When Roberta Ganeo describes her own way of working, she says she writes to find out what she is thinking. Not to record what she already knows. To find out. By the time she has finished a draft of Il Viaggio dell'Eroe Altamente Sensibile — The Journey of the Highly Sensitive Hero — the book she is writing is not the book she sat down to write.
Most of the first-time authors I work with are in the same place. They just do not know it yet. Writing clarifies their thinking, and the book's real message comes into view through the writing, not before it.
They come to me with what they are sure their book is about — clean, articulable, the version they could pitch over coffee. They have it down to a sentence. Sometimes they have had that sentence for years. And then they begin writing, and somewhere around chapter seven the manuscript begins to disagree with them.
The author still believes they are writing the book they pitched me. They are not. The book has started to be about something else, and the pages already know it. The author is the last to find out.
The First Draft Is Not What You Think It Is
Most writing advice frames the first draft as the rough version of the book — the messy attempt the rewrite will tidy into the real thing. That framing is wrong, in a way that costs first-time authors months.
The first draft is not the rough version of the book. It is the conversation in which the book tells you what it wants to be.
That distinction changes what the second draft is for. If draft one is the messy version of a known book, draft two is a polishing job. If draft one is the conversation in which the book reveals its real subject, draft two is a different cognitive operation on the same material. Not a polish. A different kind of thinking on the same words.
Have you ever sat down to write a short message — for a wedding, a leaving card, a note to a friend going through something difficult — and discovered, halfway through the third sentence, that you felt something different from what you sat down to say? The page made you honest. A book does this too. It just takes longer.
This is also why I rarely meet a first-time author afraid of editing. Most are afraid of starting. The fear of the first chapter is its own subject, but it matters here: most of the authors who never get past it are waiting for a clarity the page is supposed to give them.
Why Writing Clarifies Thinking
The reason a manuscript can disagree with its author is plain.
When a thought stays in your head, it is fast and slippery. You cannot easily examine it from the outside, because it is still part of the apparatus doing the thinking. The moment you put it onto a page, something changes. The page is slower than thought, and more permanent. The sentence stops being part of you and becomes something you can read.
And once you can read it, you can disagree with it. You can notice it does not say what you meant. You can notice it says something you did not know you thought. The page lets thought get larger than the head it came from.
Researchers Veerle Baaijen and David Galbraith have shown that the moment in drafting where the writer finds out what they think is not the same as sentence-level polish — it correlates with revising the structure of the work, and with generating new sentences as you go, not with smoothing prose you have already written. You cannot edit a sentence into telling you what your book is about. The book finds that out through chapters reshaping each other. My co-founder has written more on that idea — the structural view of a manuscript — through a different lens, but the underlying point about structure carrying meaning is the same.
The Words That Show Up When You Are Figuring Something Out
There is a quiet linguistic signature when a writer is genuinely thinking on the page. James Pennebaker and his collaborators have spent decades documenting it: words like realise, because, reason, consider, think start showing up more often — often enough to mark the moments where the page has stopped being a transcript and started being a tool for finding something out.
It is easier to see in a small example. An author writes Leaders make decisions under pressure. That is knowledge-telling. The same author, weeks later, writes I keep thinking that the leaders I admire are the ones who admit, in the moment, that they are not sure. That is knowledge-transforming. The shift is not in the topic. It is in the words I keep thinking, admit, not sure. The author has stopped reporting and started thinking with the reader as a witness.
When I read a first draft, I notice those words too. Not because I am counting. Because I can feel the pages where the author has gone from telling me what they know to working out what they are still figuring out.
How to Find Your Book's Real Message
A distinction from Carl Bereiter and Marlene Scardamalia helps here. Most beginning writers do knowledge-telling — they write down what they already think. Experienced writers do knowledge-transforming — the writing changes what they think. This is not stylistic. It is developmental. It is the move almost every first-time author has to make, often without realising it is happening.
A knowledge-telling chapter has a particular feel on the page. The prose is confident, even fluent. The sentences land where the writer expected. The author is not surprising themselves anywhere. Read enough of these in a row and the chapter starts to feel airless — as if the author already knew what they were going to say before sitting down, which is usually because they did.
A knowledge-transforming chapter reads differently. The rhythm changes. Sentences trail off mid-thought and pick up again from a new angle. The author writes a confident line and then writes but actually and contradicts themselves productively. In the margins of a printed draft I have read, an author once wrote I do not know in pencil next to her own paragraph. That is the moment the chapter started transforming her instead of recording her.
Your first draft may very well be a knowledge-telling draft. That is the starting condition, not a failure. This is also why "just outline harder" is bad advice for someone wrestling with what their book is about. An outline is, by definition, knowledge-telling. Only the chapters themselves can do the rest.
A Different Author, A Different Discovery
Roberta's is one flavour of this. Here is another.
I was working with an author who had pitched me a book on leadership — what she had learned from a long career in a difficult sector. She knew her material. The first three chapters were good in the way I described above: confident, fluent, airless.
Then she sent me chapter four. Chapter four was about the people she had lost on the way — colleagues who had left the work, mentors who had died, a friend who did not survive the years she described. She had not planned to write about them. The chapter was, in her email, "an aside."
It was not an aside. By chapter eight she was no longer writing a leadership book. She was writing about grief, and about what grief teaches you about leading — which turned out to be the book she had needed to write for years and had not known how to begin. The leadership frame was the runway. The book was about loss.
She fought it for a while. Most authors do.
Why Authors Resist When the Book Pulls Them Somewhere Else
When a manuscript starts pulling its author toward a different subject, the author almost always resists. It is not laziness. It is not a lack of discipline.
It is the discomfort of letting the book be smarter than you were when you started.
The author who pitched me a leadership book had been saying for years that she was writing a leadership book. She had a sentence she could use at dinner parties. To follow the manuscript into the territory it was opening up — grief, loss, what it costs to keep going — meant unsaying that sentence, and trusting that the new book was real and not a chapter-eight wobble.
Most first-time authors have a version of this. The pages start pulling them somewhere they did not plan to go. Their first instinct is to pull the pages back. That is the instinct to watch for. The resistance is not telling you the manuscript is wrong. It is telling you the manuscript has begun to know more than you do — and that is almost always the sign the work is going well.
What This Means If You Are Sitting on a Book
You do not know what your book is about yet. And that is not a problem to solve before you start writing. It is the work the writing itself is going to do. The author waiting to start because they do not yet have the perfect sentence is waiting for an output the process itself is supposed to produce. The path becomes visible after you have walked it, not before.
Joan Didion put it more bluntly: "I write entirely to find out what I'm thinking, what I'm looking at, what I see and what it means." That is not modesty. It is method.
A Calmer Way to Begin
If you are at the start of a manuscript and do not yet know what your book is really about, here are three pieces of guidance — not a checklist, just a direction.
Keep writing past the point where you would normally stop and re-outline. The instinct to tidy the plan is usually the instinct to retreat from the page back to the safer territory of the outline. The discovery rarely happens there. It happens in the chapter you have not written yet.
Let the second draft be a different cognitive operation, not a polish. Ask, of every chapter, what is this chapter actually doing for the book the manuscript wants to become — not is this prose good. The prose can wait. The structure cannot.
Pay attention to the chapter where the words you are using start changing. When think, because, realise, not sure begin showing up where they did not before, do not edit them out. That is the chapter the rest of the book is probably going to bend toward.
This is also why careful human reading is not a luxury. The pattern of an author starting to think on the page is something a reader can feel and name — and one of the reasons why ChatGPT cannot write your novel for you: discovering what a book is about needs a reader who knows the author, not a tool that does not.
If you are sitting on a manuscript that has started disagreeing with you, that is usually a good sign. It means the page has begun to think back.
Marion Eickmann-Tomasini is co-founder of my-book.ai, where she reads every chapter of every book project personally. She is the author of The Splinter of Memory, the first volume of the Velirion Chronicles. Before books, she spent more than fifteen years helping leaders see their own thinking clearly — a habit she now applies, chapter by chapter, to first-time authors. We read first drafts that have started to know themselves.