Conforming chapters

First, let me say that I find your software a pleasure to use and have employed it to write and export a recently published novel.

There is one feature, however, that I and perhaps others would find useful. When moving chapters to a different position within the binder, it would be helpful if the associated chapter numbers are automatically changed to the new corresponding position. And it would be nice if the associated change, perhaps as a option, is in addition extended to the corresponding chapter headings within the document. Without this feature, if one decides that chapter 50 should really be moved to the position of chapter 4, all subsequent headings and chapter numbers must be changed manually to maintain proper order.

Thank you in advance.

Hi Ron,

Check out “Auto-numbering” in Scrivener’s Help.


Auto-number is indeed your friend, and in 2.0’s new compile, you’ll be able to set up numbering dynamically as a part of the compilation rules. You’ll be able to set top-levels as “Part <$w>” and second level folders as “Chapter <$n>”, and everything below that as “<$hn>” which will give you 1.3.12 style numbering—all without touching a single title name in the Binder.

Thank you for your reply.
But unless I have been misusing the compile function, it is to be used in preparation for export. My dilemma relates to the organization and numbering of chapters during the creative process. There are occasions where a new chapter or chapters must be added to the mix and once that is accomplished, all subsequent chapters must be renumbered both in the binder and the individual files (documents). If there is an existing function within Scrivener that I have overlooked and that is capable of overcoming this issue I would be delighted to to be informed.

And thank you to the poster who mentioned the “auto number” function. What exactly does it do? I’ve noticed that the function is not available from the binder but is active within the corresponding document.

Thank you once again.

The idea of auto-numbering is indeed that it gets applied when you compile your manuscript for export or print. In Scrivener 2.0 you can hide these auto-number tags entirely out of the way and have them applied only at compile time - the idea is that until then, you can name the folders anything you want. You don’t have to use numbers at all during the creative process - surely chapter numbering is for the final product and not part of the writing itself?
All the best,

Having been a writer/novelist for over 10 years, I respectfully disagree with your statement in re chapters. Keeping tabs on specific chapters for a long project is an essential requirement, at least for my writing style, for maintaining a cohesive flow of events and character evolution. If one is simply creating an outline, or brainstorming, then perhaps the allocation of information to specific chapters is unnecessary, nay, even superfluous. But for this novelist and I suspect many others, chapters are an essential means of keeping track of one’s progress and additionally, offers the benefit of allowing one to create an effective timeline via some method external to this software.

I think you misunderstand me. I wasn’t saying that keeping track of chapters isn’t important to anyone, I was merely talking about the actual numbering of chapters. Instead of having “CHAPTER ONE” as a folder title in Scrivener, you could just use the chapter subtitle, or something meaningful - a title that tells you what that chapter is about - and then have “CHAPTER ONE” inserted during the compile process.

But if you personally find you need the chapter numbers in the binder for your own way of working, then the answer to your question is indeed that there is currently no way of doing this, and there’s not likely to be, not in the near future at least. Not because I don’t think it’s a good idea, but more for technical reasons. Auto-numbering in Scrivener cannot work as auto-numbering does in a word processor. In a word processor, the whole document is open at once, so whenever you make any changes, the word processor can quickly iterate through all occurrences of your number fields and update them. This is more complicated in Scrivener, though. Auto-numbering can occur in titles, in the text, in notes; and you can choose to exclude documents from Compile and so on. So to keep live numbering up to date in Scrivener, Scrivener would need to check the text of every single document in the project, the titles, the notes and so on, and update them each time “include” is unchecked… It would be cumbersome and problematic. Thus auto-numbering is left as a function of the Compile process, when it can be accurately calculated.

Hope that makes sense.

All the best,

Not to necessarily disagree with you, as everyone has their own system and is entitled to it, but when you say things like: “Keeping tabs on specific chapters for a long project is an essential requirement…”, that causes me to think of all kinds of things other than numbers. I think of project level favourites dispersed throughout the item navigation and targeting interface; visual highlighting; automatic collection based on type—basically I’m thinking of ways the interface can envelop the concept of structural mind-based highlights, rather than construction mental indexes of those highlights.

I’m speaking rather abstractly here—but inline numbering and referencing by that numbering is a form of mental indexing, or association by assigning a symbol to a deeper construction of items. “This set of items, which we’ll call a chapter, are assigned to the number 5; chapter five”. It’s a way of linking and simplifying, and it is convenient because it maps to the end product, but it has its weaknesses. If all that deeper structural stuff gets moved, it’s no longer chapter 5. The mind has to re-index and keep up with topology shifts. If, on the other hand, that structural stuff was assigned an interface tool, the interface basically picks up the mental task of indexing. You now have all of that structural information being simplified into an abstract thing, like a list of items you can quickly navigate to; quickly move new items to; quickly open in a window; quickly see in a list with other such items. This kind of indexing is resilient to topological tampering because it is defined by the actual structural elements themselves. If the chapter moves, it doesn’t matter in the slightest to the highlighting interface.

The last thing you states is definitely the largest disadvantage to this way of thinking and working. If external tools must be brought to bear upon the internal structure in Scrivener, then application supplied indexing is nearly useless. It doesn’t matter how slick things are inside of Scrivener, if you need to place an entry in a timeline referring to parts of this topology there are no mental hooks for doing so. Of course, numbering still has the same weakness here as well, and perhaps even more so, for a topology shift not only screws up the internal Scrivener mental indexing, but any external references as well. You have to fix “see chapter 5” notes in Scrivener and everything else outside of it—but at least you have something.

I can think of some ways around this problem, but I’d have to think about it a bit more before pontificating on it further.

Hi Keith,
Thank you for the explanation.
I submit that each of us works differently and you have brilliantly taken note of this fact in your equally brilliant software. However, some months back I transitioned from the PC environment and a writing package called PowerWriter (Windows only software), to the Mac and Scrivener. While I have no intention of ever going back to the aforementioned software, I do miss the feature in question. And while I have no knowledge of the technical aspects of the PowerWriter program, it appears to be organized in a fashion similar to Scrivener, in as much as it is a file based structure sporting a format quite similar to the binder. That said, repositioning a chapter in the binder equivalent is accompanied by the appropriate renumbering of all subsequent chapters. And it was with this in mind that I raised the question of allowing for a similar function in Scrivener.
Thank you once again for your response and this wonderful writing tool.

Oh, and by the way, index card numbering is coming in the upgrade as well. If you click on Draft and view it as a corkboard, you’ll see “1. Introduction”, “2. Et cetera”, and of course if you move those cards around the numbering will update. It’s relative numbering though. “Introduction” is not implicitly “1”, it’s just the first card in the corkboard. In a search result it might not be the first card.

I am smiling at your quite erudite explanation of the subject at hand–impressive. I further recognize the validity of your construct. However, that said, many novelists––myself included–– work with a timeline and since that feature is not available from within the software one must find an alternative. And such alternatives exist in many forms, including costly light boards, blackboards, and software, the latter being my preferred approach. If one does subscribe to the TIMELINE approach, it would appear that the most efficient means of creating said timeline would be with the introduction of chapter numbers or some other chapter identifier.

Why do I use chapters during the creative process? Saves me time at the end and makes for a uniform timeline. Could you call each block of writing something other than a chapter? Of course, but somewhere along the line you will have to rename it as a specific chapter, assuming you intend to have it published, so why not save time and do it from the onset?

But in the end, we all do what works best for ourselves and that is why an extremely flexible program like Scrivener is so valuable.

Thanks for the kind words, much appreciated.

Actually, my point was that you don’t need to rename it, not yourself anyway - that is something that Scrivener 2.0 can do for you at the Compile stage. Scrivener 2.0’s freeform corkboard might also be useful for arranging timeline-like data, especially in conjunction with the new Collections feature.

All the best,

Well, of course I have not seen version 2.0, but for my style of writing, the Timeline is something that I find useful during the creative process, constructing one at the end would be counterproductive. By the time I have reached the level of compilation my novel is ready for export and editing.
But, on the other hand, the free-form cork-board might serve as an alternative to the external timeline and will be a welcome addition to the software.


Oh, it would definitely be pointless to construct a timeline at the end, yes - that wasn’t what I meant. I merely meant that you might be able to use the freeform corkboard for arranging ideas in a manner similar to arranging a timeline. That chapter titles can be added automatically at the end was intended as an entirely separate point.

All the best,

Oh, and just to clarify, I didn’t so much mean to imply one could easily abandon chapters. While Scrivener’s outlining lets you, to a degree, work above and around the book structure itself, it still does a great job of defining the book structure, and in fact I work that way. In my current project, I have three big part folders, another for appendices, and a bunch of chapter containers beneath that. Below that level things break down to be a bit less rigid, but the top three layers are definitely “table of contents” material.

My erudite ruminations ( :slight_smile: ) were merely upon the reliance of numbers as a way of addressing chapters. With Scrivener’s new re-titling feature, it’s easy to even completely obscure internal titling for the “public face”. Does the timeline implementation need to address “chapter 5”, or does it need to address, “the train derailment” chapter—and within Scrivener, what can be done to make that folder-cum-chapter more useful, and I think there are some neat tools on the way for that. To myself, the latter seems less fallible. The odds of that chapter no longer being about a train derailment are much lower than it no longer being chapter five.

But anyway, on the notion of using compile creatively (and not just as a book production device), because of this new “procedural” approach that you can optionally apply, it would be pretty easy to export a numbered outline of titles and maybe synopsis into an RTF periodically. Maybe import that back in as a reference, and use it as your gold standard. If the structure changes, just compile out an outline again. There is even a handy compile preset for just that sort of thing.

No attempt at erudition here, but I personally find it best to assign a descriptive name to each chapter. Then not only do you not have to renumber chapters when changing order, you can easily tell the content of said chapter, even after the fifteenth rearrangment :slight_smile: You can add numbers later, when the order is fixed, ie just before compile…


I think one of the points being made above is that you don’t even have to “add numbers later”. If you write one document per chapter, you can start each document with “CHAPTER <$n>” and during compile, that will change into “CHAPTER 5” or whatever number in the sequence it should be. If your chapters are folders of multiple documents, and since folders are just a special kind of text document, you can insert the “CHAPTER <$n>” text into the folder’s textual content. As you add folders, you can simply add that text to the newly inserted folders. Rearrange, split, insert or delete chapters however you want as many times as you want… during compile, the chapter numbers will always be in the correct sequence.

The outline-like document titles in Scrivener can be completely ignored during compile time, which allows you the freedom to describe the theme/content of your chapters instead of using the generic label of “Chapter 5”. So in a timeline, you can say “1999 - The prince had a party”, and the chapter, instead of being labeled “Chapter 7” in the binder, can simply be “The prince had a party”, while still allowing you, from the very beginning, to prepare the document to be titled “Chapter 7” (or whatever number it ends up being) when it’s compiled.

Quite right. Thanks for the clarification.


Your thoughts are well taken, however, I suspect that your suggestions are based upon the upcoming version 2.0 which I currently do not have but will acquire when available.

Once again, I feel compelled to emphasize that each of us are individuals and as such have established different approaches to the craft of writing. While many of the functions that you have described, such as exporting RTF and then reimporting, are indeed feasible from a technical perspective but are time consuming. I tend to write from Prologue, if one is required, straight through to the end of the story before compiling, exporting and sending the resultant file off to my editor.

However, while designating chapters with descriptors rather than numbering may have its advantages, it would interfere with my tried and true timeline.

In any event, I fear that this thread has gone astray from the initial question/issue; that being a simple request for a feature that would allow for automatic renumbering or reordering of chapters/folders in the binder when a chapter/folder is moved. While I appreciate the various workaround and alternative method suggestions, I do not desire to change the way I have worked for years. That said, I anxiously await the latest version of the software.

Addendum: Robertguthrie’s post appeared after my post. I recognize the numbering capability of the software that becomes available with the compiling process but for my purposes that is an after-the-fact scenario.