I feel like I should pause to comment that the "trusting the text" line is a tongue in cheek reference to something that one of my professors, Dr. Paul Menzer, said last semester. I will here paraphrase: when someone says "trust the text," you ask "which text?" "what do you mean by 'trust'?" and "what do you mean by 'the'?"
Pursuing a degree in Renaissance Literature in Performance (i.e. Shakespeare) means being caught between two worlds. On the one hand, we have a deep, abiding respect for the text: the verse, the prose, and the stage directions. On the other we have a menacing distrust for "the text." Influenced by the likes of Gary Taylor and his generation of bibliographers (the new new bibliographers?), we recognize that the texts as we know them are a collaborative process that carry with them the influences of the printing houses in which they were first printed.
"Text" is a product of the technology it is disseminated in, and in the case of The Merry Devil, that means six quartos over the course of roughly 50 years. The first in 1608, and then in 1612, 1616, 1617, 1631, and 1635. The 1612 and 1616 quartos are functionally reprints of the 1608 quarto, and in perfect honesty I'm not far enough along in my editorial process to talk about similarities and differences between the others. Keep in mind that this is before typewriters. To print a text, someone had to manually set the letters, cover them with ink, apply paper, and press, and you had to do this for each page. Time is money, so you may have to do it quickly, and your typesetter may be more or less experienced. In addition to the typos that a typesetter might make, they would sometimes need to change a word, or the layout of a word, to make sure it fit on the page. They may interpret a series of prose lines as verse, or vice versa.
The product of an author thus passes through the filter of the printing house. Even so, our post-enlightenment conception of "the author" (to borrow a phrase from Dr. Tom Berger, our scholar in residence) does not necessarily apply to the Renaissance London theatre scene. Several writers may commonly be hired to develop different scenes of a play from a single plot outline, the quicker to have a work ready for the stage. It was also a common practice to hire a writer to revise the work of another: remember, when Mr. Henslowe hires you to write a play, he gives you money and you give him a play; that is now Mr. Henslowe's play, not yours.
All this is to say that "the genius of Shakespeare" owes much to the media through which it is transmitted to us, as does "the smarts of whoever wrote The Merry Devil." Trying to look through that to some magical authoritative hand is pointless and inane. A fool's errand.
For the moment, I'm not trusting any text. I'm starting with a transcription of the 1608 quarto; my "control text," and line by line comparisons of that to the others will yield insights on how to best prepare my own text. Ultimately, I want to be able to give my actors a text so that when they ask me "which text," I can very simply tell them "the one I gave you." That text might not even be the equivalent of the one that I would offer up as my own "new critical edition;" it may simply be a derivative of that edition that I find to be slightly more performable.
If you think that's a contradiction... well... don't. We know, for example, that Ben Jonson published some of his plays, advertising that he had included scenes that the players had cut. The copy of the text that is sold in the book shop for reading and analysis need not be, and maybe should not be, the same as the one that actors perform on stage.
At the moment, this text is too young to trust; not even long enough to verify.