Summary
- The PP-26 proposal system gives
NOCtwo official meanings. PlainNOCclarifies that no proposal is made for the marked text; underlinedNOCis itself a proposal to maintain specified text without change. - ITU says the underlined form may be used when Member States want to emphasize that provisions should remain unchanged, and that reasons for maintaining them should be given.
- The live Proposal Management Web showed the distinction in use:
USA/23/1andUSA/23/2were underlinedNOCproposals covering the whole ITU Constitution and Convention. The complete submission supplied a stability rationale for both. - The public record does not prove that ITU has lost, mistranslated or misclassified any proposal. It does prove that removing one visual property makes two different procedural acts look identical in plain text.
- ITU should preserve distinct action identifiers, scope, reason, source, version and lineage in a public proposal receipt. The familiar letters and underline can remain as presentation, while the semantic state survives exports, indexing and downstream reuse.
Three letters, two acts
The action table in the PP-26 Proposal Management Web looks like a drafting key. ADD adds text. MOD modifies it. SUP suppresses it. Near the bottom, however, the legend contains two rows carrying the same three letters.
The first row is plain NOC. ITU defines it as text for which no change is proposed. The note says the symbol may be used to clarify that no proposals are made for the marked text, and that the text need not be reproduced. This is an absence made explicit. It says: within this contribution, there is no proposed intervention on this part of the instrument.
The next row is underlined NOC. ITU calls that a proposal for text to be maintained without change. Member States may use it when they want to emphasize that a provision, or selected provisions inside a wider article, should remain untouched. The guide says the reasons why the text needs to be maintained should be given.
Both forms can lead to the same textual outcome: no amended words. They do not record the same procedural act. Plain NOC reports that the contributor is not making a proposal about the marked text. Underlined NOC places preservation itself before the conference as the requested disposition. One is a boundary around the contribution. The other is a position within it.
That is why the underline matters. It is not emphasis in the ordinary editorial sense. In this legend it carries semantic state.
Strip the styling from the two labels and both become NOC. A text-only copy cannot show which row said “no proposal” and which said “proposal to preserve”. A search index that stores only visible characters encounters the same collision. So can a CSV column, an accessibility conversion, a screen scraped dataset or a hand-built comparison sheet if it does not receive an explicit action value.
This is a structural risk, not a finding that any particular ITU system has failed. The public sources do not reveal the backend identifier used for either action. They do not establish that a six-language document, an export or an assistive technology has collapsed the distinction. The provable point is narrower: the official human-facing semantics depend visibly on formatting, and those semantics deserve a durable carrier of their own.
The preservation proposal is not hypothetical
At the research cutoff, the PP-26 PMW displayed two documents and three proposals. Ghana’s Document 35 supplied an ADD proposal for a draft new resolution on protection of civilian telecommunication and ICT infrastructure. United States Document 23 supplied the other two rows.
USA/23/1 is an underlined NOC mapped to CS, the Constitution of the International Telecommunication Union. USA/23/2 is an underlined NOC mapped to CV, the Convention. Their detail pages identify the United States as source, show 11 March 2026 as the receipt date and offer submission and proposal downloads in the Union’s six official languages.
The proposal-only English files show the underlined symbol beside each proposal number. The complete English submission makes the intent even clearer. It proposes “No Change” to the whole Constitution and Convention in order to maintain what it describes as the fundamental stability of the Union’s basic instruments. It argues that the current text supplies sufficient clarity, completeness and flexibility, and that changes to general policy or strategic direction can be made more effectively through relevant Plenipotentiary Conference resolutions and decisions.
That reason may persuade or fail to persuade the conference. This Article does not judge its merits. Its relevance is evidentiary. The file is not a silent omission. It is a reasoned institutional choice about which legal layer should remain stable and which layer should carry future change.
If the underline disappears, a reader could mistake that position for the statement that the United States offered no proposal on the Constitution or Convention. The error would not merely concern graphic fidelity. It would reverse the status of participation: an affirmative preservation demand would be read as an absence of demand.
The reverse error is possible as well. If plain NOC were treated as the underlined form, a contribution that merely clarified its scope could be inflated into a formal request that the conference entrench existing language. The text outcome remains “no change” in both cases, which makes the procedural error easy to miss.
A contribution is a chain, not a Word style
The February 2026 CPI guide shows the path a proposal takes before it appears in PMW. A delegate selects provisions from the authoritative Basic Texts and assigns actions. CPI generates a correctly formatted Word skeleton. A Reasons clause is added by default after each proposal, although it can be removed; the guide recommends retaining it so the final file can explain the purpose of the action.
The submitter then selects the source. Depending on the case, that may be the Member State attached to the user’s TIES account, an associated regional group, or selected Member-State co-signatories. For common proposals, the guide also calls for a separate account of co-signatories and the provisions to which they agree. Source is therefore part of the authority record, not decorative letterhead.
CPI assigns a provisional number and a unique ticket. The resulting Word file opens on the user’s computer and is also sent by email. It becomes the basis of the contribution. The guide warns that the proposed changes are not saved in the web interface: the author must save the Word document locally. It also says ADD, SUP and NOC proposals should not contain tracked changes.
Official submission is a later act. Administration Member State focal points send final contributions to the PP-26 secretariat. Once officially submitted, the provisional document number is replaced by the final contribution number under which ITU publishes the item. The proposal code then combines action, source, document number and the proposal’s position within that document.
The PP-26 submission page explains what happens next. ITU says using CPI facilitates secretariat processing, publication in the Union’s six languages and availability through PMW. PMW can search and sort proposals, retrieve them in those languages and inspect their “inheritance”.
Each transition adds institutional meaning: selected text, action, reason, source, co-signature if relevant, version, provisional identity, official submission, final identity, language manifestations and later inheritance. The underline is only one display element inside that chain. Making it the sole visible separator between two action types is out of scale with the record around it.
Six languages do not solve a format collision by themselves
Multilingual publication is an important safeguard. A proposal addressed to a global treaty conference should not exist only in the language of the contributor. ITU’s submission page explicitly connects CPI use to publication in Arabic, Chinese, English, French, Russian and Spanish. The two United States proposal pages expose downloads across all six.
But six renderings are not automatically six independent semantic guarantees. If every version inherits the same display convention and a later system strips the same formatting, replication has multiplied the surface without eliminating the dependency. Translation quality and action typing are separate controls.
A language file needs to know more than the visible token. It needs to know which action concept the token instantiates. That lets each locale render the concept faithfully, lets a screen reader announce the difference in words, lets an export preserve it as a value and lets a comparison tool group unlike acts separately.
The official documents page makes this point by analogy. PP-26 is paperless, but its records follow different publication regimes. Contributions and reports are published in six languages; administrative and information documents are English only; temporary working documents generally stay in their original language unless they contain input to the Final Acts; limited-distribution drafts follow another rule. A paperless system therefore depends on metadata that travels with the file. Document class is not something a reader should infer from font choice.
Proposal action deserves the same discipline. NOC can remain the concise professional symbol. Underlining can remain the familiar visual cue. The underlying value should nevertheless say in words and data whether the item means “no change proposed” or “proposal to maintain without change”.
A dual NOC receipt
The repair does not require a new decision procedure. It requires a public receipt for the one already operating.
First, assign separate stable action identifiers. Their exact names are an implementation choice. For illustration, one could mean NO_CHANGE_PROPOSED and the other MAINTAIN_WITHOUT_CHANGE_PROPOSAL. These are recommendations, not claims about present ITU codes. A display layer could map the first to plain NOC and the second to underlined NOC in every supported language.
Second, bind the action to scope. The receipt should identify the Basic Text, article, provision or complete instrument to which it applies. USA/23/1 is not a generic preference for stability; its displayed mapping and source file place it on the Constitution. USA/23/2 separately addresses the Convention.
Third, preserve the reason as a distinct field linked to the action and scope. The public record should show whether a reason is supplied, where it appears and whether it applies to one or several proposals. It should not infer a reason from surrounding prose or manufacture one when the contributor omits it.
Fourth, keep the identity chain. Source administration, co-signatories where applicable, provisional number, final contribution number, proposal number, receipt time, language version and exact content hash should remain joined. An inheritance or successor link can then show how a proposal is combined, superseded, amended, carried into a working document or disposed of by the conference.
Finally, preserve disposition without overwriting the proposal. A later record may say adopted, rejected, withdrawn, merged, deferred or otherwise treated under the controlling conference process. The original proposal remains a true historical act even when the conference chooses another outcome.
The receipt is small compared with the material already produced. Its purpose is not to turn diplomacy into database administration. It is to keep a reasoned Member-State act from changing identity when it moves from a formatted Word page into another representation.
Deadlines and the boundary of this evidence
ITU lists 9 March 2026 as the deadline for proposals to amend the Constitution or Convention, 9 July for other proposals and 19 October as the firm contribution deadline. The detail pages show the two United States proposals as received on 11 March. This Article does not infer from those dates whether the items were timely, accepted under an exception, administratively processed after an earlier submission or subject to any other rule. Receipt display is not enough to reconstruct deadline compliance.
Nor does the present PMW count define the final PP-26 docket. The observed page is a live proposal view and can change. PP-26 is scheduled to meet in Doha from 9 to 27 November 2026. As of the research cutoff, neither United States proposal has a conference disposition.
No source inspected here shows a lost action, a mistranslated NOC, an accessibility defect or an incorrect export. The Article does not allege wrongdoing. The public evidence establishes the two definitions, their visual distinction, the live underlined examples and their reasoned source document. The governance recommendation follows from the fragility of placing procedural meaning in one typographic property.
Lu Heng’s “Policy Mirror” asks institutions to show the mechanism by which a public description becomes an operative state. Here the mirror is unusually literal. Two identical letter sequences face the reader; a line beneath one carries the difference between non-proposal and proposal. A reliable system should make that difference visible, translatable and recoverable even after the mirror changes format.
Sources
- Lu Heng, “The Policy Mirror”
- Lu Heng, “The Multi-Stakeholder Mirage”
- ITU, PP-26 Proposal Management Web
- ITU, PP-26 proposal detail USA/23/1
- ITU, PP-26 proposal detail USA/23/2
- ITU, PP-26 CPI User Guide, February 2026
- ITU, PP-26 submission of proposals
- ITU, PP-26 official documents
- ITU, PP-26 key dates and deadlines
- ITU, About PP-26
Member Briefing
Deeper Profile Context
Sign in with the right membership level to unlock the full briefing and source notes.
Only for Strategic Circle
Strategic Circle
Open to all readers. Unlock profile briefings after joining and signing in.
Join Strategic CircleOnly for Leadership Alliance
Leadership Alliance
For qualified IP-asset owners and management; sign in to unlock alliance briefings.
Join Leadership Alliance
