Here's what I have figured out about it.
1. There's a certain amount of hubris to pulling together a chapbook. I mean, you have to assume that somebody might want to read it, right? And who am I to go about thinking I'm good enough to be read? (Hmm . . . maybe this should be there's a certain amount of self-loathing to pulling together a chapbook?) I feel like I'm constantly having to get past my own self in order to commence work on the project at all. Nevertheless, I push on.
2. I have to pick and choose among my own work. This is difficult, because, of course, on any given day I may like (or loathe) every single poem I've ever written. Also, on the one hand I am trying to select my best work, and on the other hand:
I am thinking that the poems that I consider my BEST work that haven't been published yet ought to get submitted to journals first, where odds favor them being repeatedly rejected, but still, sometimes poems get accepted
I am thinking that a chapbook made up entirely of poems that have already been published someplace wouldn't be that interesting
I can't always figure out which are my best poems, although I have my suspicions
I have poems that I am especially fond of, which may not actually be my "best" poems, that I kind of want to include
poems that seem perfect one day and therefore must be included don't appeal to me after I've let the collection sit for a while
my chapbook is supposed to be, say, around twenty pages, and I have only picked 12 poems for it; or, conversely, I have picked 37, and can't seem to decide which of those ought to go in the collection and which ones ought to go back in the file
other stuff and bother
some or all of the above.
3. Why bother? I might get rejected even after all this to-and-fro-ing. And I might not.
4. Once I have selected the poems that I think ought to be in the collection, I have to put them in some sort of order. Suddenly, I realize that I have 14 wry or humorous poems and four that are super-heavy, sad poems (or some other, similar conundrum) and I cannot for the life of me figure out how to intersperse things without seeming unhinged or, worse, creating an expectation that "this is what these poems are like", which turns out to be a false one, but it's too late, people already put the collection down and walked away. So I take some out and put some other ones in and start over. Or I put the whole thing down and slowly back away.
Or I figure out that some things go together, subject-matter wise or thematically or whatever, but not all of the poems work that way, and then I have to figure out what to do about it. At this point, I resort to pulling down various and sundry chapbooks from my bookshelf, to see how they were organized, and then I fall into reading the poems again, because they are there, and my, they are so very good - are mine anywhere near this good? - and then I start rocking quietly, wondering why it is that I thought this was a good idea anyhow?
5. There is no "right" answer. This is what I've finally figured out, after spending the past two months repeating the actions in the previous two paragraphs. Today, I'm resolved to push past/through it and arrive at an order. Then it's on to the next step . . .
6. A chapbook needs a title. Of course it does. And as many writers will tell you, coming up with titles is hard. Books should be written on the topic, but as far as I can tell, they haven't been. A common practice is to pick the title of one poem from the collection to stand as the title for the work. Even Billy Collins, Kay Ryan, and Mary Oliver do this. Probably I will do this, too. But did I mention that it's hard? Which one title speaks for the collection as a whole? Speaks for me as a poet? I guess eventually I will pick one that seems best (or least bad) and go with it.
7. As has already become evident to you if you've read this far, there's a lot of whining involved in the process. Maybe not for everyone. Maybe, once I have done this more than once, or more than more than once, I will get good at it. Perhaps it will be a speedier, simpler process. I will only find out if I finish this one and move on.
And now, I am off to sort and compile and shuffle and whatnot. But first, there will be tea. It can only help.