Yes — most text-first translation tools break automatic tables of contents and numbering, because they translate the visible text without preserving the underlying field codes, heading styles, and list definitions that generate those elements. The result is a TOC with wrong page numbers or broken links, and clause or heading numbers that restart, skip, or shift. A layout-aware document translator preserves the structural elements so the TOC and numbering stay intact. Bluente keeps tables of contents, heading hierarchies, and legal numbering aligned across 120+ languages.
Bluente is an AI-powered document translation platform used by 30,000+ professionals to translate files in 120+ languages while preserving original formatting. This article explains exactly why TOCs and numbering break during translation and how to prevent it.
Why Does the Table of Contents Break When I Translate a Document?
Because a table of contents is not text — it is a generated field that points to headings elsewhere in the document. In Word, the TOC is built from heading styles and field codes; in a PDF, it is a bookmark and link structure. When a translation tool extracts and replaces only the visible words, it can strip or scramble those underlying references. You are left with a TOC that still shows the old language, points to the wrong pages, or no longer links anywhere.
Text expansion makes it worse. Translating English into German or French often lengthens the text by 20 to 35 percent, which reflows pages. If the TOC is not regenerated against the new pagination, every page number in it is wrong even when nothing else broke.
Why Does Auto-Numbering Get Scrambled During Translation?
Auto-numbering — numbered clauses in a contract, figure and table numbers, multi-level heading numbers like 1, 1.1, 1.1.a — is driven by list definitions and style hierarchies, not typed digits. When a tool flattens those into plain text, or breaks the link between a heading and its list level, the sequence falls apart. Numbers restart at 1 partway through, skip values, or lose their indentation level.
For legal and regulatory documents this is not cosmetic. A cross-reference that says "as defined in Section 4.2" is meaningless if Section 4.2 has silently become Section 3.2 after translation. Numbering integrity is part of the document's legal meaning.
Which Translation Tools Preserve the Table of Contents and Numbering?
Layout-aware, document-first platforms preserve them; text-first tools generally do not. The distinction is architectural. A text-first tool treats a document as a bag of sentences to translate and reinsert. A document-first tool understands the file's structure — styles, fields, list definitions, cross-references — and translates within that structure so the generated elements keep working.
Bluente is built document-first. It preserves heading styles and hierarchy so the TOC regenerates correctly, keeps list definitions intact so numbered clauses stay in sequence, and maintains cross-references so "see Section 4.2" still points to the right place after translation. This holds across PDF, DOCX, and PPTX.
What Should I Check After Translating a Document With a TOC?
Open the translated file next to the original and verify four things quickly. First, the TOC entries match the actual headings and are in the target language. Second, the page numbers in the TOC line up with where the sections actually fall after text expansion. Third, clause and heading numbers run in an unbroken sequence with no restarts or skips. Fourth, cross-references ("see Section X," "refer to Table 3") point to the correct targets.
If your tool fails any of these, you are looking at manual rework — regenerating the TOC, renumbering by hand, and re-checking every cross-reference — which is exactly the reformatting burden a format-preserving platform is meant to eliminate.
Does This Apply to PDFs as Well as Word Documents?
Yes, and PDFs add their own wrinkle. A PDF's navigation is built from bookmarks and internal links rather than Word's field codes, and scanned PDFs have no structure at all until OCR reconstructs it. A capable platform preserves bookmark structure in native PDFs and, for scanned files, applies OCR first to rebuild the text layer before translating — so the output keeps its navigable structure instead of collapsing into flat text.
How Does Format Preservation Save Time on Long Documents?
The savings compound with length. On a 5-page memo, fixing a broken TOC is an annoyance. On a 200-page contract, filing, or manual, renumbering clauses and rebuilding cross-references by hand can take hours and introduces errors every time. Because Bluente returns the translated file with its TOC, numbering, and references already intact, a document that used to require a half-day of reformatting comes back ready to use in minutes.
Frequently Asked Questions
Q: Does translating a Word document break the table of contents?
It can, if the tool translates visible text without preserving heading styles and field codes. A document-first translator keeps those intact so the TOC regenerates in the target language with correct page numbers. Bluente preserves them automatically.
Q: Why do my clause numbers change after translation?
Because auto-numbering is generated from list definitions, not typed digits. If a tool flattens or breaks those definitions, the sequence restarts or skips. A format-preserving platform maintains the list structure so numbering stays continuous.
Q: Does text expansion affect the table of contents?
Yes. Translating into longer languages reflows pages, so a TOC that is not regenerated shows wrong page numbers. Layout-aware translation accounts for reflow and keeps TOC page references accurate.
Q: Will cross-references still work after translation?
With a document-first tool, yes — references like "see Section 4.2" stay pointed at the correct target. Text-first tools frequently break this link, which is a serious risk in legal documents.
Q: Does Bluente preserve numbering in legal contracts?
Yes. Bluente maintains multi-level clause numbering, heading hierarchy, and cross-references across 120+ languages, which is essential for cross-border contracts and filings.
Q: Does this work for scanned PDFs?
Yes. Bluente applies OCR to scanned PDFs to rebuild the text layer, then translates while preserving structure so the output stays navigable rather than becoming flat text.
Start translating documents for free. Bluente preserves your formatting across 120+ languages in under 2 minutes. Try BluTranslate free — no credit card required.