I must admit I absolutely hated the work, both the words and the music. I do not consider myself a prude, but a woman with a beard I think goes too far. The early box office was most encouraging... but later performances were played to many empty seats and in the next year the subscribers were furious -- people literally spat at the box office, and we couldn't give tickets away. We offered to send heated cabs, but nobody would come.
As the above passage implies, there was quite a lot of hostile reaction from audiences at the time. What seems weird about that, from today's standpoint, is that The Rake's Progress is one of the most musically accessible operas written after WWII: As the culmination of his "neoclassical" period, Stravinsky and his librettists (W.H. Auden and Chester Kallman) made a deliberate return to the form and style of 18th and early 19th century opera: individual numbers separated by harpsichord-accompanied recitative. Ever since Wagner, the ideal of opera had been to make an opera seem through-composed, with no clear distinction between aria and recitative; Wagner did this by writing whole operas in what was basically a heightened form of accompanied recitative. Stravinsky revived all the formulas that Wagner had supposedly rendered obsolete: relatively simple melodic lines, songs that are carried by the voice and in which the orchestra has a subsidiary role; even – in Anne's aria – the old 19th century Italian form of a slow cavatina followed by a fast cabaletta. Stravinsky was also reaching back to the 18th century ideal of making musical points in simple, short terms. Wagner had come up with all sorts of complicated ways to make music express ideas and portray physical objects; a gigantic orchestra, new and strange harmonic techniques, the leitmotif system. Stravinsky in Rake chooses older, simpler techniques for making music express something concrete. At the climax of Tom's first aria, when the words refer to horses ("Come, wishes, be horses/This beggar shall ride"), Stravinsky doesn't introduce a "horse motif," nor does he try to find a musical equivalent for the metaphor that the words are expressing; he just has the orchestra play some very simple melodic figures that are suggestive of horses, and then speeds it up a little when the words refer to "riding" those horses. It's the kind of very basic, simple representational music that Haydn did in The Creation and The Seasons, and which was commonly dismissed as hopelessly naive and silly by Wagner's era.
Of course, because of all this calculated simplicity, the musical avant-garde condemned Stravinsky as hopelessly out of touch; that, along with the failure of Rake, is part of what led him to start writing serial music in the '50s. But one would have thought that an opera with simple, attractive melodies would have appealed to general classical music audiences, and to the majority of music critics (who generally were anything but avant-gardists). Instead the Rake's Progress got lukewarm box-office and reviews that ranged from ambiguous to outright hostile. Why?
One problem may have been that Stravinsky and Auden and Kallman were harkening back to a kind of opera that was not very popular at the time. In the late '40s and early '50s, Mozart was still not a major part of the standard repertoire, and when his operas were performed, it was often in camped-up versions that tried to smooth out the ambiguities of tone -- so opera houses would present Cosi Fan Tutte as a pure featherweight farce, and Don Giovanni as either a gloomy tragedy or a campy sex romp; the thing that defines Mozart operas for many of us today, that they are serious and funny at the same time, was considered a flaw. The most popular operas were pure comedies (Barber of Seville) or, especially, pure tragedies, with direct emotional appeal. The biggest inspiration for The Rake's Progress is Mozart's Cosi Fan Tutte, with its shifts between farce and seriousness, and its shifting signals as to how we're supposed to respond to the characters. But that was the kind of thing that made Cosi a hard sell at the time. Stravinsky was bringing Mozart's ambiguity and irony back to opera at a time when audiences weren't accustomed to finding such things in an opera.
Conrad L. Osborne, with his usual brilliance, summed up another part of the reason for the mixed reactions when he wrote (reviewing a recording of the opera in High Fidelity, January 1965):
The source of our ambivalence in feeling towards the opera (I say "our" and mean "my," but apparently it's a widespread reaction, to judge from the quizzical stutterings of most critical response) lies in the fact that while the characters are placed in the perspective of fable, and the form is carefully calculated to keep our minds working, Tom emerges as a person, too. And while all the operatic hocus-pocus is being pointed at and even ribbed, it is working the old tricks on us; it uses the old forms for their own sakes, as well as for purposes of framing. This leaves us at something of a loss. Should we laugh or cry? Snicker or cheer? Remain detached, or give in?
This problem -- the ambivalence about how to respond to the opera -- was exacerbated by two things. One was Stravinsky's tendency to say, in public and in print, that his music was supposed to be unemotional, unrelated to anything but itself, etc. It's hard to know how seriously to take such statements, but in any case, critics did take them seriously and came to Rake assuming in advance that we were supposed to distance ourselves from the music and the characters. The other problem was that Stravinsky was paying tribute to operatic techniques that, for the most part, were no longer taken seriously, and that therefore seemed to send out a signal that the whole piece wasn't to be taken seriously. The thing about The Rake's Progress is that while Stravinsky's use of these musical anachronisms may sound, at first, like parody, it eventually becomes clear that he's not always using them ironically; he's just using them because they work, musically and theatrically. Osborne continues:
When we hear the harpsichord, we are immediately conscious of the anachronism -- that is unavoidable. Yet Stravinsky has chosen the harpsichord not merely for its status as an antique or even its functional value accompanying secco recitative, but for its descriptive coloration as well, and in this role it creates a chilling effect in the graveyard scene. When Anne launches her cabaletta, we are aware that a fine tradition of romantic opera is being sent up, as the English say; but we also respond to the genuine impetus of the writing, and wait for this high C as impatiently as for any in Verdi.
There is, as Osborne says, a problem with this bouncing between using musical devices for ironic effect and using them for their own sake; sometimes we may feel, as he puts it, that we're "being toyed with." But it's not a major flaw in the opera. What made it seem like a major flaw is that audiences and critics weren't used to taking these devices seriously, so Stravinsky's use of them was interpreted as pure parody. Evaluated just as an aria, Anne's aria is a very effective piece; but because it's cast in the form of a 19th-century Italian opera aria, and because that form was considered to be discredited by that time, it was hard to hear it just as an operatic number, rather than a spoof of an operatic number. This he-can't-be-serious attitude applied to a lot of Stravinsky's "neoclassical" pieces: Oedipus Rex, his great opera-oratorio, is filled with very sincere homages to Verdi, whose music Stravinsky loved; but few critics took Verdi seriously at the time Oedipus premiered, so it was interpreted as a bitter parody of older music. Similarly, Stravinsky's ballet The Fairy's Kiss was a loving homage to Tchaikovsky, but many critics called it a spoof or criticism of Tchaikovsky, because at that time, Stravinsky was held in much higher critical esteem than Tchaikovsky was, and few could seriously entertain the thought that the most celebrated modern composer had anything to learn from the composer of Swan Lake.
The problem with applying that kind of analysis to The Rake's Progress is that it's just too long to be a pure parody. Oedipus Rex is only 50 minutes, so if you interpret it as a big nasty joke, it can just about sustain that joke. But Rake is a full-length opera; unless you lose yourself in the music and become engaged with Tom as a character, it doesn't work. Part of the reason Rake has gotten more popular, I think, is that we are more willing now to accept the greatness of the older composers Stravinsky was paying tribute to -- Mozart, Haydn, Handel, Verdi -- and therefore more willing to hear the work as an opera that happens to use some old-fashioned operatic devices, rather than Igor and Wystan's big expensive joke.
Most of the recordings of Rake seem to be out of print, but the John Eliot Gardiner recording is still available, and very good; the conductor and singers take the work seriously and play up the emotional effectiveness of the music, rather than trying to make it sound dry and parodic. (The most famous singer in the cast, Bryn Terfel, actually overdoes it by snarling and barking his way through the role of the villain, Nick Shadow, but at least that's a non-ironic, non-parodic take on the character.) There's a DVD of a Salzburg production that is unfortunately not very good -- a "revisionist" production of an opera that's already plenty revisionist on its own. Hopefully a better production will find its way to DVD eventually.