Summary

  • On 1 June 2017 in Nairobi, AFRINIC’s annual general members’ meeting heard the 2016 audited report and recorded no objection, but its minutes recorded no motion, seconder or vote adopting the statements. Twenty-two days later in Johannesburg, legal counsel advised using a written-resolution route through the directors’ separate status as Registered Members; one director declined to vote, and Resolution 201706.362 was recorded as passing by 78 per cent.
  • That sequence presents a strong benign cure case. Written member resolutions were contemplated by the Constitution and Mauritian company law, the cited threshold was a substantial 75 per cent, and a written route could avoid the cost of calling another meeting. But dual status did not make Board capacity and member capacity interchangeable. The cure depended on proof that the right electorate assented to one identified text under the member rule.
  • The surviving public record gives the rationale, attendance picture, abstention, proposer, seconder and result. It does not display the separate written instrument or like-form assents, the exact eligible-vote denominator, version identity, individual assent times, a distinct passage time, or proof of delivery to non-signers within seven days. The durable answer is a compact cure receipt that preserves both the missed vote and the later member act without turning accounts adoption into a waiver, release or claim of public authority.

L3 — The vote that never happened

At 14:23 on 1 June 2017, AFRINIC’s annual general members’ meeting opened in Nairobi. The financial update included presentation of the 2016 audited report. The chair asked whether there were objections, and the minutes record none. Then the record moves on without the formal sequence that normally makes an institutional decision visible: no adoption motion, no seconder and no passed resolution for the statements. Those features do appear around other decisions in the same meeting record. Their absence here is therefore meaningful.

It does not prove what every person in the room thought, but it does show what the durable corporate record failed to capture.

The omission was not invented by a later observer. AFRINIC’s own subsequent resolution said that no formal vote to adopt the financial statements had been taken at the meeting. That admission matters because it prevents the phrase “no objections” from doing work it cannot properly do. A room can listen without voting. A chair can invite objections without putting a motion. Members can be comfortable with a presentation without performing the legal act assigned to them. Silence may be evidence of no expressed opposition; it is not adoption, ratification, waiver, release or discharge.

Twenty-two days later, on 23 June, the scene shifted to Johannesburg. Eight directors were present at a meeting of a Board composed of nine directors, including the chief executive; one director was absent with an apology, and legal counsel attended. The chair recalled the missing formal adoption. Counsel advised that section 117(1) of the Mauritius Companies Act could be used by the Registered Members. Subramanian Moonesamy said he would not vote because he had not attended the Nairobi meeting. Sunday Folayan proposed Resolution 201706.362, Seun Ojedeji seconded it, and the minutes recorded passage at 78 per cent.

That account is neither nothing nor everything. It is meaningful evidence that the omission was noticed, a specific route was selected, one participant’s non-vote was recorded and a result above the cited threshold was entered into the minutes. It supports a reasonable reading that the institution was attempting to cure a procedural miss rather than conceal it. Yet the very route selected makes the identity of the actor essential. The same people were gathered as directors, but the power invoked belonged to members acting under a written-member procedure. A minute held in a Board-meeting record can report a member act.

It cannot, merely by naming one, erase the need to show how that member act occurred.

The distinction is practical, not semantic. AFRINIC’s 2016 Constitution made a director a Registered Member upon assumption of office and required the relevant membership forms. It also made the chief executive one of nine directors. The natural persons around the Board table could consequently possess two corporate capacities. In one capacity they could deliberate and vote as the Board. In another, they could sign or assent as Registered Members.

The Constitution applied different decision rules to those capacities: an ordinary Board resolution used a majority of eligible Board votes; a written Board resolution used a two-thirds rule; a written member resolution used a 75 per cent floor. Calling all three “a resolution” would hide the rule that actually authorised each act.

Dual officeholding is common enough in private organisations and is not inherently suspicious. A small membership corporation may sensibly give directors a membership status that lets necessary corporate acts be completed without assembling a much larger constituency for every correction. Efficiency, however, depends on clean records. When one individual changes capacity without changing seat, location or meeting, the document must do the separating. It should say: this person is not now casting a Board vote under the Board rule; this person is assenting as an eligible Registered Member under the member rule.

The more visually continuous the meeting, the more explicit the documentary change should be.

The Nairobi record establishes the first half of the chain. It identifies the statements presented and preserves the absence of a formal adoption vote. The Johannesburg minutes establish much of the second half. They name the statutory advice, Registered-member route, abstention and percentage. The unresolved issue lies in the join between those records. Did the eligible members sign or otherwise assent to a frozen written resolution? What precisely counted in the denominator? When did the threshold become satisfied? Were all assents attached to the same text and the same statement set?

Were non-signers sent the resulting resolution within the prescribed period? Those are not demands for ceremonial excess. They are the facts that distinguish a written member act from a Board minute describing one.

Later treatment of the minutes cannot replace that join. The Board subsequently approved the 23 June minutes with amendments through Resolution 201709.365, though the surviving set does not identify which passages were amended. The 2017 annual-meeting minutes were later published and treated in the 2018 meeting context as the formal account of the earlier gathering. These later steps reinforce the status of the minutes as records of what AFRINIC said occurred. They also reinforce the original omission: the approved annual-meeting account still contains no adoption vote.

Authenticating a record is not the same as supplying any underlying assent document, signature, notice or version marker that the record does not reproduce.

It helps to put the narrow question in ordinary language. Suppose a private association forgets to ask its members to approve a document at the meeting where the document is presented. Its rules allow the same matter to be decided in writing. The association promptly recognises the mistake, circulates a resolution, secures the required percentage and informs those who did not sign. That is a sensible repair.

But if its public account shows only a later committee meeting where the committee members, who also hold membership status, are said to have passed something, a reader needs a modest bridge: the written text, the eligible electorate, the assents, the arithmetic and the notice. Without that bridge, the account is plausible but not fully inspectable.

This is why the missed vote must remain visible. A clean institutional history does not rewrite the Nairobi silence into approval after the fact. It says first that presentation occurred and no objections were recorded; second, that the adoption act was omitted; and third, that a separate written-member procedure was used in response. Preserving all three facts makes the cure stronger. It avoids the temptation to treat institutional continuity as a licence to smooth away an inconvenient break in the record. The ledger serves reality only if it records the break as well as the repair.

AFRINIC’s function supplies context but no enlarged power. It is a private, membership-based technical bookkeeper and coordinator for Internet number-resource records. Accurate registrations, dependable directory services and disciplined administration are useful to operators and members. Those functions do not confer sovereignty, legislative jurisdiction, public regulatory power, policing, punishment, confiscation or adjudication over number resources. Resolution 201706.362 concerns a private corporate act: adoption of identified financial statements through a member mechanism.

Its credibility rises or falls with the proof of that mechanism, not with any larger rhetoric about a region or a community.

The narrowness works both ways. Rejecting inflated authority does not make lawful private powers unimportant. A company’s Constitution and the law governing it can assign real powers to directors, Registered Members and Resource Members. Those powers affect corporate administration and should be taken seriously. Precisely because they are real, their boundaries matter. A member power should not be absorbed into a Board label, just as a Board procedure should not be mistaken for a public mandate. Respect for private ordering begins by naming the principal, capacity and instrument correctly.

The public record therefore supports a measured preliminary finding. The 1 June meeting did not leave a recorded adoption act. The 23 June material records a deliberate response under a written-member route and supplies several concrete facts that favour the benign account. What it does not supply publicly is the complete documentary chain contemplated by that route. The right inquiry is not whether institutional language sounds confident. It is whether the members’ act can be reconstructed from one text, one eligible electorate, recorded assents, transparent arithmetic and completed notice.

The point is especially important because “78 per cent” looks exact. Precision can reassure, but a percentage is only the visible end of a calculation. To evaluate it, a reader needs the numerator, denominator and eligibility rule. The nine-seat composition, eight directors present, one absence and one abstention make seven affirmative votes out of nine a strong arithmetic inference: seven divided by nine rounds to about 78 per cent. The minutes, however, do not publish an express roll call saying that seven named people assented, nor do they establish publicly that nine was the entire eligible electorate for this particular member act.

The inference is useful; it must remain an inference.

That restraint protects everyone involved. It avoids converting a gap in publication into an allegation that the underlying act never happened. It also avoids converting a persuasive minute into proof of formalities the minute does not show. Sunday Folayan’s proposal and Seun Ojedeji’s second are recorded facts, not grounds for speculating about motive. Subramanian Moonesamy’s non-vote is attributed to his absence from the Nairobi meeting, not to disagreement with the statements or the chosen route. Ashok Radhakissoon’s advice is a position reported in corporate minutes, not a judicial determination.

Careful attribution is the difference between analysis of a record and invention around it.

The cure question can thus be framed without accusation. Was there a written or otherwise properly assented member resolution tied to the same statements, supported by at least the required share of eligible votes and followed by the required communication to non-signers? The published materials provide a substantial outline, but not every item needed to answer that question from the public record alone. That is a proof gap, not a verdict.