Matterhorn 18-002 · PDF/UA-1 UA1:7.8-1
Running header artifacts with no Header or Footer subtype
The content is an artifact but carries no /Subtype /Header or /Subtype /Footer, so a processor cannot offer to read running content on demand.
The condition, as the protocol states it. “Header or footer artifacts are not classified as Header or Footer subtypes.”
Matterhorn Protocol 1.1, checkpoint 18, index 18-002, section UA1:7.8-1. The protocol marks this as requiring human judgement.
What a screen reader actually does
The user loses the option to hear the running head when they want it.
What is in the file
An artifact is marked content tagged /Artifact, and it may carry a property list. For running content, ISO 32000-1 (14.8.2.2) defines the entries that classify it: /Type /Pagination, and for pagination artifacts a /Subtype of /Header, /Footer or /Watermark. There are three shapes a running head can take in the content stream:
/Artifact BMC % no property list at all
BT /F1 9 Tf 1 0 0 1 72 810 Tm (Annual Report 2026) Tj ET
EMC
/Artifact <</Type /Pagination>> BDC % typed, but no subtype
…
EMC
/Artifact <</Type /Pagination /Subtype /Header>> BDC % what this condition asks for
…
EMCThe first two are both condition 18-002. They are out of the reading order — 18-001 is satisfied — and they do not say which band they belong to.
What it costs
Not much that is audible today, which is worth saying plainly. The screen readers most people use skip artifacts, and a subtyped artifact is skipped the same way as a bare one. Nobody hears the running head interrupt the text in either case.
What the subtype buys is the option. A running head carries information a reader sometimes wants — the chapter, the section, which part of a long report this page belongs to — and the subtype is the only entry in the file that says this is the running head rather than some other piece of page furniture. Without it, a processor that wanted to offer “read the header” on request has nothing to find. The protocol treats the omission as a warning, and that is the right weight for it.
Where it comes from
Artifacting by a tool that does not classify. Marking content as an artifact is one step; filling in its property list is another, and nothing requires the first tool to do the second. Acrobat’s Accessibility Checker rule Tagged content asks only that every piece of content be tagged or marked as an artifact (Adobe’s description), so the file passes it either way.
Producers that artifact the header band without a property list. /Artifact BMC is the shortest legal way to write an artifact, and the obvious one for a program that only wants the content out of the tree.
How to check it free
Acrobat Pro’s Content panel. Open it from the hamburger menu (Windows) or the View menu (macOS): Show/Hide → Side Panels → Content. Select the running head on the page and choose Find Content From Selection from the Options menu to locate its container. Adobe documents Edit Container Dictionary as available only for containers that have a dictionary — so if the command is not offered, the artifact is the bare /Artifact BMC kind, and if it is, the dictionary it opens shows whether /Subtype is there.
Reading the content stream. qpdf --qdf --object-streams=disable in.pdf plain.pdf decompresses the page contents. Search for /Artifact and look at what follows: BMC, a dictionary without /Subtype, or one with it.
Not the validators. Acrobat’s checker has no rule about artifact subtypes, and veraPDF’s PDF/UA-1 rule set has none under clause 7.8. Both are consistent with the protocol leaving this to a person.
How to fix it free, in Acrobat Pro
Adobe’s Content panel documentation describes Edit Container Dictionary as the place to specify a container’s dictionary, and warns that a mistake in it can damage the PDF. It does not describe artifact properties. With that understood, and on a copy:
- In the Content panel, find the running head’s container as above.
- Choose Edit Container Dictionary.
- Make sure the dictionary has a
Typeentry with the namePagination, and add aSubtypeentry with the nameHeader— orFooterfor the bottom band. - Repeat on every page. There is no batch path.
The keys and values come from ISO 32000-1, not from Adobe’s documentation; Adobe documents the dialog, not this use of it.
If the command is not offered, the artifact has no dictionary to edit. Acrobat’s documented commands are Remove Artifact and Create Artifact, and Adobe does not document what property list Create Artifact writes — so after re-creating an artifact, check it with Edit Container Dictionary before assuming anything.
For a long document this is slow enough that the source is worth trying first: put running heads in the authoring application’s header and footer regions, export, and check one page with the steps above. If the producer omitted the subtype once, it will omit it again.
Where Taggart differs
Taggart finds running content by repetition. Text in the top or bottom 11% of the page that repeats on three or more pages, with digits collapsed so Page 3 of 40 and Page 4 of 40 match, is a running head or footer. If that text is still tagged on some page, it reports 18-001. If it is artifacted everywhere, it looks for any artifact in the document that declares the matching subtype, and if there is none:
The header band is artifacted but no artifact declares /Subtype /Header, so a processor cannot offer to read the running content on demand.
Three consequences of that design. It is reported once per band, not once per page. One subtyped header anywhere in the document satisfies the header band for the whole file, so a document where only the first chapter’s heads are classified will pass. And a document of one or two pages is not checked, because nothing can repeat three times in it.
There is no repair for this condition. Taggart’s repair for 18-001, Artifact running headers and footers, works on running text that is still in the reading order: it turns it into /Artifact <</Type /Pagination /Subtype /Header>> BDC (or /Footer), so running content Taggart artifacts never produces this condition. Run on a file whose running content is already artifacted without a subtype, it finds nothing to work on and says so:
no running heads, footers or page numbers were found to artifact
So for this condition Taggart reports, and the subtype is added in Acrobat or at the source as above. The finding is a warning; in the application it goes to the review queue, where you can record what you did about it — or that you have decided the subtype can wait.
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 each one as warning for you to settle in the review queue. A heuristic finds what it was built to find and can miss the rest; what this one looks for is described above.
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
The header is already an artifact. What is still wrong?
Only its label. Being an artifact takes it out of the reading order, which is the part that matters most and is condition 18-001. A pagination artifact with /Subtype /Header also says what it is; a bare artifact says only that it is not content.
Does any screen reader use the subtype?
In the readers most people use, an artifact is skipped whatever its property list says, so on its own this changes nothing a listener hears today. PDF/UA asks for the subtype so the information is in the file for a processor that can offer running content on request, and the protocol marks this a warning rather than a blocker.
Where does a page number go?
With the band it sits in. ISO 32000-1 defines three subtypes for pagination artifacts — Header, Footer and Watermark — and a page number at the foot of the page is part of the footer.
Will Taggart add the subtype for me?
There is no repair for this condition. Taggart's repair for 18-001 writes /Type /Pagination and the subtype on running content it takes out of the reading order itself, so its own work never produces 18-002. Content that is already an artifact without a subtype is not something that repair touches, and the subtype has to be added in Acrobat or at the source.
Where this sits in the standards
| Standard | Reference |
|---|---|
| Matterhorn Protocol 1.1 | Checkpoint 18 (Page Headers and Footers), index 18-002 |
| 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
- 18-001 Headers and footers are not marked as pagination artifactsRunning heads, footers and page numbers are tagged as real content, so they repeat in the reading order on every page.
- 01-005 Content is neither marked as Artifact nor tagged as real contentEvery 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 artifactContent 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.
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 — 75 of the 87 Matterhorn conditions software can decide, with the other 12 listed for you to confirm; no watermark, no expiry, and no account needed up to 60 pages.
Checked with Taggart on a generated file that fails this condition on . The Acrobat steps follow Adobe’s documentation and were not walked through by hand. Last updated .