Matterhorn 15-005 · PDF/UA-1 UA1:7.5-2

Complex tables: when Scope is not enough

In a complex table — merged cells, stacked header rows, split spans — Scope alone cannot express the associations, so explicit /Headers references are required.

What a screen reader actually does

Cells are announced with the wrong headers, which is worse than no headers because it is confidently wrong.

The table that breaks Scope

The cell containing 3 is governed by three headers: North, Q1, and 2026.

/Scope can express two of those. The 2026 cell spans columns four and five, and there is no /Scope value that means “columns four and five”. /Scope /Column on it would claim it heads its own column only, which is wrong, and a reader following that claim announces 3 under the wrong year.

What the file needs instead

Every header cell gets an /ID, and every data cell lists the headers that govern it:

% the 2026 header, spanning two columns
28 0 obj
<< /Type /StructElem /S /TH
   /A << /O /Table /ColSpan 2 >>
   /ID (y2026)
   /K [ 14 ] >>
endobj

% the cell containing 3
34 0 obj
<< /Type /StructElem /S /TD
   /A << /O /Table /Headers [ (north) (q1b) (y2026) ] >>
   /K [ 22 ] >>
endobj

The /ID values must be unique across the document, and the /Headers array references them by name. Order does not matter to conformance, but readers announce them in array order, so put the most specific first.

The two triggers

Taggart reports 15-005 when a table has either:

Spanning cells — any /ColSpan or /RowSpan greater than 1, and no /Headers anywhere in the table.

Stacked header rows — two or more consecutive rows whose cells are all TH, and no /Headers.

Both mean the same thing: the associations in this table cannot be expressed by /Scope, so whatever /Scope values are present are at best incomplete.

How to fix it free, in Acrobat Pro

Acrobat supports it, one cell at a time.

1. Tags panel → right-click the <Table> → Table Editor.
2. Right-click a header cell → Table Cell Properties → set an ID. Give it something you can remember: y2026, q1b, north.
3. Right-click a data cell → Table Cell Properties → Associated Header Cell IDs → add the IDs that govern it.
4. Repeat step 3 for every data cell.

On the small table above that is eight dialogs. On a 5×20 table it is a hundred, and there is no multi-select for it. This is the single most tedious task in PDF remediation, and it is why the honest first question is whether the table can be simplified instead.

The alternative worth considering: split it. Two tables — one for 2025, one for 2026 — each with a single header row, need only /Scope. They are also easier to read on screen and in print. Restructuring in the source document is often ten minutes against an hour of Headers references, and produces a better document.

Upstream: neither Word nor InDesign generates /Headers references on export. This one genuinely has to be done in the PDF, which is another argument for avoiding the structure in the first place.

Where Taggart differs

Taggart reports it as a distinct condition with the reason, rather than folding it into a generic table warning:

This table uses spanning cells or stacked header rows but does not use /Headers ID references. /Scope alone cannot express the header associations unambiguously. spanningCells: true, stackedHeaderRows: 2

And crucially, when it detects this case it does not write /Scope values into the table as part of “Fix All Safe”. A /Scope on a spanning header would clear condition 15-003 and leave the reader announcing wrong associations — conformant-looking and worse than before.

Writing the /Headers arrays requires knowing which headers govern which cells, which is a reading of the table rather than a property of it, so it stays with you.

How Taggart handles it

This one is a judgement call. Taggart puts it in the guided review queue with the relevant page region highlighted, records the decision you make and who made it, and prints that decision in the remediation report.

The fix is Set table headers, Scope and Headers references. It changes what the document means, so Taggart proposes it and waits for your approval rather than applying it unattended.

Where this sits in the standards

Related conditions

• 15-003 A TH cell does not contain a Scope attribute Header cells must declare whether they head a row, a column, or both, unless the table instead uses explicit /Headers ID references on every data cell.
• 15-001 A row has a header cell, but that header cell is not tagged as a header The leftmost (or, in RTL, rightmost) cell of each row labels the row but is tagged TD instead of TH.
• 15-002 A column has a header cell, but that header cell is not tagged as a header The top row labels the columns but is tagged as ordinary data cells.
• 09-004 A table-related structure element does not conform to ISO 32000-1, Table 337 The table grammar is broken: a TD outside a TR, a TR outside Table/THead/TBody/TFoot, a Table with no rows.

Other conditions in checkpoint 15

Checkpoint 15 covers tables. These are its other failure conditions; the ones without a link are in the rule set but do not have a written page yet.

• 15-001 A row has a header cell, but that header cell is not tagged as a header The leftmost (or, in RTL, rightmost) cell of each row labels the row but is tagged TD instead of TH.
• 15-002 A column has a header cell, but that header cell is not tagged as a header The top row labels the columns but is tagged as ordinary data cells.
• 15-003 A TH cell does not contain a Scope attribute Header cells must declare whether they head a row, a column, or both, unless the table instead uses explicit /Headers ID references on every data cell.
• 15-004 Content is tagged as a table for information not organized in rows and columns A table used purely to position content on the page — a two-column layout, a form arrangement — has been tagged as a real table.

Verified against a real document on 2026-09-09 . Last updated 2026-09-10 .