Recently, a former student who had once been part of an ongoing workshop I ran asked if I still had the list of recommendations for critiquing manuscripts that I used to give out. I’m so far from having a copy of the list of suggestions that I had forgotten it ever existed. But her request got me to thinking, and so I’m going to try to resurrect it, if in somewhat abbreviated form.

My suggestions for giving critiques boil down to two:

First, always begin your comments by naming something about the manuscript you like. This is imperative, not only for the ego of the writer receiving the criticism, but for your own understanding of what you have read/heard. It’s easy, far too easy, to leap into a manuscript and begin to pick. But it’s only when you consider first what the writer is doing well that you are truly seeing the piece whole. And only then, when you have seen and acknowledged its strengths, can you be useful in pointing to the places where a work in progress might be improved.

I used to warn my students not to be intimidated after reading to a group if the first this-is-what-I-like comments were slow in coming. My experience is that the stronger a manuscript is the harder it is to define just what makes it work and why. And for some reason, the more impressed workshop participants are with what they have heard the more likely they are to pick. (My most charitable explanation for that phenomenon—we’ll put pure jealousy aside—is that such a response comes from a powerful need to be helpful.) But having to define what works and why slows down the rush to demonstrate our critical insights.

So if you belong to a writers’ group—or exchange manuscripts for comment—I recommend that you make this a hard and fast rule. Each person who speaks must first tell the writer what she likes about what she has heard or read. Only then does she have permission to suggest improvements.

Second—and this rule is a harder one for most of us to follow—don’t try to fix whatever problems you see. You may have excellent ideas for improving someone else’s manuscript, but the revising is not your job. Tell the writer what you don’t understand. Tell her what you want more of or not so much of. Tell him how you respond to his characters so he can decide if that’s the way he wants you to respond. Tell her where the movement seems abrupt. Tell him where the story trajectory loses momentum. But don’t tell her how to fix it. Stand back and give her room to climb back in to find her own solutions.

Even the best, most creative suggestions—“Wouldn’t it be better if . . . ?”—are apt to make it harder for the writer to find her own place in her work when she returns to it. That good idea came out of your creative energy. What we all need when we are revising is to tap back into our own.

So when you are offering a critique, say what doesn’t work for you and why, then leave it at that. Don’t problem solve. The writer conceived the piece you are responding to. She is capable of conceiving a solution, too.

That’s exactly the kind of help I want from an editor, and it’s the kind of help that is most useful between writing peers as well.

Those are the only rules that matter when offering critiques. Respect the manuscript, which is another way of saying tell the writer what he is doing well. And don’t intrude. Offer your insights about what work still needs to be done, but don’t try to explain how to do it.

Next week I’m going to consider the other side of the coin, how to receive criticism, whether in a writing workshop or from an editor.