When Does Collective Authorization Become Legitimate?
Voting is often treated as if it automatically authorizes what follows. But does it?
The controversy surrounding FIFA’s proposed Forward Enterprise (FFE) is usually described as a problem of governance, transparency, or valuation. These are reasonable questions, and FIFA has given reasonable people reason to ask them. But they treat the vote itself as settled — as if the only open question is whether the terms are good, once we accept that a vote on them is the right kind of event to be having.
Boundaryism asks a prior question. Not: was this decision consented to. But: what kind of decision was it, and does a vote of this shape authorize a thing of that shape at all.
Consent to an act is not authorization of an agent
If I sell my house, I have consented to a transaction. The transaction has a beginning and an end. Once it closes, nothing further is asked of me, and nothing further is decided on my behalf.
If I authorize someone to sell my house on my behalf, going forward, whenever they judge the time is right — that is a different act. I am no longer consenting to a sale. I am transferring the standing to make future decisions I will not individually approve. The difference is not one of degree. A completed sale binds me to one outcome. An authorization binds me to a stream of outcomes I have not yet seen, made by a judgment I will not be asked to ratify again.
FFE is not the first kind of act. The 211 member associations being asked to vote are not being asked whether $20 billion is the right number. They are being asked to transfer standing over a set of ongoing questions: who governs FFE, who interprets the valuation as circumstances change, who decides on further financing, who amends the terms once the entity exists. The disclosed proposal does not specify that these questions will return to a vote. What is being requested is not approval of a transaction. It is authorization of an agent.
This distinction matters because the two acts fail for different reasons. A transaction can be criticized for its terms. An authorization has to be criticized, first, for whether the conditions under which handing over standing to someone else could ever be legitimate were present at all — before any question of terms arises.
What a legitimate authorization requires
An authorization of this kind — a principal handing an agent standing to make future binding decisions — is a familiar structure. It is what happens whenever a shareholder appoints a board, a citizen elects a representative, or a client signs a power of attorney. None of these are inherently illegitimate. What makes them legitimate, when they are, is not the presence of a vote but the presence of conditions that keep the transferred power answerable to the party who transferred it.
At minimum, four conditions do this work.
A known agent. The principal must know who is being authorized before the authorization is given — not necessarily every future officeholder, but at minimum the structure through which the agent will be selected and held to account. A power of attorney naming no attorney authorizes nothing; it is a blank check with no name on it.
A bounded scope. The authorization must specify what the agent may decide, not merely that the agent may decide. A board given authority to manage a company’s operations has not been given authority to sell the company; the scope of the grant is what makes the grant reviewable at all, because only a bounded grant can be checked for whether the agent stayed inside it.
Revocability. The principal must retain some realistic path to withdraw the authorization if the agent’s conduct warrants it — resignation, recall, sale of the shares, a vote of no confidence. An authorization with no exit converts what looked like a delegation into something closer to a permanent transfer of standing, made once and never revisited.
Standing oversight. Short of revocation, the principal must retain some ongoing capacity to observe and act on what the agent does with the power — a seat, an audit right, a reporting requirement, a mechanism to object. Without this, the interval between authorizations, however rare the vote, is a period in which the agent is, in practice, unsupervised.
Where all four are present, an authorization can be legitimate even though the principal will not vote again on each individual decision the agent makes — this is the ordinary condition of functioning representative institutions. Where more than one is absent, the vote that produced the authorization was doing more work than the principal could have knowingly consented to.
Where FFE stands against this
On the materials presently disclosed, FFE fails, or leaves unanswered, each of these conditions, and does so in the same direction each time: toward removing the very thing that would let the associations act on the power they are being asked to hand over.
No agent has been named. FIFA’s own materials confirm that FFE will be run by a management team and board not yet appointed — the associations are asked to authorize an entity whose future decision-makers cannot be identified, let alone evaluated, before the vote closes.
No scope has been bounded. The proposal describes what FFE will hold — the commercial rights bundled from the World Cup and other properties — but not what limits, if any, constrain how those rights may later be used, refinanced, or further diluted. A grant with no stated ceiling cannot later be shown to have been exceeded.
No revocability has been offered. Once equity is sold to outside investors, that transfer is not something a future vote of the member associations can unwind. Whatever standing is given up in October is given up for the life of the entity, not for a term subject to renewal.
No standing oversight exists. The associations have been given no board seat, no independent access to the numbers behind the valuation, no defined right to challenge how FFE is later governed. What oversight there is runs through FIFA’s own reporting, to the associations, about an entity FIFA itself will have helped create and staff.
Each of these, individually, might be an oversight correctable before the vote. Together, they describe an authorization that asks the associations to transfer standing over a named set of future decisions while removing every mechanism that would let them know whether that standing was later honored.
The sequence this belongs to
None of this requires bad faith on FIFA’s part, and Boundaryism does not need it. A threshold theory asks whether the structure of a decision made meaningful authorization possible, not whether anyone involved intended otherwise. The same four gaps, present for entirely defensible reasons, would fail the same test.
Nor does this settle whether outside capital is a bad idea for FIFA’s commercial rights. It might not be. What the four conditions specify is not a verdict on the transaction but a standard for the act of authorizing it — a standard that, once met, would let the associations find out later whether the authorization was honored.
That last point is worth sitting with. An authorization that satisfies these conditions does not just protect the principal at the moment of the vote. It preserves their standing to ask, later, whether the agent’s conduct stayed inside what was granted — and to have that question answered by someone other than the agent itself. An authorization that fails these conditions forecloses that question in advance. The agent who receives an unbounded, unrevocable, unsupervised grant of standing does not just gain the power to act. It gains the power to decide, on its own account, whether its actions were ever in question at all.
That is where this essay’s question meets the one that follows it: once an authorization of this shape has been granted, who is left in a position to ask whether it was ever exercised legitimately?
This essay refers to FIFA's July 2026 public materials on FIFA Forward Enterprise and contemporaneous reporting on the proposal. The distinction between consenting to an act and authorizing an agent draws on a distinction familiar from agency law and, in a different register, from Hobbes's account of authorization in Leviathan — where the represented becomes, formally, the author of what the representative does. Neither is discussed here at length; the aim is the structural distinction, not the genealogy.