There is a question more basic than what the United States-Iran deal means: what does it say? As of late June 2026, that question does not have a single answer, because the two parties — and the news organisations briefed by them — are circulating materially different versions of the text. Reuters, briefed by an Iranian official, reported that a draft of the memorandum includes the United States releasing twenty-five billion dollars of frozen Iranian assets. The version seen by Bloomberg contained no such stipulation. Bloomberg’s draft specified a minimum of three hundred billion dollars in reconstruction funding contingent on a final agreement; Iran’s semi-official Mehr news agency countered that negotiations would not even begin until half of Iran’s frozen funds were released, oil sanctions suspended, and the naval blockade lifted. The United States describes a sequenced deal in which Iranian compliance earns staged rewards; Iran describes a near-simultaneous exchange. These are not nuances. They are different agreements.
This site’s subject is the Strait of Hormuz, and the strait provisions happen to be among the more consistent parts of the text: all versions include reopening within thirty days, the lifting of the United States blockade, a sixty-day nuclear-negotiation window, and sanctions termination on an agreed timeline. But the consistency of the strait clause does not insulate the strait from the inconsistency of the rest, and the deeper point is structural. A deal whose text the parties dispute is, at the level of the document, the same pathology this site identified at the level of the waterway in the post on the strait’s contested status: the absence of an authoritative reference. Just as the strait had four answers to “are you open?”, the deal has at least two answers to “what did we sign?” This post reads that parallel, because it reveals something about why the institutional question is so hard to close.
Two documents, one signature
Begin with the plain fact. The memorandum was signed — electronically, as the post on the deal’s text noted — and yet the parties present different accounts of its contents. This is possible because a framework memorandum negotiated at speed, under mediation, across a deep trust deficit, can be drafted with deliberate ambiguities, side understandings, and provisions that each side reads differently. Each party then briefs its own constituency on the version most favourable to it. Tehran tells its domestic audience that twenty-five billion dollars in frozen assets is coming and that sanctions relief is near-immediate, because Tehran needs to show its hardliners a concrete win. Washington tells its audience that the rewards are sequenced and conditional on compliance, because Washington needs to show its hawks that Iran is being held to account. Both can point to the text, because the text was built to be pointed at from both sides.
The signature, in other words, did not produce an agreed document. It produced a document each side agrees to interpret in its own favour. That is a familiar feature of difficult diplomacy, and it is not always fatal — many agreements have been implemented despite ambiguity. But it means the deal does not yet have an authoritative text in the way a treaty registered and published has an authoritative text. The parties are operating from different copies, and the difference is not in the punctuation but in the twenty-five billion dollars.
The pattern: no authoritative reference at any level
Here is what connects this to the strait. The defining feature of the Hormuz crisis, in this site’s reading across more than fifty posts, has been the absence of an authoritative reference at every level of the chokepoint. There is no authority that says what the transit fee is, so the fee is whatever Iran charges. There is no authority that says whether the strait is open, so its status is whatever four contending parties claim. There is no authority that says who may transit on what terms, so access is whatever the bilateral carve-outs produce. And now, at the level of the deal meant to resolve all of this, there is no authoritative text, so the agreement is whatever each party says it signed.
The pattern is the same at every level: a question that should have one authoritative answer has as many answers as there are interested parties, because no institution exists to produce the single authoritative answer. The strait has no chokepoint authority, so the strait’s status, fee, and access are all contested. The deal has no agreed authoritative text, so the deal’s terms are contested. The absence of an authoritative reference propagates upward from the waterway to the document meant to govern it. A crisis that began because no one was in charge of the strait is being resolved by a deal that no one can authoritatively state.
Why this makes the strait institution harder to build
The disputed text bears directly on the prospect of building the chokepoint institution the site has argued for. The institution is supposed to be constituted through the sixty-day dialogue the deal’s text promises. But if the parties dispute what the deal’s text says — if Iran believes the deal entitles it to immediate frozen-asset releases and near-term sanctions suspension, while the United States believes those are conditional on sequenced compliance — then the foundation on which the strait dialogue is supposed to proceed is itself contested. The strait institution cannot be built on an agreement whose terms the builders dispute.
The 22 June closure over Lebanon, analysed in the companion post on the strait as hostage, is partly a product of exactly this. Iran closed the strait claiming the United States violated the deal by failing to rein in Israel; the United States presumably reads its obligations differently. When the text is disputed, every party can claim the other violated it, and the strait — being the most powerful lever — becomes the venue where the dispute over the text is litigated. A disputed deal produces a contested strait, because the strait is where the parties apply pressure when they disagree about what they signed.
What an authoritative text would require
The remedy at the document level is the same kind of thing as the remedy at the waterway level: a single, authoritative, published reference that all parties are bound to. Treaties achieve this through registration, publication, and a recognised text in defined authentic languages, deposited with a neutral authority. The Geneva or Bürgenstock signing could, in principle, have produced such a text — a single registered document, published, with the strait provisions and the asset provisions and the sequencing all fixed in agreed language. It did not, because the deal was a framework signed in haste with deliberate ambiguities, briefed differently to different audiences.
The forthcoming final deal — the one the sixty-day roadmap is meant to produce — needs to be that authoritative text if it is to support a durable institution. A final agreement with a single published text, fixing the strait’s administration and the asset releases and the sequencing in agreed language, would give the chokepoint institution a foundation to be built on. A final agreement that reproduces the framework’s ambiguities, with each side again briefing its own version, would leave the institution to be built on contested ground, and would leave the strait available, once again, as the lever the parties pull when they disagree about what was agreed.
The deeper lesson
The two versions of the deal are a small story — a discrepancy over twenty-five billion dollars that careful diplomacy will probably paper over. But they illustrate the deepest lesson of the whole crisis. Institutions exist to produce authoritative answers to questions that would otherwise be contested by interested parties. A chokepoint authority produces the authoritative answer to what the fee is, whether the strait is open, who may transit. A registered treaty produces the authoritative answer to what was agreed. The Hormuz crisis has been, from the beginning, a demonstration of what happens in the absence of such authoritative answers: every question fragments into as many answers as there are parties, and the strait becomes the place where the fragmentation is fought out. The deal with two versions is the latest instance. The remedy, at every level, is the same: an institution, and an authoritative text, that all parties are bound to. The comparison page sets out the institution. The rate schedule sets out the fee that would no longer be contested. The calculator prices a transit against it. A deal no one can authoritatively state is the document-level version of a strait no one authoritatively governs.
Sources: Fortune, “Iran pushes differing versions of deal as U.S. sticks to timeline,” 14 June 2026; Reuters and Bloomberg reporting on divergent drafts of the memorandum of understanding, including the disputed twenty-five-billion-dollar frozen-asset release and the three-hundred-billion-dollar reconstruction provision; Iranian Mehr news agency statements on preconditions for negotiations; CSIS, “The United States and Iran Announce a Deal to End the War”; this site’s prior analyses on the strait’s contested status (22 June), the deal’s no-toll text (22 June), and the companion post on the strait as hostage to the Lebanon front.