Problem with footnotes

I wondered whether there is a minor bug in compiling footnotes. The situation I have s that on page 1, there is a long footnote that correctly spills over on to page 2. However, on page 2, there are two footnotes, both numbered “a”, one of which appears on that page, the other on the next page. On page 3 there is a single footnote, labelled “b”. So we have:

page 1: Footnote marker a; beginning of footnote a.
page 2: Two footnote markers, a and a; end of previous footnote and footnote a
page 3: Footnote marker b; footnotes a (from previous page) and b

It takes a while to work out that the sequencing of markers and notes is correct, but it is confusing, especially having two identical markers on page 2.

Are you on Mac or Windows? What format are you compiling to?

Knowing that would make it easier to help.


Very sorry. Mac (Intel) using Ventura 3.3.1 compiling to RTF.

Is it possible that one of those footnotes is actually a comment? Thus, it is possible that there are two numberings (one for footnotes and one for comments), if you have it set up to look the same in compile.

No,that is not the problem. There are no comments on those pages, and I am not compiling comments into notes.

I should also have mentioned that these are inline footnotes. I am compiling inspector notes for endnotes, which are given a separate numbering sequence.

What app are you opening the RTF in?

Do you have footnote numbering for the inline stream set to restart each page? As you’re using a, b, c etc. it seems to me you might. If so, the note from the first page is being repeated (presumably for clarity) and the second is the new start on the page in question.

I’ve not encountered this, as I’ve never had footnotes that long. I’m not at my computer at the moment, so I can’t experiment, but I will later. I use Nisus Writer Pro, and I have Pages, so I can try those.


I certainly did my main processing in MS Word for Mac. I might have done the initial processing in Nisus Writer, until I discovered that only MSW could cope with my 900 page document.

Yes, footnote numbering is set to restart each page, with a single sequence of endnotes through the whole document (until I divide it into sections using MSW).

As I answer your questions I realise I should have created another document to check that I can replicate this problem before posting.

Thing is, I’m pretty sure the RTF file created by compile from Scrivener is not paginated and the long footnote will be a single block of text. It is NWP or word which does the layout and allocates space for footnotes.

In NWP, you can set/change the maximum space for the display of footnotes, so I presume you can in Word. What I don’t know, but will check, is whether you can set NWP/Word, not to repeat the footnote number when one goes over onto the following page.



PS If I were dealing with a 900 page document, I’d use Affinity Publisher/InDesign; 900 pages is going to tax any word processor. :blush:

Indeed, the actual numbering and placement of footnotes is being handled by Word, not Scrivener. So, first up hypothesis is that Word is having trouble doing what you asked it to do.

Handling footnotes is a tricky business for any word processing program. Having long footnotes (page-spanning footnotes) compounds the problem, especially consecutive long footnotes. To get a sense of what is happening, I would play around with the text and footnotes in the compiled output in Word. See what changes to footnote length, page breaking, resolves the weirdness.

If it is right that the challenge you face is endemic to the situation:

  1. You might want to consider /not/ having the footnote numbers restart on each page. Restart them per chapter or per section. Why not?

  2. If your work involves lots of long footnotes (and the format decisions are yours to make), you should consider using endnotes instead – perhaps shunted to the end of sections.

Thank you for this clarification. I had not realised that it was the word processor that was responsible for this, though not you have pointed it out, it is obvious that it must be the case.

Yes, the problem of formatting footnotes must be a nightmare to program–I cannot even begin to think about how one would sort out the logic.

My response will be to ignore the problem on the grounds that, when the file eventually goes to a publisher, their software will have to deal with the problem. I am putting in occasional very long footnotes because there are some points that do not belong in the text, but which some readers will find it helpful to have available. I consider Endnotes for references that need to be there but which most readers will not need to consult, or will consult only occasionally.

Thanks to everyone who has tried to help me, and apologies if the outcome is that this is an issue for which this was not the correct forum.

1 Like


Just FYI, I have tried in NWP and it doesn’t repeat the footnote number when extended onto subsequent pages. I wonder if you can turn it off in Word, but as I don’t have Word… (the last time I used it it kept crashing every 15 minutes on a paper I was asked to edit, which was formatted in columns, with diagrams, formulae, tables, footnotes etc.)

Zipped nonsense file attached :smiley: (9.1 KB)

To segue slightly, I used to do as you do with references in inspector footnotes, and commentary as inline footnotes, but it suddenly struck me that it would be better the other way round as references tend to be much shorter and therefore less intrusive on the main text.



Here’s a PDF of how it looks in NWP; of course when you open it in Word, that will change things! (87.4 KB)

Thanks for this. However, I would say better from a techncal point of view, but not better from the perspective of the reader. Commentary is what is more useful to readers.

I meant that way round while working in Scrivener. I have inline footnotes set to compile as endnotes, and inspector footnotes set to compile as page-bottom footnotes.


Sounds right! If the result will go to a publisher for typesetting, then yes definitely do not worry about it. Their pagination will be all different anyway, so ad hoc fixes in Word would be undesirable.