‘Independence of the composer from the poet’:
A comparative analysis of eight settings of Goethe’s
Nur wer
die Sehnsucht kennt
Ruben Adriaan Meijerink (10424156) – Master’s Thesis February 16, 2020
Thesis supervisor: Dr. R.M. Helmers Second reader: Prof. dr. J.J.E. Kursell
Table of contents
Introduction 3
Context of the poem and its settings 9
The poem itself 14
Method 16 Analysis 28 Mignon (lines 1-2) 28 Mignon (lines 3-4) 35 Mignon (lines 5-6) 41 Mignon (lines 7-8) 47 Mignon (lines 9-10) 54 Mignon (lines 11-12) 61 Conclusion 67 Appendix: scores 74
Introduction
It is commonly held that music is a type of language. Since music and language both rely on organized sound, it is hard to make an absolute distinction between the two. We speak of the musicality of language, or even that music has a grammar. Musicality of language is
particularly relevant in poetry. When reciting a poem, it is easily recognized that poetic language is ‘musical’, with its rhythms, pitches and rhetorical ebb-and-flows. In some poems, the sounds of the words are even more important than merely communicating their lexical meanings. Indeed, these non-referential components of poetry make the comparison with music especially apt: both a piece of music and a poem can be seen as a composition of ordered sounds.
Poetry and music are often combined in song. In setting words to music, song composers, when presenting their musical material, need to grapple with an appropriate diction, as well as the poem’s rhythms and specific sounds. In studying text-to-music settings, then, it is an interesting analytical question how those ‘systems’ of music and language collaborate. In other words, how are the words accommodated within the vocal melody? What note goes with what syllable—is such text underlay arbitrary? Does rhythmic
differentiation in the music align in one way or another with speech rhythm; or, divisions into syllables with divisions into musical beats?
It has often been debated which of these domains—language or music—has the upper hand. The relationship between language and music is generally considered to be very strong in nineteenth-century art song, especially the German Lied. Since the eighteenth century, the idea of a union between music and poetry had been reinvigorated. Throughout the course of the nineteenth century, this union was actualized by reflecting textual details in musical settings.
Many composers of the German art song in this period were deeply interested in the relationship between poetic language and its musical setting. Lied composers turned to a wide range of poetry to set to music, some works being particularly fashionable. For example, Johann Wolfgang von Goethe’s Mignon Lieder from his novel Wilhelm Meister’s Lehrjahre were set to music many times. As the novel designated these poems as ‘songs’ in the first place, such texts obviously reinforced composers' interest in the interrelations of poetry and music. The present thesis is about one of these Mignon Lieder, called ‘Nur wer die Sehnsucht kennt’—abbreviated as ‘Sehnsucht’. It is a song sung by the character Mignon and constitutes
a single lyric moment without dramatic development. It is particularly short with an
economic form that subtly balances regularity and irregularity. Many composers have been drawn to this short poem; it has even inspired some composers to write multiple versions. In this thesis, I look at eight settings of ‘Sehnsucht’, each by a different composer, to examine the interrelations between the sound and structures of language and the musical presentation thereof.
On the basis of this poem, literary scholar Meredith McClain has set out an argument about music’s “independence” from lyrics in an 1987 article. McClain illustrates the development 1
of the German Lied by reviewing seven settings by six composers of ‘Nur wer die Sehnsucht kennt’ (from here on designated as ‘Sehnsucht’). She addresses the question to what extent the words are followed by the music or whether the composer’s music becomes independent. In her analysis, this selection of settings of the poem serves to illustrate a general “evolution” towards the independence of the composer from the poet.
She starts tracing this evolution with Johann Reichardt’s strophic setting of ‘Sehnsucht’, a setting commissioned by Goethe himself. This setting would still be in
accordance with Goethe’s own views, namely that music should follow the words. Similarly, the second composer that McClain discusses, Carl Friedrich Zelter, composed a simple strophic setting too. She explains that Zelter, just as Reichardt, was a “member of the Second Berlin Song School, so it is not surprising that his music supports and follows the poetic text.” What is more, these two early composers are quoted as saying that they read the poem 2
out loud repeatedly, in order for the correct expression in the melody to arise naturally, instead of writing an instrumental melody first and couple music and words later. 3
Indeed, Goethe himself famously endorsed such an aesthetic view, according to which words and music should be unified in the Lied. With the settings that follow, McClain 4
describes a gradual divergence from this aesthetic. This development starts with Beethoven and Schubert: we are told, although the settings are still fluently singable, and articulate the syntax musically, one can glean something of the “evolving complexity and independence of
1Meredith McClain. ‘Goethe and Music: “Nur wer die Sehnsucht kennt’’’. Johann Wolfgang von
Goethe: one hundred and fifty years of continuing vitality. Ed. Ulrich Goebel. Lubbock, Texas: Texas University Press, 1984: 201–227.
2McClain, p. 207. 3Ibid.
the musical form in relation to the verbal structure”. In addition to a relatively more complex 5
musical form, Beethoven “takes a liberty” to repeat and interpolate words, for example. 6
Significantly, then, McClain discerns a watershed in Schumann’s setting, which for her exemplifies the “independence of the composer from the poet”. Schumann would 7
represent a new direction, characterized by a great range of musical detail. McClain states that for this setting it is too difficult to “make a meaningful simple graphic reduction of its form”. Indeed, she designates the setting as completely through-composed, where the music 8
“evolves without attention to syntactic breaks”. In other words, the text is made subservient 9
to the composer’s will. McClain’s indeed states that “the music is in control of the text.” 10
Consequently, McClain closes her discussion with Tchaikovsky's setting, published almost four decades after Goethe’s death and twenty years after Schuman’s setting. Whereas the early settings were designated as simple strophic and Schumann’s as through-composed, Tchaikovsky’s musical form is so complicated in articulating the poem’s structure, McClain holds, that it is now simply impossible to outline graphically. McClain’s reason for ending 11
her survey with Tchaikovsky’s setting is the following:
Of all the settings presented it is the only one in which the music was such a
successful entity that it lives on in performances of instrumentalists [...] without the words and many people who today would recognize the melody would have no idea who Goethe was or that it was his poetry which had inspired the music 12
Of course, it is not this proposition itself that is interesting (let alone plausible) but rather the idea that is implied: Goethe’s poem as a poem can hardly be recognized anymore because the music has become independent to such a large extent. She indeed writes that: “For this reason this setting completes the evolution we have been following since the discussion of the original Reichardt setting” However, a skeptical reader might reply: when such a piece of 13
music exists independently from the words it was based on, what does this fact necessarily
5McClain, p. 210 6Ibid, p. 209 7Ibid, p. 211 8Ibid. 9Ibid. 10Ibid., p. 212 11 Ibid. 12Ibid. 13Ibid.
imply? Cannot it both be an independent piece of music and fit on the words as well? Indeed, McClain does address an interesting issue, namely what fate textual details have over time and to what extent the music may come to reign independently. The picture of development towards through-composed songs and a more independent role for the piano are indeed commonplace in histories of music. However, the particular idea that a poem dissolves in 14
the musical setting, ceases to be poetry and becomes music as it were, has been criticized by recent authors.
A recent article by Stephen Rodgers takes aim at such a view in musical analysis, dubbed the ‘assimilation model’ by Kofi Agawu. Rodgers explains this model as follows: 15
Analysts who adhere to this view argue that a poem ceases to be a poem when it is set to music: ‘purely’ poetic elements, such as line length, meter, enjambment,
assonance, alliteration and caesuras, are so thoroughly absorbed by the music that they cease to be very relevant to an understanding of the song. 16
This paradigm, Rodgers proceeds to show, has been challenged before by various
musicologists, such as Berthold Hoeckner (quoted by Rodgers) who writes: “Even when a poem has been molded into a through-composed song; even when its words have lost the rhythm of their original meter; and even when its text has been altered by the composer: the poetic text still remains an independent component of a song” . This insight is important for 17
Rodgers, as he contends that precisely this independent component of the poem deserves more attention in the study of text-to-music settings. In other words, he wants to give proper due to the sounds and structures of the poem, apart from what the words are thought to mean. For this purpose, he draws strongly on the work of Don Michael Randel and Robert
Snarrenberg. All three authors have in common, that they first consider such independent 18
linguistic material of the poem, its sounds, and subsequently study the composers response.
14J. Peter Burkholder, Donald Jay Grout, and Claude V. Palisca. A History of Western Music: Tenth
International Student Edition. New York: W.W. Norton & Company, 2019, p. 590.
15Stephen Rodgers. ‘Song and the Music of Poetry.’ Music Analysis 36.3 (2017): 315-349; Kofi
Agawu. ‘Theory and Practice in the Analysis of the Nineteenth-Century ‘Lied’.’ Music Analysis 11.1 (1992): 5.
16Rodgers, pp. 316-317. 17Ibid., p. 317.
18Don Michael Randel. ‘Congruence between Poetry and Music in Schumann’s Dichterliebe.’
19th-Century Music 38.1 (2014): 30-52.; Robert Snarrenberg. ‘On the Prosody of German Lyric Song.’ Journal of Music Theory 58.2 (2014): 103-154.
It is particularly interesting that these recent authors under review focus on composers from the Romantic era—Schumann in particular. Crucially, the authors document that nineteenth century composers, too, carefully read out loud the poems before setting them to music as well, much in keeping with the statements of the Second Berlin School composers that McClain had quoted, which I will elaborate on in the next chapter.
Significantly, the ideas from Rodgers, Randel and Snarrenberg diverge from the ‘assimilation model’ as well as familiar music-and-text perspectives that predominantly look for semantic correspondences between words and music. At times, however, these recent authors still use their observations as a technique of text interpretation: they do not intend to exclude
hermeneutics. Conversely, a strength of McClain’s comparative analysis is that she employs a single analytical yardstick—independence of the composer from the poet—for multiple settings of the poem, not taking occasional detours to the interpretation of the poem’s
meaning, for example. However, in sketching the picture of independence, McClain does not analyse systematically or in detail what features and structures of the poem are purportedly lost, and to what extent this happens over time. Relying on single observations, she inevitably sketches her model of development with a broad brush. In order to study this general idea of independence more precisely, I believe it is warranted to look closely and systematically at the poem’s words and structures as they appear in declamation: likely such features direct compositional choices. These textual features include meter, poetic structure, pronunciation and even the phonetic sound of words. Richly inspired by the three recent authors mentioned above, I aim to analyse musical settings from such a viewpoint in this thesis.
While McClain is thus not ultimately my central point of reference, I do take some of her hypotheses into account: those about irregularity, syntactic breaks and the relation between poetic structures and musical structures, as these bear on the idea of independence. Taking my cue from Rodgers, Randel and Snarrenberg, then, it is interesting to analyse the treatment of the composer with these aspects.
In short, following McClain, it is valuable to analyse settings of ‘Sehnsucht’
comparatively, in order to see how musical settings respond to textual features as mentioned
above. By studying multiple settings, I can make a stronger argument as to how a musical setting can be tied back to textual elements and to what extent this is so, and to what extent a chronological development can be detected in this matter. For this purpose, I take up a selection of analytical observations from the literature and apply them to multiple settings, in
order to investigate whether such observations can be generalized or remain unique.
My own comparative analysis of multiple settings, then, ought to provide more insight into the issue of “independence” across the chronology. I aim to reconsider and re-evaluate the idea of development towards independence of the composer from the poet: not to ‘debunk’ it per se, but to study if and how such an evolution to independence might be mapped. What textual features are ignored and to what extent? Alternatively, what textual features are supported—or interacted with—by the vocal melody? And seen from a historical perspective, could this shed light on any developing ‘independence’ through time? These angles may help us to understand the general idea of independence better with respect to these particular Lied settings, and secondarily, the viability of the methods of Rodgers, Randel and Snarrenberg. This thesis steps in to answer these questions, for which I take the most useful elements of the articles described above. I will elaborate on these studies and set out my own method in detail below, but first I provide some contextual information on the poem that is the object of this study.
Context of the poem and its settings
In this chapter I will briefly provide some information on the poem and the settings that I have included for analysis. The poem under review is taken from Goethe's, classic Bildungsroman Wilhelm Meister's Lehrjahre, published in 1795. The novel is about the protagonist Wilhelm Meister's journey to self-realization. Embedded within the narrative, this poem constitutes one of four famous “Mignon songs”. Mignon is a mysterious waif-like character in the novel, about 13 or 14 years of age. She was rescued by Wilhelm from a troop of dancers who had abducted her from her homeland. The girl willingly accepts his protection and forms a close bond with Wilhelm, and accompanies him on his adventures.
This song is located near the end of book four. While Wilhelm is resting in his room, suffering from wounds inflicted on him by robbers, he falls in a state of dreamy longing. From outside of his window, he overhears the poem being sung, its words echoing his own yearning for a beautiful Amazon that had saved his life. Mignon sings the poem together with a character called the Harpist: according to the narrator, it is an “irregular duet sung with the most heartfelt expression.” Although it is designated as a duet, the words attest to distinctly 19
personal expression. In fact the poem has been associated with Mignon almost exclusively, and is often called simply Mignon’s Lied.
Composers were fascinated by the Mignon as a character and the highly expressive words of this poem, spurring many interpretations. Indeed, at least 56 settings are known by
professional composers. The poem undoubtedly attracts composers due to the preeminence 20
of Goethe as a poet, and the increasing interest in the bond between music and poetry. This affinity between music and lyrics that Goethe himself espoused would continue to fascinate composers throughout the nineteenth century. As the poem is not recited in the novel but actually sung, it is no wonder composers were attracted to this text.
My own choice for selecting this poem is partly pragmatical. As a famous poem that has been set multiple times, material is readily available for comparative analysis. The form of the poem, too, lends itself well for comparison, due to its tight structure and recurring
19 Thomas Carlyle. Wilhelm Meister’s Apprenticeship and Travels: Translated From the German of
Goethe. London: Chapman and Hall, 1888, p. 211.
20Jack M. Stein ‘Musical Settings of the Songs from Wilhelm Meister.’ Comparative Literature 22.2
features. What is more, the poem has no dramatic development or narrative, which makes the poem especially conducive to formalist analysis.
Most importantly, of course, I want to specifically follow up on the issues that McClain addresses. This is why I have included all but one her selection of settings reviewed in her article. Indeed, of her selection of settings I include: Reichardt, Schubert, Beethoven, Schumann and Tchaikovsky. I have omitted Zelter. To these I have included three additional settings, from Ferdinand Hiller, Hugo Wolf and Gustav Jenner. First, I have selected these particular settings for roughly two reasons: to represent the whole century as much as
possible, that is to say: find an even distribution of settings through the century. Second, more pragmatically, I had to consider whether a score was available. In what follows I will briefly introduce the settings of the eight composers under review.
The setting of Johann Friedrich Reichardt (1752-1814) from 1795 was commissioned and approved by Goethe himself. In fact, the very first edition of Wilhelm Meister's Lehrjahre included Reichardt's music for all songs in the novel—eight in total. Goethe's influence on his composition is echoed in the remark by Reichardt quoted in the introduction. This
composition is the only setting under review that is actually a duet. However, I will only analyse the upper voice (i.e. Mignon), which differs in melodic shape and one single rhythmic irregularity.
In 1810, Ludwig van Beethoven published this setting together with three other settings of the same poem. This reflects his typical experimental spirit and dissatisfaction with his own creative works. The setting chosen for the current analysis was his last and, according to McClain, his most successful attempt. Although it is the only through-composed setting of 21
his collection, according to Jack Stein all the settings “are extremely simple and brief, very strongly in the eighteenth-century tradition.” According to the New Grove, however, his 22
next 1816 cycle would be closest to the Volksweise endorsed by Goethe. Beethoven said 23
that the rhythms of Goethe’s poetry had great power over him, and that melody realized the sensuous element of poetry in general and the song of Mignon in particular. While Goethe's 24
21McClain, p. 209 22Stein, p. 132.
23 Douglas Johnson, Scott G. Burnham, William Drabkin, Joseph Kerman, and Alan Tyson.
‘Beethoven, Ludwig van.’ Grove music online.Oxford music online. Oxford University Press. 2001. 14 November 2019.
opinion on this setting is not known, McClain quotes a “telling” reaction to his setting of
Kennst du das Land, where Goethe reproached Beethoven for not writing simply enough. 25
It is documented, too, that Goethe was disappointed with Franz Schubert’s interpretation. Schubert's 1826 setting under review is the last of no less than six versions. Schubert is often seen as the quintessential Lied composer. A general view in his lifetime held, that he
introduces more formal freedom and complexity to the folk-like strophic style of the Second Berlin School. Modern scholarship stresses the variation in Schubert's compositional method throughout his career and that his poem settings are indeed stylistically divergent. 26
Robert Schumann spent a fair amount of time of his life on the composition of Lieder. Himself conversant in literature, he strongly believed that music and poetry were interrelated as art forms. As such, he sought for a fusion of poetry and music, in which composer and poet are co-creators. 27
His setting of ‘Sehnsucht’ from 1849 is part of his song cycle Lieder und Gesänge aus
Wilhelm Meister. Stylistically, the songs attest to very different styles compared to his earlier
song cycles. The songs employ more chromaticism and dissonance and are more
declamatory, almost operatic. The cycle is considered to be of comparatively low aesthetic 28
value, compared with his earlier Lied cycles. As are most works of Schumann’s ‘late style’, the songs are commonly seen as a deterioration as compared to his previous output. Quite a few commentators, however, do see value in Schumann Wilhelm Meister settings. For example, Stein writes that, “The songs are richer in texture, the harmony is more complex, the musical declamation more subordinate to the poetic line, and the songs in general more dramatically conceived.” Schumann, a watershed in McClain’s article, and the three 29
composers that follow in the chronology, will figure as ‘late composers’ in my discussion.
Ferdinand Hiller (1811-1885) was a close friend of Schumann, who, together with Felix Mendelssohn influenced Hiller as a composer. This setting is from 1867, published around
25Ibid.
26Maurice J.E. Brown, Eric Sams and Robert Winter. ‘Schubert, Franz (Peter).’ Grove music online.
Oxford music online. Oxford University Press. 2001. 11 February.
27Jürgen Thym. ‘Schumann: reconfiguring the Lied.’ The Cambridge Companion to the Lied. Edited
by J. Parsons & J. Cross. Cambridge: Cambridge University Press, 2004, p. 126.
28Ronald Taylor. Robert Schumann: His life and work. London: Granada Publishing, 1982, p. 127. 29Stein, p. 139.
the same time as Tchaikovsky’s. Though Hiller is fairly unknown today, he was an important figure in his own time. The composer is considered a graceful composer especially in the smaller forms. When he went to Weimar to become one of Hummel’s pupils (from 1825-27), he met Goethe and played concerts at his home. A prolific writer, he also wrote a biography called Goethes musikalisches Leben. This song is part of his 12 Gesänge, Op.129.30
Pyotr Tchaikovsky is generally associated with his instrumental compositions. However, he wrote a fair amount of vocal music, often less well-known outside of Russia. This setting of
Nur wer die Sehnsucht kennt is the last of his collection Six Romances op. 6, composed in
1869. This collection was successful in its time, and the setting of ‘Sehnsucht’ is perhaps the best known today. While Tchaikovsky is not seen as an equal to the greatest of Lied
composers, the New Grove believes it is hard to deny the vocal excellence in many of his songs. 31
For this setting, Tchaikovsky used Lev Alexandrovich Mey’s translation of the German words into Russian. The translation preserves the original metrics of the German, and the scholar Richard Sylvester believes that “Tchaikovsky's music is perfectly suited to Goethe’s text, including the lines which Tchaikovsky repeats, as can be seen in the frequent recording of the song in German.” A different language raises all sorts of questions and qualifications. For the present project, I simply use the score provided by McClain with German text, in order to follow her argument in which Tchaikovsky’s setting has a key role, as Tchaikovsky marks the endpoint of her discussion. For this thesis, I have included some later settings to further investigate the issues at hand.
In my view, Hugo Wolf is an essential composer to include. In contrast to Tchaikovsky, he devoted most of his life to vocal music, particularly the art song. His absence from McClain's selection is interesting, as he is often named in general overviews of the Mignon lieder. For example, Jack Stein, an author heavily referenced by McClain, includes Wolf “among the most important of the countless settings that have appeared.”32 What is more, Wolf's Mignon settings, including this setting from 1888, are considered the among best he wrote.
30Reinhold Sietz. ‘Hiller, Ferdinand’ Grove music online.Oxford music online. Oxford University
Press. 2001. 14 November 2019.
31Roland J. Wiley. ‘Tchaikovsky, Pyotr Il′yich’ Grove music online.Oxford music online. Oxford
University Press. 2001. 14 November 2019.
In general, his Lieder lean towards opera idiomatically, as he was strongly influenced by Wagner, an influence that is most obviously seen in the intense chromaticism that was in vogue. Generally, for his dramatic and lyric ambitions, Wolf did not believe the strophic form was suitable. In contrast to Schumann, he has a reputation to meticulously adhere to the poetic text. Compared to most Lied composers, he valued perfect declamation to a large 33
extent. A distinctive practice of his, is that he prefaced the performance of his Lieder34 with a
recitation of the words. In this he reasserted the centrality of the poem, where he strived for an equality of poetry and music. 35
The latest setting from 1900 that I discuss is from a largely unknown composer, Gustav Jenner (1865-1920). He was a composition student of Brahms and one of many who praised the union of music and words that Brahms could achieve. Brahms scholar Heather Platt summarizes Jenner’s comments in which he summarizes his master’s view that “the music must reflect the structure and meaning of the original poem, and he [Jenner] notes that such relations between the text and music include form, musical and verbal syntax, declamation, word painting, and harmony.” According to Platt, 36 “Jenner begins by emphasizing the necessity of thoroughly knowing the text before beginning to compose a Lied, and he recalls that Brahms specified familiarity with all of the intricacies of a poem including being able to recite it correctly.” Such statements inform my analytical angle. In keeping with them, I will 37
consider the text of the poem itself in the next section.
33Lawrence Kramer. ‘Hugo Wolf: Subjectivity in the Fin-de-Siécle Lied.’ German Lieder in the
Nineteenth Century, edited by Rufus Hallmark, Routledge, 2009, p. 187.
34Kramer, p. 189 35Ibid.
36Heather Platt. ‘The Lieder of Brahms.’ The Cambridge Companion
to Brahms. Edited by J. Parsons. Cambridge: Cambridge University Press, 2004, p. 185.
The poem itself
In this section, I discuss the text of the poem itself. I only sketch a general overview of the poem’s structural features. I will not zoom into individual lines or line pairs much, let alone words and phonemes—this is largely reserved for the analytical chapter.
1 Nur wer die Sehnsucht kennt A
2 Weiss, was ich Leide! B
3 Allein und abgetrennt A
4 Von aller Freude, (B)
5 Seh’ ich ans Firmament A
6 Nach jener Seite. C
7 Ach! der mich liebt kennt, A
8 Ist in der Weite. C
9 Es schwindelt mir, es brennt A
10 Mein Eingeweide, C
11 Nur wer die Sehnsucht kennt, A
12 Weiss, was ich Leide! B
The poem has only 12 verse lines, each line being rather short. In the novel, the poem is printed as one single stanza. Its first two lines are repeated at the conclusion of the poem, which gives the whole a ‘cyclical’ form. From the onset, the poem alternates lines of six and five syllables, where the initial line ends on a stress, and the answering line ends on an unstressed syllable, i.e. masculine and feminine endings. In effect, lines of three and two beats alternate. On the above account, according to McClain, the poem can be divided in two segments as “the basic rhythmic structure indicates two stanzas based on the same pattern.” 38
As indicated, the rhyme scheme is AB, A(B), AC, AC, AB, AB. In this view, the poem can be seen as one stanza with three quatrains, where the four middle lines are one rhyme group ACAC. Because ABAB returns at the end, the framing function of the repeated
final lines is underscored. Even taking into account the half-rhyme of (B), the two rhyme schemes are rather similar. Indeed, the constant return of A further creates a sense of regularity. Grammatically, each line pair constitutes one syntactic unit. Lines 3,4,5,6 form one sentence of four lines. The subordinate clauses of this sentence are connected through enjambment and a comma (3-4) and again an enjambment and a comma (5-6).
Following McClain, and other commentators, I will divide the poem in two stanzas as well. An additional good argument in favor of this division is that line 7 has a syntactic structure similar to line 1.
The verse feet are not completely axiomatic either. That becomes apparent in the phrasing when McClain’s writes that “one way to describe the pattern of metric feet is to assume that the basic foot is iambic (u /), but that the rhythm is suspended or counterbalanced by word stress in in the first foot of lines, 1, 2, and 5 in stanza one and of lines 1, 5, 6 in stanza two, causing trochaic feet (/ u)”. Scholar of German literature Harry Seelig, designates the lines 39
of the poem as “dactylic-trochaic” without explanation. He means that the verse feet are 40
either (/ u u) and (/ u) throughout. As to both views, the difference lies in only where one marks off the verse feet, that is to say when a set of stressed and unstressed syllables begins and closes. In this sense, interestingly, the accents in both accounts do not necessarily conflict. As there are no indications of verse feet printed alongside the poem, however, the lines require some additional explanation, which I provide in the analytical chapters that follow.
But first I will set out my own method.39Ibid.
Method
Before I present my analysis, I will review the literature named in the introduction and set out my own method. These studies provide the general theoretical framework that I draw on and respond to. The more recent authors in question—Randel, Rodgers and Snarrenberg—can
teach us how to pay attention to poetic sound and sound structure present within the poem, contending that composers have attended to these immanent features of the poem before setting the words to music. These authors offer various analytical observations to investigate the composers’ interaction with the poem, a helpful supplement to McClain’s questions and observations. These studies will partly guide my own project, as they illuminate a particular way of thinking about text-to-music setting.
It is crucial to note that these three authors describe the relation between musical setting and the words as one of ‘response’, ‘interaction’, ‘convergence of sound worlds’ and how the materiality of the poem ‘constrains’ the setting. In this sense they diverge from McClain, as the above terms contradict the hierarchy that McClain implies in her phrase “independence of the composer from the poet” and the composer being “in control”. 41
By contrast, Randel tellingly describes his approach as viewing poetry and music on more “parallel footing”. 42
This parallel footing is found in the the fact that the two systems both have sound. Having established this view, the authors want to show what this can reveal in individual works: that is how the sound world of music brings out the “sound world” of a poem. In keeping with this, Robert Snarrenberg focuses on the declamation of poetry in general and prosody in particular: where the ‘reading’ of a poem by the composer means not only ‘interpreting’ but also quite literally a reading out loud before setting it to music. Another 43
purpose of these three authors is to show how composers draw attention to particular textual elements and why. This way of studying the compositional process of Lied composers is interesting to pursue, as it differs from the traditional emphasis on lyric content that has basically defined the field of song studies.
Goethe’s ‘Nur wer die Sehnsucht kennt’ has not been studied systematically in this way, although McClain is an important point of departure for the present study with her focus
41McClain, pp. 211-212. 42Randel, p. 32.
on text dependency and comparison. My own method is generally inspired by all four authors, but as stated in the introduction, both my method and my aim are different. I look at settings of the same text by many composers comparatively, where I investigate what light such a focus on the ‘material’ of poetry and the musical interaction can shed on a collection of settings across a time span. As mentioned above, to focus my topic, I look more closely at the settings that McClain discusses and include additional settings. First, I will now review three authors thematically, with emphasis on matters that are important for my own project. Secondly, I will set out my own method.
The four main authors all begin their analyses with a discussion of the particular poem under review. Indeed, they analyze the words as a sound structure or sound notation, distinct at first from musical analysis. Rodgers, Snarrenberg and McClain explicitly emphasize that a variety of Lied composers are known to have read poems out loud before setting them to music, as I will dwell on below. Rodgers notes in the opening page of his article that
One would think that analysts of art song would be especially attuned to what I am calling the music of poetry.There is plenty of evidence, after all, that composers of art song were so attuned. Schubert, Schumann, Brahms and Wolf were all known to have read poems aloud before setting them. 44
In a similar vein, Snarrenberg illustrates his main theme at the start of his article: Brahms advised his pupil Jenner that before writing one note of a musical setting, he should
memorize the poem and read it out loud, to attend closely to its declamation and placement of pauses. As described in the introduction, McClain quoted Reichardt and Zelter as following 45
a similar procedure. These examples illustrate a main theme in these studies: the text is not merely a carrier of meaning, but also specific material that can be declaimed. In other words, the poem constitutes a notation that is in itself quasi-musical, where elements like rhythms, words, sounds, accent patterns recur and alternate. Note how this resembles the way that musical motifs and themes are studied as recurring and alternating features in musical analysis.
In addition, reading poetry out loud is a physical performance, where the concrete materials of a poem will be, indeed must be, articulated in recital. In short, there is a range of
44Rodgers, p. 1. 45 Snarrenberg, p. 103.
specific textual elements and materials in poetry (and prose), and hypothetically, in the setting in one way or another. Such analyses of texts that these authors offer are interestingly detailed. They all include comments on the meter, syllable count, various regularities and irregularities in rhyme scheme and line endings. Randel, Rodgers and Snarrenberg go into much more detail than McClain. They also dwell at, for example, punctuation marks and syntactic breaks (including pauses such as enjambments). At a structural level, they look at the number of syllables that appear at every recurring accent, which might alter per line, and the unique moments this creates within the broader sound structure. I mention this to call attention to ways in which poems structure sound that resemble such structuring in music, where ‘unique’ moments stand out against previous patterns, and hence are not studied entirely in isolation but in relation to the structure. One example is when Randel discusses Heine’s ‘Aus meinen Tränen sprießen’ as set by Schumann. The poem’s second stanza’s first three lines have a more active syllabic rhythm than what went before and also shift the prosodic pattern. Schumann, accordingly, sets these lines distinct from the the first stanza, However, in the poem the final line (line 8) of the second stanza returns to the simple syllabic rhythm (“rhythmic relaxation”). Schumann reinforced this return with his setting, which returns a simple pattern that he used in the first stanza. This contrast gains more particular relief because he had shifted the prosodic pattern of lines 6 and 7 in his setting. I provide 46
Randel’s outline of lines 7 and 8 below (figure 1), where it can be seen that 7 is prosodically more complex and the only place in the song where Schumann used a triplet, while 8
suggests a more basic rhythm, which Schumann reinforces with his setting.
Fig. 1 Outline of Heine’s “Aus meinen Tränen sprießen” lines 7-8 (Don Michael Randel)
Another example in this vein, is a line in Schumann’s setting of ‘Wenn ich in deine Augen seh’ that Randel and Rodgers both comment on. The line is unique among the others in that it substitutes a trochee for an iamb and has a mid-way caesura. This caesura—a colon—
would afford a rhythmic irregularity in a recitation of the poem as a whole. Randel writes that
Heine’s text “does not vary the syllabic rhythm, but Schumann’s setting has varied it quite strikingly at moments and in ways that can be thought about in relation to other features of the text”. Rodgers, speaking about the same poem, disagrees with this interpretation. He finds it curious that Randel does not discuss the rhythmic regularities such as the “striking” colon in this line. Indeed, Rodgers argues that varied poetic rhythm is “precisely one of the ‘features of the text’ to which Schumann seems most attentive.” 47
Following Rodgers, I take such ‘features of the text’ and the degree towards a composer appears to be attentive as my point of departure. That is to say, I map out aspects such as varied poetic rhythm that are heard in reciting the poem, and then investigate if and how these match with musical features. On the whole, then, I am more interested in simple correspondences between music and and text, and I believe this makes the analysis more tractable. Indeed, it is congruent with my focus on word dependency, which I take to mean that something inherent in the poem is recognized in the music. Helpful is Rodgers’ phrase “most attentive” [emphasis mine], implicating that there is both a wide range of textual properties and also degrees in being attentive to them.
This concern with meter and rhythm is widespread in all authors, of which
Snarrenberg and Rodgers (Randel to a lesser extent) also pay attention how to pronounce words in speech and what syllables might be emphasised. Such emphasis might run against the verse feet that are used in the poem by convention. This attention to speech rhythms and its flexibilities, helps to study musical meter and rhythm in more detail. For example, Snarrenberg dwells at some length at the word ‘Waldeinsamkeit’. Referring to phonological work, he notes that different syllables of this word can be stressed. Both ‘Wald’ and the first syllable in ‘Einsamkeit’ invite to be accentuated: this stress clash can be resolved by
lengthening ‘Wald’, for example. In German speech, such a clash is often avoided shifting a stress to another syllable. In the settings, Schumann and Brahms stress different syllables in fact, (see figure 1), and according to Snarrenberg “their decisions reflect the ambiguity of the word’s phonological structure”. 48
47Rodgers, p. 27. 48Snarrenberg, p. 112.
Fig. 2 Eichendorff’s “In der Fremde,” lines 6b-7, as set by Schumann and Brahms
Another reason that Snarrenberg suggests, is that in both composers, their stress replicates a rhythmic “parallel” that exists with another line. Attention to pronunciation on the one hand, 49
and the between the local and structural is important for my analysis as well.
Whereas Snarrenberg mainly focuses on the placement of stress, Rodgers and Randel also delve more deeply at phonetic sounds and patterns thereof in a poem. That is to say, the appearance of vowel and consonant sounds, what textures they create and how they are physically pronounced. Rodgers goes the furthest in this: for example he minutely maps out occurrences of [a] and [ɛ] vowels. He notes that Schumann “reserves some of his longest and highest notes for the poem’s prominent [a] and [ɛ] vowels.” He also concludes his article with his statement that whereas sometimes Schumann “remained true to the poet’s song, echoing it with his own, but just as often he managed to uncover hidden strains of poetic music, revealing beautiful phonetic patterns that might not be heard without the aid of the composer’s art”. 50
I will return to the question of phonetic sound in the second part of this chapter. But in general, as I will explain below, I take as a point of departure the affordances of the text, and then study the composer’s interaction. Conversely, I do not look at how the composer is as it were a director of our experience of the poem, let alone “uncover hidden strains of poetic music”. That is to say, a particular texture of phonetic sounds in the poem is made more explicit by the music, by emphasizing and amplifying a constellation of sounds that is rather implicit in the poem. I believe this aim is not relevant to my research question and a bit too esoteric. Recalling the previous discussion syllabic rhythm, I instead look more at
49Ibid.
interactions, or where the composer is indeed “true to the poet’s song.”. As will become clear when I set out my specific approach to studying the words, I do include individual phonetic sounds and phonetic patterns within a single line pair (for example, a sound is repeated often, e.g. internal assonance). Of course some textual feature might not be supported by the
composer’s music; this, however, does not mean that the composer is not attentive to the words on the whole: he as it were ‘negotiates’ with the words.
Hence I suggest that the poetic material can be properly seen as affordance, a concept introduced by James J. Gibson. That is to say, the linguistic material of the poem and its 51
structures might afford (make possible) compositional choices, or direct them in a certain way. The affordance concept is often explained with reference to a chair. A chair in the environment, by dint of its form, ‘invites’ one to sit on it. That is not to say, however, that you cannot possibly stand on it. Indeed, such objects are not absolute properties of the environment, but can be suited to various needs or not. For example, a twig of a tree affords support for a bird, but not for humans.Analogously, textual features are half-normative, as they can reasonably give direction for composers. Ultimately, the composer’s musical
reaction to the words can sound as reflecting the poetic sound structure. However, the present project leaves open how composers do that, which is what I aim to investigate.
The three authors under review do not use the same analytical questions for each song they analyse. The initial comments on the text are not always used as a test case, but rather as a first general overview of the poem’s sonoric shape. Indeed, sometimes new textual
observations occur in the course of their discussion of the actual musical setting: in this sense their discussion feels exploratory. Conversely, I hope to investigate whether such
observations are unique to single composers or songs, or to what extent generalisations can be made about a broader group of composers. Can we ascertain whether compositional choices are inspired by textual features, or, to put it differently, is there a correlation between textual details and the way composers respond to them? The purpose of my method is to shed more light on this.
Structure of my analysis
Now in this second part of the methods section, I will more concretely describe my
investigation of the poem’s setting. Each analytical chapter of my thesis focuses on one line
51 James J. Gibson The Ecological Approach to Visual Perception. New York: Psychology Press,
pair, that is to say two verse lines and their musical setting. I do so because these line pairs form grammatical units, often a single sentence. What is more, sets of two lines constantly alternate six with five syllables. Each line is rather short on its own. By studying units of two lines, I can anticipate parallels with other line pairs.
I begin each chapter with a short overview of the line pair. This contains an outline of the meter and its possible ambiguities, and I note structural relationships with other lines or line pairs, which I will tell more about below. I mention significant textual details such als caesurae, punctuation marks, enjambments. In addition, I map out what phonetic properties stand out in this line compared to other lines. Then the next step is to examine the musical settings of the line pair comparatively, in which I analyse the vocal line only. My aim here is to pay attention to how composers respond to the textual details that I have outlined in advance. This response is framed around what Rodgers calls “support” and Randel “fit”, that is to say when musical sound and poetic sound conjoin. As I said, this implies that a negative answer is possible too in this respect. Such a negative answer must be taken into account as well, since it bears on my research question. I investigate how forms of music—musical
rhythms, musical meter, melody—in some way match textual forms or not.
To this end, I present my analysis under four headings, each representing a different theme. These themes are: accentuation, structural relations, structural differentiation, phonetic sound. The discussion of themes in turn constitute different subheadings in the chapters for each line pair. For each theme, I generally try to answer the following cluster of questions. Can a shift in compositional approach be recorded over time in the settings of the poem, and if so, where does that lie in? Conversely, what does remain the same across settings? To what extent do earlier composers adjust more to the course of the text? Can I map out the impact of textual details: and, in turn, when looking at a range of composers, can this qualify the idea of a development? Can this shed more light on McClain’s idea of independence: what causes such independence more precisely? In what follows, I will clarify the four themes in turn, and give an explanation of how each theme relates to the aim of the present project.
Accentuation
In the first theme called ‘accentation’, I analyze to what extent the layers of poetic meter and musical meter align. That is to say, is there any alignment in the accented and unaccented sounds between, on the one hand, poetic verse, and, on the other hand, musical meter? Conversely, these layers can also conflict in settings.
These terms—alignment or conflict between such layers—are used in an 1817 article by Hans Georg Nägeli, who foretold a “new epoch in the art of song” and a “higher style of Lied”, that would feature “an as of yet unrecognized polyrhythm, so that the rhythm of speech, singing, and playing will be subsumed into a higher artistic whole”. Indeed, Nägeli 52
correctly presumed that the Lied was evolving from Volkstümlichkeit, where the layers of poetry and music are customarily aligned, towards complexity, where they become more independent and even will conflict. This description strongly echoes McClain’s notion of a development, indeed “evolution”, towards complexity and independence. I take up Nägeli’s observation to make McClain’s historical sketch about an evolution towards independence concrete: that is to say, I use his analytical angle—independence of layers—to study a possible
evolution.
One layer found in a (recited) poem that I use from this discussion is meter. Indeed, building on Randel, Rodgers and Snarrenberg, declamation of poetry is a basic point of departure, where I look to what extent declamatory schemes are reflected in the settings of verse lines. I highlight declamation, because when a declamatory trait is reflected musically, 53
the music will to some extent keep representing the delivery or diction of a text. In this sense the distinction between music and poetry is less marked: this invites us to speak of “equal footing”, instead of the music being merely in control.
To study such alignment, I take seriously the possible ambiguities of meter in the poem. Conversely, I call some accents in the poem unambiguous. Two factors decide whether I call an accent unambiguous. First, there is no discussion about the accent in the literature on this poem. Second, leaving this syllable unaccented would sound very strange in speech, phonologically. Both these unambiguous accents and more ambiguous ones are explicitly mentioned in the overview of each line pair. As a synonym for an unambiguous accent I will use the term ‘main accent’. In the background lies what the most straightforward rhythm fitting for the time signature would be and, secondarily, how verse feet can be distributed straightforwardly in measures. In the main, I investigate here: to what extent the vocal 54
melody employs accents that are analogous to poetic accents? This question must answer to what extent the poetic meter still directs the melody in the text-to-tune settings or,
alternatively, how they diverge. This examination entails the following questions: are
52Malin, Yonatan. Songs in Motion: Rhythm and Meter in the German Lied. New York:
Oxford University Press, 2010, pp. 29–30.
53Cf. Randel, p. 34.
unambiguous accents allocated to a stressed beat, either the first downbeat or second stressed beat? Conversely, how are stressed syllables alternated with non-accented syllables? Are non-accented syllables allocated to particular pieces of the measure, most straightforwardly to an unstressed part of the measure?
Throughout, the following question remains implicit in my examination: to what extent are later settings of the poem “irregular” as to the course of the vocal melody? With this frame in mind, I search for developments or other patterns, discovering whether a development or not as to independence of the composer from the poet can be mapped out in this respect.
When I analyze the treatment of poetic accents, then, I zoom in on musical accents, that is to say how those accents are accentuated musically. For this, I focus on a range of possible musical accentuation: I account for registral, agogical and dynamic accents and constellations thereof (called a peak by Snarrenberg) in each accent under review. These are 55
easily quantifiable parameters that all authors use.
In addition, I include poetic caesurae found in the lines under review. Obviously, they often occur in tandem with poetic accents and such caesurae have a significant impact in the declamation of a text. For a composer, in turn, there are various ways to set a caesura to music or neglect it (that is to say, to what extent composers have no attention to “syntactic breaks” as McClain suggests). For example, one way to set a caesurae to music is by lengthening the tone before the pause instead of the pause itself, which Snarrenberg calls “little caesurae”. 56
In other words, I focus on concrete musical parameters while retaining openness in looking at how accents and caesurae are treated. Ultimately I will discuss important
observations about common features (or the lack of them) and developments (if any) found in the selection of settings with respect to these accents. Indeed, for all themes, I will frame my discussion around the most important observations, not necessarily pick out every detail in my report.
55Snarrenberg, p. 107 56Ibid., p. 106; cf. 327
Structural correspondences
After having dwelled on key accents in verse lines on a more local level, the next theme is about poetic structure more broadly conceived. The aim in these sections is to look at parallels and structural correspondences that pre-exist in the poem. That is to say, concrete places where lines parallel each other strongly and textual elements recur. How do I decide where such structural parallels are found? Lines can be formally related in having the same implied meter; because they are positioned at a similar point in the stanza; they share syllable counts and or subdivisions in syllables, particular line endings (end rhyme) and so on. I have to clarify why I relate lines, but they are readily shown to be rather straightforwardly related by formal and prosodic elements in the poem. In my initial discussion of the line pair, I have explained in advance significant structural relations with other lines or line pairs.
Thus, by analyzing the musical settings comparatively, I examine to what extent the music echoes such structural relations that pre-exist in the poem. Again looking at the standard set of musical parameters, to what extent are those related lines similar in the
musical setting? Are such parallel lines declaimed similarly? Are rhyming words echoed with musical rhyme, for example the same cadences at places with end-rhyme? Is a repeated word set in the same way elsewhere? If yes, this suggests dependence, because poetic forms are 57
reflected in the music. When the same melody is used for different words, I would like to speak of a ‘pattern’. I aim to investigate to what extent such a pattern really “fits” the words. Or does the composer implicitly say: I am going to use this pattern, no matter what? If so, I argue, that this would signal independence. An additional question then, is where such use of patterns can be located, not just in single instances but possibly across a timespan. McClain sketches a schematic image that in the course of time, settings would articulate the form of the poem (division in stanzas) less straightforwardly, ultimately culminating a continuously through-composed setting. It is interesting to look at this in more detail.
To what extent do composers tie back with their music to textual details that form some kind of structure? Are such textual details echoed in the music? Are later composer less attentive to these structural points?
Structural differentiation
The next theme is called structural differentiation: the complementary reverse of ‘parallelisms’ between lines and line pairs discussed above.That is to say, now that the structural parallels have been mapped out, I can closely compare lines and line pairs as a foil to see at what places in the text important differences pop out, for example an imposition of a comma or other punctuation mark in an otherwise structurally similar line.
In addition, I pay close attention to verse feet, how words are subdivided in syllables, what other textual details elements occur. To do this, I first ask the question how does the line pair under review contrasts with the previous line pair? For example, when this new line or lines in the poem contrast in declamation, will the musical setting contrast as well? Again, the key issue is whether shifts or patterns can be recorded across the timespan (or not). If there is such a musical contrast, in what metrical or other textual property of the line in question is this contrast found? Focusing on the same straightforward parameters, I analyse if and how the composers respond to those differences.
I can briefly illustrate this analytical lens with an example from Randel. He showed that in one strophic setting, where the music repeats for each two stanza, parallel lines in the second stanza differs in the poem: they have one fewer syllable and accent. Consequently, a small “adjustment” was made, namely an unique non-syllabic treatment on these linest. If textual properties such as ‘syntactic breaks’ or metrical patterns influence the setting, it can be assumed that the text guides the setting and that the music cannot be wholly independent. 58
Indeed, the frame of ‘adjustment’ clearly runs counter to McClain’s idea of “independence”.
Phonetic sound
The last theme concerns phonetic sound: that is to say, the category of sound units in
language, such as vowel and consonant sounds. Traditionally, in musical analysis of the Lied, there has been little attention to these aspects. Here Randel and Rodgers provide an important contribution: they believe a composer can carefully respond to those sounds as well, by either capturing or enhancing them. For this reason I will take up some of their observations as a test case. However, the present project cannot take everything into account. For example, whereas Rodgers may map out every ‘a’ and ‘e’ vowel across the entire Lied, I limit myself to patterns found in a single line pair. Otherwise, comparison with other composers is extremely hard.
Since this project places much emphasis on declamation, I zoom in phonetic sounds as they are performed in the reading of a poem, indeed requiring physical gestures. Some sounds or sounds in succession are more effortful, for example when they require the mouth to open and in pronunciation. The main questions I ask are the following: Are phonetic similarities at points reflected or “reinforced’’ with melodic similarities at these points? 59
What happens to vowels that resemble an exclamation? Within the line pair, do particular 60
sounds fall in a specific register?
Can some melodic choices be explained by the way of producing vowels? What happens when lines have a particularly high phonetic density, that is to say direct repetition of phonemes, including assonance and alliteration. Is such monotony reflected in the music? 61
By dint of being effortful, are such phonetically dense lines set differently, for example with broader rhythms ? By contrast, does something significant happen with sudden appearances 62
of a new sound , or when phonetic patterns progress to less effortful in comparison ? 63 64
In addition, I include other phonetic properties that stand out in a line pair, such as sharp, plosive sounds, etc. Do these influence the setting? In addition, I note the use of diphthongs, where the audience might expect gliding vowel sounds in speech. Such 65
compositional techniques are relatively tractable to analysis and comparison. In short, I focus on what distinctly stand outs, and ask whether these can reasonably tied back to the concrete parameters employed by the composer.
After this analysis, I hope to have gained more insight in the handling of textual details by a range of composers, and whether it is possible to say more about the composer’s
independence from to the poet and a chronological development (or not) in the settings under review. In addition, I aim to determine the viability of the methods of the authors reviewed above. Now I turn to the analysis of each line pair, starting with the first.
59Rodgers, p. 15. 60 Ibid., p. 23. 61Randel, p. 35. 62Randel, p. 43. 63Rodgers, p. 9. 64Randel, p. 44.
65As does Rodgers, I consider the first sung vowel in the vocal melody of the diphthong as being
Analysis
Mignon (lines 1-2)
Nur wer die Sehnsucht kennt,(Weiss) (was) ich leide!
Text
The first two lines constitute a self-sufficient sentence, with a comma after ‘kennt’. Regardless of whether we interpret the meter as iambic with trochaic substitution, or dactylic-trochaic, Nur / Sehn / Kennt / Lei- will be accented in recitation. These are the unambiguous or main accents of the poem. In the view of both of the above mentioned interpretations of the verse feet, ‘weiss,’ would also be accented. This would in effect make 1-2 the only line pair that starts with a trochee or dactyl in both lines. However, the word ‘weiss,’ could be slightly more ambiguous, as will pass in review.
As to phonetic sound, there is relatively much alliteration: ‘Sehn-sucht’ / ‘Weiss-was’.
What is more, ‘sucht’ and ‘kennt’ produce three plosives in short succession.The segment ‘Sehnsucht kennt, weiss,’ is relatively effortful: it has many different sounds in short succession.
Accentuation
Most composers place Nur on the first downbeat of the first measure. Alternatively, in Schumann’s setting, the first line begins after a quarter note rest that opens the first beat of the vocal melody; here the word ‘Nur’ will be accented in practice. The principal exception, thus, is Hiller, who places a eighth note rest at the beginning of the beat, where ‘Nur’ is not metrically accented and short, unlike Schumann.
It is important for most composers across the timespan to stick to a declamatory rhythm for the first three monosyllables: ‘Nur’ is long and ‘wer die’ are shorter. For example, Schumann effectively gives a quarter note for ‘Nur’ and uses the remainder of the 3/4
measure for ‘wer die’, straightforwardly in two eighths. Indeed, a basic ‘long short short’ rhythm for ‘Nur wer die’ is distinctive for most composers: Reichardt and Tchaikovsky use a precise one and two halves-division. Beethoven, Schubert and Wolf employ a slight
hierarchy: the word 'wer' is longer than ‘die’.
words. Thus the use of such a figure might be some kind of independence from the text. These patterns have a function however. Hiller's metrically ascending line functions clearly as a preamble to the next strong accent Sehn. Jenner also makes a broad gesture to 'Sehn' with an arpeggio, including a fourth leap towards ‘Sehn’. As literary scholar Terence Cave notes, Schumann picks up a motive from the opening figure of the vocal line for ‘Kennst du das Land’“with the fourth stretched to a sixth, and then inverted in the second phrase and again at the final reprise of the line ‘Nur wer die Sehnsucht kennt’” Indeed, as the interval is 66
altered, I would suggest it is mostly a rhythmic similarity, and that a similarity between ‘Kennst du das Land’ and ‘Nur wer die Sehn-’ in speech sound too, with its dactylic rhythm and then a leap, enhanced by the music, to intensify the next accent of the expressive word ‘Sehnsucht’
In addition, it is interesting to note that such a leap to ‘Sehn’ is significantly found ‘late’ composers Schumann, Tchaikovsky and Hiller. Indeed, only Reichardt, Schubert, Beethoven and Wolf reach ‘Sehn’ by step. Significantly, this word falls on the first downbeat of the new measure, and has a relative agogic accent everywhere. Consequently, it can be seen that the word ‘sucht’ prepares an accented beat reserved for ‘kennt’. Here the interval between ‘sucht’ and ‘kennt’ is most often a stepwise descent; only Reichardt and Beethoven use an ascending step, and Schumann uses repeating tones. It is interesting to note that, conversely, ‘Sehn’ was reached by an ascending interval by all composers except in Wolf’s descending scale. I will return to this in the discussion of phonetic sound.
As expected, the word kennt most often falls on a strong beat then, but not always on the first downbeat.
The first caesura, that is the line ending of line 1 (‘kennt,’) is worth dwelling on further. Reichardt cannot articulate this line break in his setting: the melody unfolds without any musical breaks in eighth tones. Thus his distinctive direct eighth note continuation is surely strange from a declamatory perspective. Other composers do reflect the caesura effectively. For instance, Beethoven, Schubert and Wolf place a dotted quarter tone on the second strong beat, taking up half a measure. Indeed, they use this same pattern, which could indicate influence. Schumann writes ‘kennt’ at the second beat, preceded by a dotted note and short note which practically accentuates the word (some kind of mazurka rhythm); no less than two
66Terence Cave. Mignon's Afterlives: Crossing Cultures from Goethe to the Twenty-First Century.
quarter tones of rest follow, which makes ‘kennt’ ring out as well. Similarly, Tchaikovsky also places a rest after a long tone on ‘kennt’.
What is key to note is how these rests heighten the sense of pause after the line break, where it might reflect the poetic structure instead of continuous prose. There is much
variation in treatment of this caesura, that is to say how it is reflected musically. All
composers (excluding Reichardt), however, have a particular idea for this caesura or at least accentuate it.
Importantly, for Reichardt, ‘weiss’ does not sound like a caesura at all in his setting either. This raises an additional point: since the next word ‘was’ is stressed in Reichardt’s setting of line 2, could this word also be accented in a reading of the poem instead of ‘weiss’? In speech, this would make for a subtle semantic difference (weiss was): the attention shifts more to the amount of suffering. It is hard to say that it is not possible, but it would neglect the comma in the poem as a midway caesura.
But it is important to note that the composers across the time span do differ at this point. For example, Beethoven and Schubert place ‘weiss’ on the first downbeat, which mirrors the preceding caesura by using the same note value, a dotted quarter tone. Schumann mirrors his employed note value and place in the measure—the second beat—as well. Wolf makes this caesura in effect longer than ‘kennt’ and is followed by a rest. After this eighth note rest, the words ‘was ich’ constitute a syncopation on the strong beat, closely mirroring a declamatory rhythm.
Other ‘late’ composers–Jenner, Hiller and Tchaikovsky–have diverging treatments as well. Jenner places ‘weiss,’ on a downbeat, but with no particular accent, other than being the highest note of the second line. Hiller puts ‘weiss’ on the second strong beat and it is also accented agogically: a quarter tone tied over to another quarter tone. In contrast, Tchaikovsky makes 'weiss' short, where he has separated the word ‘kennt’ and weiss with a quarter rest: 'weiss' falls on the second beat of the measure (cf. Reichardt) and prepares as some kind of upbeat pattern for the final word of the line pair. Remember the pattern of Hiller, where Tchaikovsky has effectively prepared the broadened word ‘leide’ with a scalar ascent. I do not see a development of independence from the text so much, as many differing treatments that can be tied back to different aspects of the text.
There are some more commonalities to note, however. In general, the composers with ternary meter tend to place 'weiss,' on the first downbeat. Indeed, line 1 can be divided into
