Ostensibly, the central question of Richard Strauss’ last opera is “What is more important to an opera: words or music?” The central irony of this wryly subtle comedy, though, is that by the time the question is posed it has been thoroughly answered in favor of music; it would be hard to disagree either on the basis of the quality of Strauss’ music in the opera or on the basis of the opera’s construction. Compare this to an opera like Elektra, Die Frau ohne Schatten, etc. — these begin to incorporate the text almost immediately to establish the setting, the central conflict, and the characters of the opera — Strauss here establishes that the principal characters are the musical motifs, thus resolving the central conflict rather prematurely.

I hope it will be uncontroversial to say that this is not a satisfactory answer. What’s the point of establishing a question to frame the entire piece in the first scene when it was already answered in the prelude? I would suggest that the answer is not actually quite as simple as it first seems.

In defense of the text

Strauss’ argument in favor of the primacy of words is both textual and meta-textual. Textually, it actually begins in the preface, in which he states that among the highest praises that an opera conductor can receive is that “the singers were audible and the text was comprehensible1.” Strauss accuses opera orchestras in this paragraph of a tendency towards Sängermord2, or the murder of singers; the simplest explanation for this charge is that he feels that the text is something worth protecting as essential to the operatic enterprise, and that it is not merely a vehicle for music. This is reinforced by his brief recapitulation of operatic texts in history, beginning as quite separate from the music3 and shifting to the Wagnerian mode of operatic composition in which they are completely integrated.

It is therefore worth noting that among Strauss’ operas, Capriccio bears the greatest resemblance to a traditional number opera. Sure, the recitative sections are more melodic than your average Baroque opera, but a much more significant portion of this opera could be described as raw text declamation than many operas in the post-Wagnerian style4. The actual content of this text declamation, though, remains a significant source of confusion for anyone trying to definitively answer the opera’s central question.

For starters, Olivier and Flamand (the characters who represent text and music respectively), are joined by a third: La Roche, who both straddles the line between them as well as adds a third possible answer to the question (staging/production). Already in his introduction, La Roche: says music is helpful for sleep (a point against music?) but that audiences are only there to hear the tenor’s high notes and wait impatiently for them (a point in favor?); he describes a production as overflowing with learned music5 (plus?) but says that no one can hear the text over the tumult of the orchestra (minus?). He describes Italian arias filled with great music and enchanting arias (surely a point in favor) but then describes a conversation with Goldoni6 in which the latter describes [French] operas as a “heaven for the eyes and a hell for the ears” in which the listener is left waiting for “arias that all sound like recitatives” (probably a point against?). Perhaps La Roche’s biggest contribution in this first scene is the implication that it’s a matter of personal taste: some audiences prefer the lyricism and beauty of music and some prefer the drama and intellectual engagement of text, while others may value the pomp and circumstance of the production. His verdict in this scene might be summed up as: music is worth the hassle. This scene ends with the introduction of the Count and the Countess Madeleine; the former obviously represents the argument in favor of text while the latter is divided7. Given the predispositions of the other characters, the majority of the cast (diegetically speaking) falls on favor of the text side of Strauss’ divide.

Speaking of waiting for arias that sound like recitatives, there is a quite significant (by operatic standards) dramatic scene of spoken text before anything that might be called a proper aria. The first meeting of the Count and the actress Clairon is a highly effective dramatic scene and expresses a depth of feeling between them that is quite touching. This ends with the Count’s recitation of the sonnet central to this opera, which in this context sounds like a heartfelt profession of love. Clairon responds to this by resuming musical declamation, which cannot help but feel as if she is trivializing the situation and brushing off the Count’s feelings rather than really engaging with him. Olivier’s rectiation of the same sonnet later in the scene8 reads as a more melodramatic affair, especially with the addition of a harpsichordist who feels compelled to improvise an accompaniment to Olivier’s spoken-word utterance. Flamand’s musical setting of the same text is quite beautiful but feels yet another step further towards a dramatic contrivance. It is surely no accident in Strauss’ three framings of this poem (pure speech, speech accompanied by music, and song), the barest presentation of the words feels like the most authentic expression and the most musical presentation also feels the most artificial in its expression. When Madeleine responds to Flamand’s dramatic love song, her first praise is that “the poet’s words shine so clearly” — this doesn’t say anything against the music, but the magnitude of the music’s merit in her mind is still left somewhat ambiguous. In a typical display of Strauss’ ironic sense, I think the highest musical praise for the song in this scene is by Olivier: while he fumes that Flamand has destroyed the structure and beauty of his text, he does so with a wonderfully expressive lyrical line that helps to accentuate the structure and beauty of the aria.

That said, most of the points one can make in favor of the primacy of text are made meta-textually by Strauss rather than specifically enumerated. The first such point is obvious: how could one even raise this question, let alone answer it, through music alone? Absolute music can surely express some emotional states, but it requires some text for the audience to get any engagement from it other than 1) evaulating it on purely musical terms (melodic construction, harmonic progression, form, etc.) or 2) this first-order emotional engagement. In order for music to really “say something” it has to have some kind of text associated with it. Many of Strauss’ instrumental works fit this model; the tone poems are some of the best examples of music in this mold and among Strauss’ most frequently-performed works9. The addition of text allows audiences to engage with the work on a different level. To take an obvious example like the Alpine Symphony, it gives audiences something to follow along with and guide their listening while giving seasoned interpreters something else to chew on in forming their interpretations of the work — I don’t just mean musical interpretations here; by some accounts Alpine is a more Nietzschean work than Zarathustra in its text10.

Similarly, the subtitle of the opera (“A conversation piece for music”) works on multiple levels. It suggests the conversation that motivates the post you are reading right now, offers some statement about the form of the piece itself (most of the scenes are just conversations set to music), and perhaps most importantly hints at the “conversation” between the text and the music present in this opera. While there are certainly some supremely well-composed bits of music in this opera, it would be pretty boring without the ironic tension between music and text that underlies many of the scenes. Of the arguments laid forth by the characters of the opera in favor of text and/or music, most of them are directly undermined by the music. Let’s look at a few of my favorite examples of this phenomenon.

Characters who favor text often complain that it is buried by music, that singers can never compete with the bombast and furor of the orchestra, and that the clarity of the text is the first thing to go when it goes on stage. These complaints ring amusingly hollow in this opera. Strauss uses his decades of experience as a first-class arranger to create clear textures throughout that let the singers shine through with no issues whatsoever, except where he purposefully muddles the music to make a point11. The portion of the opera where the text is least clear is without a doubt the fugal ensemble12, but look at the score — bare, transparent scoring that serves to reinforce certain members of the vocal ensemble with little to no competition, except in occasional points of punctuation. What makes the text incomprehensible is not the music but rather the octet setting that overloads the listener by spewing multiple competing lines of text at once. Is this problem endemic to operatic ensemble writing? Sure. But is this the music’s fault? Not really. If anything, the orchestration in the octet helps to underscore and bring out certain lines of text, clarifying them by drawing the listener’s attention to them. The fact that we need a score (or supertitles) to understand that half of the characters are laughing and trivially yapping about food and drink while the other half discuss the question of text or music (with varying degrees of seriousness) is at best at fault of both the libretto and the score. Similarly, we have instances like Flamand’s statement at [136] that “one can experience the whole world in a single chord” ring hollow when set over a simple dominant to tonic resolution in C major. Flamand’s exhortations that ring true in text are often made to sound perfunctory and bathetic in music.

The strongest argument in favor of text is that tensions like these are laid bare by the interaction of text and music, and that the same music with a more straightforward text would result in an interminably boring opera. After all, there are really only a couple moments of music that could possibly redeem this opera in the absence of a good text. The opening sextet, Flamand’s aria, the dance numbers and the fugue, and the final scene are chief among these13, and this leaves well over an hour of empty space that would have to be filled with recitative, which would be dry in every sense of the word.

Counterpoint: ok yeah but really tho?

Perhaps we can put that a little more eloquently: while it’s true that the text is integral not only to developing a proper interpretation of this work and that the interplay between the text and music is what renders the music interesting for the vast majority of this opera, any proselytization to that end is supremely ineffective in the face of a passage like the Moonlight Music, whose transcendent beauty is hardly subject to a polemical refutation. The last scene, beyond being introduced by such a passionate yet serene passage of music, also makes a more interesting case in favor of music. This comes in the form of a stage direction after the final line of text in the opera: Madeleine gives a wry smile and a quick wink to her mirrored reflection. This acts as a sort of fourth wall-breaking “wink to the camera” — she asks “Is there an ending that isn’t trivial?” and then her wink is underscored by an extended musical passage (it should go without saying that this is a passage of beautifully lyrical music). To me this clearly suggests Strauss winking at the audience, saying “I know I’ve set up this whole dilemma and made a convincing case for both, but we all know why we’re really here.” This last scene ends with one of, in my opinion, the funniest moments in the opera: the final five measures of music. These are hardly comedy in the traditional sense (at least of the same sort as something like the Catalog aria), but the music provides an answer to the Countess’ question: “There may or may not be an ending that isn’t trivial, but even the trivial can be beautiful.”

There remains one interesting argument that the opera makes in favor of music: there are points where the music renders the text irrelevant, but never the other way around. As previously discussed, there are a handful of moments where there is speech unaccompanied by music, but these stand in stark contrast to the remainder of the opera. Much like the poignant grand pauses in a Bruckner symphony, these spaces without music remain “musical by omission,” punctuating the expression of the music by occasionally reminding the listener how much the music adds to the text, even if passively so. As for moments where the text is “rendered irrelevant” by the music, I think the best example of this phenomenon is the first entrance of the singers leading to Madeleine’s first entrance (beginning at [8]). Olivier and Flamand both describe the Countess’ beauty with an infatuated fervor, yet their description is virtually immaterial; the listener knows that Madeleine must be enchanting because the music that is “about her” is enchanting to the ear. Strauss is a master of musical characterization: no one could confuse characters like the frenetic Elektra, the clever Zerbinetta or her somber counterpart Ariadne, the heroic Apollo, or the wily Salomé on the basis of their music. Madeleine’s music is elegant and gracious but never really decisive, and the text can only serve to reinforce this.

Having made a reasonable case for the primacy of the text and a less reasonable (yet somehow more convincing) case for the primacy of music, we are left at an impasse. We can accept either answer on a not-totally-satisfying basis or accept that the answer to the opera’s central question remains ambivalent. Does this latter possibility satisfactorily answer it? Personally, I don’t think so. Instead, I think this ambivalence pushes us to seek different questions posed by the opera which might be more conclusively answered.

Towards a better question (and answer)

Though I’m sure there are many interpretations possible of the work, I think to me the second-order question most clearly suggested is “What is the purpose of opera?14” This question has additional weight given the circumstances in which Strauss wrote the opera. In a world wracked by two successive hellish wars, the near-complete destruction of both the cultural millieu and political system of Germany, the increasing perception of traditional musical forms and techniques as some combination of outdated and/or insufficient, and the economic conditions that made presenting an opera difficult15 and put opera far lower on the list of priorities of most of the would-be middle-class audience. Can an opera sufficiently address these circumstances? If there’s any argument to the idea that Strauss thought so, it’s the mere existence of this opera. The setting of this opera in the Paris of 1775 certainly suggests parallels: a crumbling economy and society in the wake of an expensive and burdensome war, on the eve of a revolution that was as much a revolution of thought that sought to incorporate new culutural and philosophical ideas as it was a revolution of government that sought to reform a defunct political system16. This period is squarely between Gluck’s most significant operas and Mozart’s17; one would be hard-pressed to come up with more than a couple periods at which the institution of opera was redefined to a greater extent (the 1600s, 1850s, and 1910s are the only serious candidates in my opinion).

One obvious place to start, at least from the point of someone writing today, is to ask what the opera has to say about the events of its day. Is Capriccio a “political” opera? Yes and no. In the way such a question is usually taken today, the answer would be certainly not, especially compared to something like Friedenstag18. The political content of the opera, as we will see, is the result of hard-fought exegesis rather than being in the text of the opera itself. It also serves as a wonderful example to composers of how to write music that is “timely” or “responds to the moment”: Strauss does not need to make any direct reference to events of his time period for us as the listener to understand that these concerns pervade the work. We don’t even need any text in the work to reach this conclusion: works like the two sonatines for winds or the oboe concerto stand in stark contrast to works like the Metamorphosen or the Four Last Songs19 as possible ways that an artist (indeed, that a person) can respond to circumstances of overwhelming trauma. Where Friedenstag is more like a Wellington’s Victory (perhaps, to be slightly more charitable, an 1812 Overture), Capriccio is more like an Eroica: the work can certainly be read with or as an interesting political statement, but the drama and music also stand on their own without reference to contemporary goings-on.

What, then, does this opera have to say about the world in which it was composed? We can only read this indirectly. Among the most relevant passages is La Roche’s suggestion of possible themes for an opera20. The first suggestion is “The Birth of Pallas Athena.” He briefly explains the plot that such an opera would take, but as a listener nothing about the musical content of this passage inspires any confidence in this hypothetical opera. Ostensibly, a work about a new way of thinking in a tumultuous time could have a certain resonance, but I think this first-order treatment often results in less interesting works than works in which these are background considerations. While Strauss praised Friedenstag at the time of its composition, I suspect this is a bit of an admission on his part that as timely as an anti-war polemic was in 1938, a musical work must be justified by musical substance. The second suggestion is “The Fall of Carthage.” This idea certainly has potential, at least as a more intimate human-scale drama concerning the travails of living in difficult times. Unfortunately for La Roche, his suggestions on how such a work might be executed are more about ostentatious tricks of the stage than how it might address the concerns of the citizens of Carthage (or of Rome, I suppose). This too therefore seems ultimately a futile effort to write a meaningful response to the conditions under which the opera would be composed.

The third and most interesting of suggestions is a series of operatic subjects already executed by Strauss himself. The first in this set, suggested by Olivier, is Ariadne auf Naxos. Flamand responds that the subject has already been too-often composed, and La Roche laments that it lends itself to too many long, tragic arias. While there are a handful of extant operatic treatments of this subject (whether from 1775 and earlier or from 1942 and earlier doesn’t actually change the calculus that much), and there is certainly a tragic element, this is a prime example of Strauss’ sense of humor. His treatment of the subject, whether one takes the original 1912 version or the 1916 revision, is wholly original and not at all over-burdened by tragedy. His version is striking in its unique blend of tragedy and comedy that some scholars cite as an early example of musical post-modernism21. Is Ariadne a sufficient response to the times? I would think so. It stands both as a comprehensive undermining of the separation between Opera Buffa and Opera Seria and as an undeniable affirmation of the power of each. The premise of the opera is somewhat farcical, but it does at its core confirm to the listener that antiquated modes of composition can have a purpose. Strauss recontextualizes Opera Buffa, an erstwhile means of trivial entertainment, as a sort of coping response to trauma, in modern parlance.

The second in this set, suggested by Flamand, is Daphne. Olivier responds that the transfiguration scene is too hard to stage, and the Count gives the curious response that “Gods and Greeks are everyday things.” As for the first response, perhaps this is true from a literal staging perspective, but one cannot help but think that this doesn’t matter given the incredible music that Strauss sets for this scene. From the brutal C# minor chord at [238] through the winding chromatic line introduced by the bassoons, through to Daphne’s ecstatic F# major arrival after [251], it is difficult to imagine a listener who is not sufficiently enthralled by the music that the exact execution of even the simplest staging is a major detriment to this scene. Does Daphne deal with everyday things? Perhaps, but not in a bad way — one of the major strengths of Strauss’ writing (and of Gregor’s libretto) is the focus on the human-scale drama. This is a commonality in Strauss’ best operas; even when there are gods involved, it is relatable human concerns that shine through and render the drama approachable for the audience.

The third subject in this set, suggested by the Count, is the Trojan war. La Roche’s response is apt: “we have too many Trojans, Egyptians, Persians, Judaeans, and Romans in our operas already.” Strauss’ Trojan war opera, Die Ägyptische Helena, largely eschews most of the conventions of this subject in favor of a “side story.” The version of the Trojan war that puts Helen in Egypt is more of a footnote in most tellings, so Strauss’ version of the subject is itself a subversion of the conventional one.

Thus we have three different models for an effective way to write an opera in the present day:

  1. Write against genre conventions to give the audience new perspective on new and old works in the genre alike.
  2. Write an opera in a classic mold but do it sufficiently well to be compelling, even if it doesn’t have anything fundamentally new to offer.
  3. Tell a familiar story in an original way.

Which is Capriccio? In some sense, it’s a combination of all three. It has many elements of Opera Buffa, but it’s more of a drama than a comedy. This seems like it should be a comedy of manners, but most of the comedy is musical and the text is a straightforward, if low-stakes, drama. No element of the text or music is particularly revolutionary; as much as I love this opera, I can’t claim that it contains any elements that are wholly original. Even so, the mere existence and length of this post indicates that there is some degree to which this opera is thought-provoking and/or well-constructed. These two points in conjunction feed into the third: it is a straightforward story in the aristocratic comedy of manners style with no fundamentally new components, and yet the unique combination of all of the elements used in this work still manage to create something that sounds fresh and new (even if it might take a few listens to grasp what’s going on under the hood).

Thus the combination of these factors gives us an answer at least to our sub-question: this isn’t a “timely” opera in the sense that it directly incorporates current events, but contemporary sensibilities and concerns are filtered through a retrospective lens in such a way that it forces the listener to consider not only the construction, message, and value of this opera, but also of opera in general.

Fans of tautology will be thus pleased to zoom out to our higher-level question, and answer it thus: “the purpose of this opera is to make you consider the purpose of this opera.” We must, though consider one further answer: “who cares what the purpose is? The music is great.” As previously discussed, I think Strauss takes this answer more seriously than one might initially think. Despite the petty bickering about whether Olivier or Flamand will win Madeleine’s heart, and the parallel quibbling about whether words or music (or something else) is more important (either to the operatic enterprise or as an art in general), Strauss bookends the opera with passages of music that the rest of the work seeks in vain to justify with drama. In reality, no amount of text is necessary or sufficient to justify Strauss’ composition, which is both the greatest strength and the greatest weakness of the “who cares?” answer. Personally, I find the tautological answer to be the less unsatisfying one, and I’m willing to adopt it.

There remains then yet one question: Is there a conclusion to this post that’s not trivial?

  1. Throughout this post the quotes will be my translation from the standard edition of the score and will cite rehearsal numbers from this edition where passages in the score are relevant. ↩︎
  2. On the other hand, he makes clear later in the preface his stance that a happy middle can be achieved, saying that no solo violin line or pp passage in winds and strings could possibly cover up a soloist. This somewhat ironic tension will feature both textually and meta-textually in the opera itself. ↩︎
  3. Strauss identifies arias before Mozart as “purely lyrical effusions of feeling, during which the action stands dead still.” ↩︎
  4. And certainly more than Daphne or Danaë. ↩︎
  5. Learnèd, to use the antiquated spelling (“mit seiner gelehrten Musik überschüttet“). ↩︎
  6. Goldoni was a real playwright and librettist contemporary to the opera’s setting. It is not clear whether La Roche’s account is based on an actual quote by Goldoni, but I wouldn’t be surprised if that were the case given Strauss’ penchant for history. Goldoni’s operatic works (even a handful of libretti used by Mozart and Haydn) are largely historical curiosities rather than significant parts of the repertoire. ↩︎
  7. Or so she claims. It’s quite possible to read this as her wanting to appear neutral while she obviously favors music internally — we’ll come back to this point later. ↩︎
  8. In the canon of the opera, Olivier has composed this sonnet, though Strauss notes in the score that it’s a translation of a poem by Ronsard. ↩︎
  9. Personally, I think his best instrumental works are those for winds, which don’t fit this mold, but that’s a separate conversation. ↩︎
  10. I don’t recall a citation for this but I believe I first encountered this idea reading Michael Kennedy’s writings on Strauss. ↩︎
  11. Such as at [13], and in many similar passages where the characters make similar complaints. ↩︎
  12. Especially from [166a]. ↩︎
  13. Plus the performance by the Italian singers, which as in Rosenkavalier is expressive only in a parodic sense. ↩︎
  14. If you think there’s a better candidate, please let me know. I’d love to consider it. ↩︎
  15. Recall that the premiere of Die Liebe der Danaë was delayed by nearly a decade until after Strauss’ death. Hitler’s favorite operas, especially Wagner and especially Meistersinger and the Ring got a bit of a carve-out. A full discussion of economic conditions of German opera in the 1940s is far beyond the scope of this post, but Alex Ross’ book on Wagner is a good starting point for a grasp of the relationship between the regime of the Third Reich and the music of Wagner. ↩︎
  16. Of course, these often go together. ↩︎
  17. Strauss singles out Alceste from Gluck and Cosí fan tutte and Figaro from Mozart in the preface, though surely Orfeo ed Euridice and Don Giovanni are also worth mentioning. ↩︎
  18. Which I think is more effective as a political statement than as an opera. It’s worth listening to as a fan of Strauss but perhaps more as a study in contrasting ways to respond to current events than as a study in how to compose a great opera. ↩︎
  19. Let’s ignore that these actually do have text. The ordering and the constitution of this final song “cycle” is a matter of some scholarly debate, which makes for interesting reading for a fan of the music. ↩︎
  20. Beginning at [163], [171], and [220] respectively. ↩︎
  21. I believe I encountered this idea in the essay collection Richard Strauss and his World, though I read a copy checked out from a library years ago and don’t care to track down another copy to confirm this. ↩︎

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.