University of Rochester Press.

Sapoznik, Henry. 1999. Klezmer! Jewish Music from Old World to Our World. New York: Schirmer Books.

Sapoznik, Henry, and Pete Sokolow, eds. 1987. The Compleat Klezmer. Cedarhust, N.Y: Tara Publications.

Segelstein, Cookie, and Joshua Horowitz. 2005. “Rocky Rhythm Exercises.” Unpublished pedagogical material.

Slobin, Mark. 1982. Tenement Songs: The Popular Music of the Jewish Immigrants. Urbana: University of Illinois Press.

———. 2000. Fiddler on the Move: Exploring the Klezmer World. New York: Oxford University Press.

Stutschewsky, Joachim. 1959. he-klezmorim: toldotehem, oraḥ-hayehem vi-yetsirotehem. Jerusalem: Bialik Institute.

Strom, Yale. 2012. Shpil: The Art of Playing Klezmer. Lanham: The Scarecrow Press.

Svigals, Alicia. 2002. “Why We Do This Anyway: Klezmer as Jewish Youth Subculture.” In American Klezmer: Its Roots and Offshoots, edited by Mark Slobin, 211–19. Berkeley: University of California Press.

———. 2013a. Masterclass: Alicia Svigals on Klezmer. 2: Playing the Zhok. https://www.youtube.com/watch?v=b1DFw3nKPjQ.

———. 2013b. Masterclass: Alicia Svigals on Klezmer. 1: Running Semiquavers. https://www.youtube.com/watch?v=muvjcMTX9Xw.

Waletzky, Joshua. 2020. “Toward an Analysis of Rhythm in Yiddish Song.” On Inside the Yiddish Folksong: An Analytical Approach to a Treasured Tradition. https://www.yiddishfolksong.com/rhythm.

Wikipedia. 2023. “Abe Schwartz.” Last modified August 29, 2023. https://en.wikipedia.org/wiki/Abe_Schwartz.

Wikipedia. 2024. “Harry Kandel.” Last modified July 20, 2024. https://en.wikipedia.org/wiki/Harry_Kandel.

Winkler, Helen. n. d. “Helen’s Yiddish Dance Page: Dances of the Jews of Eastern Europe.” Accessed March 5, 2024. https://www.yiddishdance.com/.

Wollock, Jeffrey. 1997. “European Recordings of Jewish Instrumental Folk Music, 1911-1914.” ARSC Journal 28 (1): 36–55.

 

Discography

 

Aylward, Michael and Joel Rubin. 2015. Chekhov’s Band—Eastern European Klezmer Music from the EMI Archives 1908–1913. Compact disc. London: Renair REN 0129.

Brandwein, Naftule with the Naftule Brandwein Orchestra. 1923. Doyne, ershter teyl. Victor Records 73940. Available in the Mayrent Collection of Yiddish Recordings, https://digital.library.wisc.edu/1711.dl/EECJPXCAVX6JP8B.

Budowitz. 2007. Budowitz Live. Golden Horn Records GHP 029-2.

Hoffman, Jacob with Kandel’s Orchestra. 1923. Der gassen nigen. Victor Records 77018B. Available in the Mayrent Collection of Yiddish Recordings, https://digital.library.wisc.edu/1711.dl/RYZWZH26PBIWJ8Z.

Schwartz, Abe with Sylvia Schwartz. 1920a. National Hora. Columbia Records E4745. Reissued on Klezmer Music: Early Yiddish Instrumental Music. The First Recordings: 1908–1927. Arhoolie CD 7034, 1997. Available in the Recorded Sound Archives of the Florida Atlantic University Libraries, https://rsa.fau.edu/track/5517 (beginning at 3:58).

———. 1920b. Oriental Hora. Columbia Records E4825. Available in the Recorded Sound Archives of the Florida Atlantic University Libraries, https://rsa.fau.edu/track/2862.

———. 1921. Orientalishe hore. Emerson 13172. Available in the Mayrent Collection of Yiddish Recordings, https://digital.library.wisc.edu/1711.dl/YBECWA4NLSHSH8J.

Svigals, Alicia, with Steven Greenman. 1996. “Gas-Nign.” On A Marriage of Heaven and Earth, Ellipsis Arts CD4090.

Tarras, Dave. 1926. Mayn Tayere Odessa. Columbia Records 8103 F. Available in the Mayrent Collection of Yiddish recordings, https://digital.library.wisc.edu/1711.dl/THH3M5AKRJWEZ8C.

Veretski Pass with Joel Rubin. 2019. “I Flow With Water Under The Ice.” On The Magid Chronicles. Golden Horn Records.

[1]. A full discussion of style pedagogy and the use of historical recordings in the contemporary scene would require a separate paper. We can note, however, that sheet music publications such as Sapoznik and Sokolow 1987 and Csaranko and Evans 2020 cite historical recordings with each tune. See also pedagogical materials cited below and discussions of style in Slobin 2000, chap. 5, and Strom 2012.

[2]. Rubin (2020, 223–26) has valuable observations on timing in recordings by Brandwein and Tarras, which I build on here. Joshua Waletzky (2020) provides a rich and detailed analysis of timing and rhythm in a Yiddish folksong, “Fun groys dasad,” performed by Anna Berkowitz. Waletzky’s identification of three styles of song, “zogevdik” (speakingly), “zingevdik” (singingly), and “tantsevdik” (dancingly) (Waletzky 2020, 7) could form the basis for future comparative studies with instrumental music.

[3]. See the overview of expressive timing studies and extensions to more diverse repertoire in Ohriner 2018.

[4]. Beregovski (2015, I-44–48) describes the training of klezmer musicians in the nineteenth century.

[5]. I outline relevant melodic features in footnote 13 below.

[6]. During (1997) coins the term “ovoid” (egg shaped) for meters with three variable, non-isochronous beats in Baluch and Tajik-Uzbek traditions. Feldman (2016, 110) in turn identifies such ovoid rhythms in klezmer recordings of tsimbl and solo instrument from the historical region of Galicia and in Moldavian lăutar music, particularly in the slow hora. I return to and comment on this repertoire in the conclusion.

[7]. The present study could serve as the basis for further experimental work. Polak et al. (2018, 16) outline the value of ethnographic and analytical study for comparative study of music perception and cognition.

[8]. This study of “Gasn Nign” builds on an earlier comparative study by Mark Slobin (2000, 123–32). Spellings and titles for klezmer tunes vary widely. Other spellings include “Der Gasn Nigun” (Sapoznik and Sokolow 1987, 47) and “Gas-nign” (Slobin 2000, 123, Beregovski 2015, nos. 61­–78). Other titles for tracks with this tune are given in the Klezmer Guide (Lutins, n.d.) under “(Der) Gas(sen) Nign (Gasn Nigun).”

[9]. See examples in Beregovski 2015, nos. 61­­–78.

[10]. See also Feldman 2016, 279 and 326. The Wikipedia article on Harry Kandel (2024) cites and compares all of these sources.

[11]. For accounts of Jewish life and music in towns (shtetls) of Eastern Europe, see Slobin 1982, chap. 1; Kirshenblatt-Gimblett 1995; Feldman 2016, chap. 2; and Rubin 2020, chap. 1.

[12]. Slobin (2000, 126) reproduces Hoffman’s manuscript version of the Gasn Nign. Differences between the manuscript and this transcription include a few details of ornamentation, but also distinct rhythms. Hoffman, for instance, notates bars 9 and 10 with two eighths and two sixteenths, but he and Kandel’s Orchestra perform them with a dotted-eighth and three sixteenths.

[13]. As Rubin observed (online meeting of the Jewish Musics Analysis Group, November 15, 2024), modal features of this tune align more with core Jewish repertoire than with transitional repertoire. These features include emphasis on scale-degrees ^3 and ^4 in minor, use of ^b2 leading into the cadence, and a modal idiom with a lowered ^5 and raised ^6, here Ab and Bnatural. For more on mode and idiom in klezmer music, see Rubin 2020, Malin forthcoming, and Malin and Shanahan forthcoming.

[14]. Sapoznik (1999, 91) indicates that Schwartz emigrated with his parents in 1899. The Wikipedia article on Abe Schwartz (2023) dates his emigration as “sometime between 1900 and 1902” with citations that include the U.S. Naturalization Records.

[15]. See the discussion of the player’s “concept” in Rubin 2020, 178­–80.

[16]. Ashley identifies the reverse, “delay-accelerate” as more common in jazz: “This delay-accelerate strategy is widespread in jazz performance and is more common than the inverse, anticipate-delay (although examples can be found of the latter strategy as well) (Ashley 2002, 319).

[17]. I located beat onsets by tapping with Sonic Visualiser and then manually adjusting each beat with a combination of visual and auditory cues. I used the spectrogram layer in Sonic Visualiser for visual cues, focusing on lower frequencies for piano note onsets (which sometimes were masked by the violin). This method does not account for subtleties of perception that have been explored recently by Danielsen et al. (2019) and London et al. (2019). However, Bonini-Baraldi et al. (2015, 271) find that the consistency of the method is as important for measuring inter-onset durations as the method itself. I took a second set of measurements more than a month after the first; the average difference between the two sets of measurements was 20 milliseconds (0.02 seconds; standard deviation 0.02 seconds). The average difference including the direction of offset was 0.00—so my measurements on average were neither more ahead nor more behind the second time around. The measurements are still somewhat approximate; extreme precision is not possible with historical recordings such as this one. Sonic Visualiser files are provided for verification in the supplementary materials.

[18]. My study of this rhythm in the Gasn Nign is similar to Goldberg’s (2019) study of what he calls “basic rhythms” in the Bulgarian elenino horo. Goldberg’s rhythms come from performances on a tŭpan, a large double sided drum; the rhythm here comes from characteristic rhythmic figures in the melody. Recordings of another zhok, a tune known as “Mein Tayere Odessa” (Tarras 1926) show how a <quarter, eighth> rhythm can evolve into a <sixteenth, dotted eighth, eighth> rhythm. The topic of characteristic rhythms in the zhok and other klezmer genres merits further study.

[19]. This is based on mm. 5, 6, 13, and 14 in both iterations. Measure 7 already tends towards a more even division.

[20]. This is based on mm. 9, 10, and 11 in both iterations, leaving out the repetition of m. 10.

[21]. Similar timings with this rhythmic figure can be heard in a tune known as “Oriental Hora” with recordings by Abe and Sylvia Schwartz (1920b and 1921).

[22]. Alicia Svigals demonstrates some of these rhythms in a pedagogical video (2013a); I discuss this below in connection with her recording.

[23]. Further research could compare the Schwartz and Brandwein timings with recordings of the tune by other traditional players, i.e., those who were part of the scene in the first half of the twentieth century even if they may have recorded later. This includes Joseph Cherniavsky, Max Epstein, Israel. J. Hochman, Leon Schwartz, and Dave Tarras. See the recordings cited under “(Der) Gas(sen) Nign” in the Klezmer Guide (Lutins, n.d.) and the discussion in Slobin 2000, 123–32.

[24]. For further information about Svigals’ background and aesthetics, see Slobin 2000, 47-50 and 107-110, and Svigals 2002.

[25]. I discovered Svigals’ video after completing my analysis.

[26]. Svigals discusses this timing as a common feature of klezmer performance in a pedagogical video posted online (Svigals 2013b), and it can also be understood as a form of anticipation and delay.

[27]. Slobin (2000, 127) similarly observes that Svigals finds “new solutions” for repeating figures in the B section with each repetition.

[28]. Svigals brings back the quadruplets in a later return of section A, at 1:48 on the recording.

[29]. For information about Segelstein’s background and approach to klezmer music, see Slobin 2000, 42, 44–46, and 47; the bio at http://www.veretskipass.com/veretskipass.com/Musicians.html; and the November 2017 episode of the Radiant Others podcast (Blacksberg 2017).

[30]. There are a few pitch differences between the recording and Segelstein’s notation. The notation has two options in m. 3 and the downbeat of m. 4; the recording uses a blend of these options both times. Segelstein’s sixteenths in m. 6 embellish an eighth-note descending line in her notation. And the arpeggiated cadential figure in mm. 8 and 16 of the transcription are not in the original notation. These are classic figures, which are often added at cadences.

[31]. Segelstein’s score provides both variants for this second bar of the section.

[32]. See the Klezmer Guide (Lutins n.d.) under “(Der) Gas(sen) Nign (Gasn Nigun)” for a variety of sheet music sources.

[33]. Zev Feldman described dancing the zhok with Dave Tarras in Tarras’s Brooklyn apartment, he observed that there are typically upward hand motions with a “hiccup” or delay, and that it is not as interesting to dance the zhok if the musicians are playing in an even three (meeting of the Jewish Musics Analysis Group, November 15, 2023). Information about Yiddish dances and video links can be found at the website “Helen’s Yiddish Dance Page” (Winkler, n.d.).

[34]. See Feldman 2003 for more information on this repertoire and the klezmer families from eastern Galicia.

[35]. The tune is also known as “Oriental Hora,” and recordings by Abe and Sylvia Schwartz (1920b and 1921) adopt the non-isochronous zhok timing. My initial measurements indicate basic beats in a 3:2 ratio. Additional recordings of the tune can be located via the Klezmer Guide website (Lutins n.d.) under “Oriental Hora.”

[36]. Moelants (2006) measures timings for multiple recordings of Bulgarian traditional music with meters that he identifies as 2+3, 2+2+3, 2+2+2+3, and individual examples of 3+2+2 and 2+2+3+2+2. In addition, he measures timings in performances of pieces by Bartók with meters of 2+2+3, 2+3, 2+2+2+3, and 3+3+2. In general, aksak tends to place the shorter-length beats in the beginning of measures, whereas the klezmer zhok begins with the long beat.