In the terminology of Kenneth Burke, the underlying story of a narrative is basically a matter of motivated actions. More specifically, in the framework of Burke's pentad, acts are performed by agents, who have purposes. Every act takes place in some scene, and its performance may rely upon the instrumental assistance of agencies. However, scenes and agencies tend to supplement the basic questions of who does what and why.
One of the greatest hazards in revisionist productions, particularly when a war-horse grand opera is concerned, is that the secondary overwhelms the primary to a point where the primary becomes at best insignificant and at worst ludicrous. This is precisely the problem with Marta Domingo's staging of Giuseppe Verdi's La Traviata for the Los Angeles Opera, currently being performed by the San Francisco Opera. As she explained in a one-page "note" in the program book, she wanted to transplant the tragic tale of Violetta Valéry from the "world of the demimondaines" of late nineteenth-century Paris to the "world of the flappers" during the Roaring Twenties of the United States. This may make for good eye candy, but it tended to undermine the characterizations of the key agents (Violetta herself, her would-be-poet lover Alfredo Germont, and his bourgeois father Giorgio) and their underlying sense of purpose. That undermining may explain why, in reviewing this production for the San Francisco Chronicle, Joshua Kosman saw it as a story about Giorgio in which Violetta played a supplementary role. By subordinating acts, agents, and purposes to matters of scene, Domingo offered a topsy-turvy discourse for the story.
The success of such a radical approach involves a "social contract" under which the audience is willing to accept the terms of the director. Because Dwayne Croft put so much of his own sense of character into his performance of Giorgio (as both actor and singer), we in the audience certainly had reasons to accept this contract. On the other hand the two party scenes, where the eye candy was at its richest, kept bumping over potholes of inconsistencies with the libretto, some of which were deftly (or not) paved over with rewordings in the projected titles. Ultimately, the most successful scene was the final act, in which (with the exception of a choreographic interruption that was as inept as it was inane) everything was stripped away other than the bed on which Violetta would expire. Perhaps, then, it is no surprise that this was the act in which Anna Netrebko's Violetta and Charles Castronovo's Alfredo rose to the same vocal heights that Croft had established the first time he set foot on the stage. Working together (and with conductor Donald Runnicles), the three of them made magic out of music that can too easily fall into cliché; but why did we having to slog through so much muddle before finally encountering the operatic experience that made it all worth while?
This may come down to support for the argument that, in a time of economic austerity, it is necessary to go back to the basics. That means concentrating on the agents (as both actors and singers) and that "sense of purpose" without which the acts lose their meaning. The best way to avoid missing the opera altogether is to keep the primary in its primary position.