Matterhorn 18-001 · PDF/UA-1 UA1:7.8-1
Running headers and footers must be pagination artifacts
Running heads, footers and page numbers are tagged as real content, so they repeat in the reading order on every page.
The condition, as the protocol states it. “Headers and footers are not marked as pagination artifacts.”
Matterhorn Protocol 1.1, checkpoint 18, index 18-001, section UA1:7.8-1. The protocol marks this as requiring human judgement.
What a screen reader actually does
The reader announces "Annual Report 2026 — page 14 of 88" between every paragraph across the page boundary.
What is in the file
An artifact is a marked-content sequence with the tag /Artifact and no MCID, so it has no route into the structure tree:
/Artifact <</Type /Pagination /Subtype /Header>> BDC
BT /F1 8 Tf 1 0 0 1 72 800 Tm (Annual Report 2026) Tj ET
EMC
/Artifact <</Type /Pagination /Subtype /Footer>> BDC
BT /F1 8 Tf 1 0 0 1 480 40 Tm (Page 14 of 88) Tj ET
EMC
/Type /Pagination says what kind of artifact it is. /Subtype /Header or /Footer says which band it belongs to, and that is what lets a processor offer “read running head” as a command rather than discarding the text entirely.
The failure is the same content wrapped in /P <</MCID 12>> BDC instead. It is now a paragraph in the reading order, and there is one on every page.
What the user hears
At every page boundary:
Annual Report 2026. Page 14 of 88. …the second half of the sentence that was interrupted.
Eighty-eight times in a document of that length. The reader has no way to suppress it, because as far as they can tell it is body text. In a document with a header, a footer and a page number it is three interruptions per page.
This is the defect that most reliably makes a technically-conformant document unbearable to use.
How to fix it free, in Acrobat Pro
The Reading Order tool handles this well, and there is a bulk path that most people miss.
The bulk path:
Accessibility → Reading Order.- Draw a marquee around the header on page 1.
- Click Background/Artifact.
- Acrobat asks whether to apply to a page range. Say yes, and give it the whole document.
- Repeat for the footer and the page number.
That is genuinely a two-minute job for a whole document, and it is the single highest-return free fix available.
Per page, if the layout varies: same steps without the page range. Watch for documents where the header differs between chapters, or where the first page of each section has no header — the marquee has to be redrawn.
Upstream: anything placed in Word’s header/footer region, or on an InDesign parent page, exports as an artifact automatically. If your document’s headers are in the body text instead, fixing that in the source solves it permanently.
The subtype nobody sets
Acrobat’s Background/Artifact button writes /Artifact with no /Type or /Subtype. That clears condition 18-001 and leaves 18-002 failing.
The difference is real: a plain artifact is content a processor should ignore. A /Pagination /Header artifact is content a processor knows is a running head, and can offer on request. Users who want to know what section they are in lose that.
Acrobat cannot set the subtype through its UI at all. You need Preflight’s fixups, a tool that writes it, or a producer that got it right in the first place.
Where Taggart differs
Taggart identifies running content the way a human does — by noticing that it repeats. It normalises digits, so Page 3 of 40 and Page 4 of 40 are recognised as the same string, and looks for text appearing in the top or bottom 11% of the page on three or more pages:
The text ‘Annual Report #’ repeats in the header band on 88 page(s) but those pages contain no /Artifact sequences, so the running content is in the reading order on every page.
Where the band is artifacted, it checks for the subtype separately and reports 18-002 instead.
The fix is not in “Fix All Safe”, deliberately. Artifacting removes content from the reading order, and a tool that does that on a heuristic will eventually delete a section heading that happened to sit at the top of three pages. Taggart lists the candidates with the pages they appear on and requires each to be approved, then records the approval in the remediation report.
How Taggart handles it
The protocol marks this condition as needing human judgement, so no validator can decide it outright. Taggart applies a heuristic to find the likely cases and reports them as warnings for you to confirm — it never passes them silently, and it never fails them silently either.
The fix is Artifact running headers and footers. It changes what the document means, so Taggart proposes it and waits for your approval rather than applying it unattended.
Whatever Taggart changes, the page still looks identical. Fixes edit the PDF object model, not the content streams, and every remediation is re-rendered and compared against the original at structural similarity ≥ 0.999 before you get the file back. A tool that quietly reflows your document is worse than no tool.
Frequently asked questions
Should page numbers be artifacts too?
Yes. A page number is pagination, not content. Tagged as content it interrupts the text at every page boundary; artifacted, a reader can still surface it on demand because the artifact is subtyped as Pagination.
What if the header contains something a reader needs?
Then it is not a running header, it is content that happens to sit at the top of the page — a document title on page one, a section name that appears once. Artifact the repeating furniture, keep the content. The test is whether it repeats mechanically or says something specific to that page.
What is the difference between 18-001 and 18-002?
18-001 is whether the running content is an artifact at all. 18-002 is whether the artifact declares /Subtype /Header or /Footer. Both are needed: the first keeps it out of the reading order, the second lets a processor offer it on request instead of discarding it.
Why is Taggart cautious about applying this automatically?
Because artifacting removes content from the structure tree entirely. Get it wrong and the content is not mislabelled, it is gone — condition 01-002, the most damaging mistake in remediation. Taggart identifies the candidates by finding text that repeats in the same band across three or more pages, and asks before removing anything.
Where this sits in the standards
| Standard | Reference |
|---|---|
| Matterhorn Protocol 1.1 | Checkpoint 18 (Page Headers and Footers), index 18-001 |
| PDF/UA-1 (ISO 14289-1) | Clause 7.8-1 |
| WCAG 2.1 | 1.3.1 Info and Relationships — Level A |
| EN 301 549 / Section 508 / ADA Title II | All three point at WCAG 2.1 Level AA for non-web documents, so a Level A or AA criterion here is in scope for each of them. |
Related conditions
- 18-002 Header or footer artifacts are not classified as Header or Footer subtypes The content is an artifact but carries no /Subtype /Header or /Subtype /Footer, so a processor cannot offer to read running content on demand.
- 01-005 Content is neither marked as Artifact nor tagged as real content Every text-showing, image-painting and path-painting operator on a page must sit either inside a marked-content sequence that maps to the structure tree, or inside an artifact sequence. Content that is in neither is orphaned.
- 01-002 Real content marked as artifact Content that carries meaning has been marked as an artifact, which removes it from the structure tree entirely. This is the most damaging single mistake in PDF remediation because the content becomes invisible rather than merely mis-labelled.
- 01-001 Artifact tagged as real content Decoration — a rule line, a page-border image, a watermark, a background flourish — has been given a real content tag, so assistive technology announces it as if it were part of the document.
Other conditions in checkpoint 18
Checkpoint 18 covers page headers and footers. These are its other failure conditions; the ones without a link are in the rule set but do not have a written page yet.
Check your own file. Taggart’s validator is free and unlimited — every machine-checkable Matterhorn condition, no page cap, no watermark, no account needed.
Verified against a real document on . Last updated .