Summary

  • On 10 May 2018, “None of the above” received 78 votes in AFRINIC’s Western Africa Seat 2 election, against 56 votes for Ousmane Moussa Tessa and 43 for Adewale Adedokun. The seat remained vacant. That outcome is the strongest evidence that the emergency nomination process could present choices but could not compel members to accept one.
  • The emergency was genuine. A mid-cycle vacancy placed the Western Africa seat before a standing Nominations Committee that already contained two people domiciled in Western Africa. Article 9.1 of the 2016 bylaws made them ineligible to serve on a NomCom handling an open seat from their region.
  • Resolution 201803.401 created a separate, Seat 2-only committee after the Board considered replacing the conflicted standing members, sought member comments and called for volunteers. The arrangement preserved the standing committee’s other election work, excluded Western Africa residents, kept staff out of candidate-merit discussions and returned the final choice to members.
  • The cure nevertheless occupied a bylaw gap. The 28 March minutes recorded legal advice that the bylaws contained no provision for an additional NomCom, even though the advice reportedly found no legal barrier. The Board appointed all four members, gave its representative a vote, retained power over guidelines, commented on candidate criteria and the draft slate, and approved nominations as inputs.
  • The distinction between input and output is essential. The Board approved three nominations subject to the ad hoc committee eliminating any that failed applicable requirements; the committee later finalised a two-candidate slate. The public record does not establish why one nomination disappeared, and it does not support a claim that the Board chose the final slate.
  • AFRINIC’s authority in this episode was private and corporate. A sound internal election could authenticate an office-holder within the company; it could not turn a technical registry bookkeeper into a sovereign regulator, punishment authority or public-law court for African networks or number resources.

L3 — The vacancy that conflicted the standing committee

The election ended with an arithmetically clear refusal. At the 10 May Annual General Members’ Meeting, Adewale Adedokun received 43 votes, Ousmane Moussa Tessa received 56, and “None of the above” received 78. There was no winning director for Western Africa Seat 2. In a process designed under intense time pressure, the most important institutional fact was therefore not that AFRINIC managed to put two names on a ballot. It was that the ballot still allowed members to decline both.

That conclusion began with a conflict the ordinary 2018 election plan had not been built to absorb. The standing NomCom’s regional affiliations had been corrected in February by Resolution 201802.390, which recorded Komi Elitcha and Wale Adedokun as its two Western Africa members. A Western Africa Board vacancy then arose unexpectedly in late March. Resolution 201803.399 sent that vacancy to the May members’ meeting and tasked the NomCom and staff with organising the election.

But the act of sending it into the normal nomination channel exposed a problem: Article 9.1 barred a person domiciled in a region whose Board seat was open for renewal from serving on the NomCom.

Two people on the standing committee were therefore regionally ineligible for this one additional contest. Leaving them in place for Seat 2 would have run directly against the eligibility rule. Yet removing them from the standing committee altogether could have unsettled its work on the elections already under way. The Board faced a practical collision between compliance with a specific regional restriction and continuity in the rest of the annual nomination process.

The 28 March special-meeting minutes show that the Board did not treat the answer as obvious. It considered asking the two conflicted members to resign and appointing replacements. It also considered creating a parallel committee for Seat 2 while leaving the standing body to complete the other elections. The minutes recorded legal advice in an unusually revealing form: the bylaws had no provision for an additional NomCom, but the advice identified no legal barrier to establishing one. The first part of that sentence matters as much as the second.

A reported absence of prohibition is not the same thing as an express grant of power, particularly when the improvised body controls access to an elected office.

The Board chose the parallel route and opened a short member-comment period, ending on 30 March at 20:00 UTC. On 2 April, AFRINIC issued a public call for three community volunteers for the ad hoc NomCom, with expressions of interest due the following day at 12:00 UTC. Resolution 201803.401 was then approved by email on 5 April; later minutes recorded the vote as unanimous. It appointed Mustapha Ben Jemaa, Douglas Onyango and Didier Kasole as community representatives and S. Moonesamy as the Board representative. The three community representatives were drawn from Northern, Eastern and Central Africa rather than Western Africa.

The speed is visible without embellishment. Member comments closed roughly two days after the Board discussion. The public volunteer window ran from 2 April until midday on 3 April. The resolution followed on 5 April. Guidelines dated 10 April required a final slate no later than 25 April so that the notice for the May meeting could issue on 26 April. A compressed schedule may have been understandable given the approaching meeting, but it reduced the time available for members to scrutinise the legal construction, volunteer pool, appointments and rules.

The record does not prove that a longer process would have produced different people or a different slate. It does establish that the public-facing parts of the emergency mechanism were brief.

The strongest benign case deserves to be stated in full. The conflict was created by a vacancy that emerged after the ordinary election machinery had been constituted. Article 9.1’s regional rule could not simply be ignored. Replacing two standing members could have disrupted nominations for other seats and would still have required Board-controlled replacements. A separate committee isolated the conflict instead. Its remit was limited to Seat 2, its community members came from outside Western Africa, and it was expected to act independently of the standing NomCom.

Staff could support the process but were barred from discussions of candidate merits. Most importantly, the Board did not use the emergency to fill the vacancy itself. The process led back to a member ballot that included a rejection option.

That is a credible case for an emergency bridge. It is not a licence to treat every design choice as legitimate merely because the underlying difficulty was real. Institutional necessity can explain why an actor moved; it cannot supply authority that the governing instrument omitted. The lack of an express additional-committee clause made narrowness, transparency, fixed procedure and an effective end point more important, not less. The committee could be defended only as the least discretionary machinery sufficient to cure this particular conflict and return this particular vacancy to the electorate.

The counterfactuals clarify that standard. Allowing the conflicted Western Africa members to process Seat 2 would have contradicted Article 9.1. Replacing them on the standing committee might have avoided creating a second body, but could have disturbed other elections and would not have removed the Board’s appointment power. A direct Board appointment under a casual-vacancy mechanism would have produced faster continuity, but it would have displaced the member election that Resolution 201803.399 had triggered.

The parallel NomCom stood between those alternatives: more faithful to regional eligibility than ignoring the conflict, less disruptive than reconstituting the annual committee, and more respectful of membership choice than a direct appointment.

Its legitimacy consequently depended on conditions that could be checked. Was the mandate confined to one seat? Were conflicted regional participants excluded? Did the committee retain responsibility for the final slate? Could the Board’s interventions be identified rather than hidden? Did the mechanism expire when its assigned election ended? Could members reject the whole offering? On the available record, several of those controls were present. Others were weakened by the Board’s structural proximity and the short timetable. The correct analysis is therefore neither institutional celebration nor blanket condemnation.

It is a bounded examination of who controlled each stage.

That examination must also separate corporate governance from public authority. AFRINIC may organise nominations, maintain membership records, conduct a meeting and determine who occupies an internal Board seat under its private governing arrangements. Those are meaningful corporate functions. They can affect operational confidence because network operators rely on accurate, stable number-resource registry services. But they do not confer legislative, regulatory, police, prosecutorial, punitive, confiscatory or public-law adjudicative power.

No resolution number, consultation window, committee title or member vote transforms a technical bookkeeper into a sovereign institution.

This distinction is not a semantic qualification placed at the edge of the story. It determines the scale of the power under review. The ad hoc NomCom controlled access to candidacy for a private-company office. That gate deserved scrutiny precisely because AFRINIC’s registry service matters operationally and because Board composition can influence the reliability of the company that provides it. Yet the consequence of a sound election was an internally authenticated director, not a ruler of a territory or regulator of African networks. Member participation was important within the company; it did not turn a service region into a polity.

The February correction to the standing NomCom’s regional affiliations further shows why careful records matter. Article 9.1 attached a concrete eligibility consequence to domicile when a seat from the same region was open. Once Seat 2 unexpectedly entered the cycle, the recorded affiliations ceased to be descriptive trivia. They became the reason the ordinary committee could not process the new vacancy unchanged. Accurate corporate bookkeeping made the conflict visible. It did not determine the cure, and it did not validate the cure by itself.

Nor should the official character of the source material be mistaken for proof of neutrality. Board resolutions establish what the Board adopted. Minutes establish what the organisation recorded about its discussions, votes and later steps. The volunteer call establishes the terms and deadline it announced. The guidelines establish the procedure it prescribed. None of these records independently proves that every member comment was considered, every potential volunteer had a fair opportunity, every deliberation was neutral or every legal conclusion was correct.

The records are indispensable evidence of acts and claimed safeguards, but institutional authorship sets a limit on what they can prove.

There are material unknowns. The available record does not reproduce the legal opinion whose conclusion the minutes summarised. It does not reproduce all member comments. It does not establish the number or identities of every expression of interest. It does not disclose the committee’s candidate scoring or deliberations. These gaps do not justify an accusation of misconduct. They do mean that confidence must rest on the visible design and observable outcome, rather than on an assumption that an official process is self-validating.

The opening problem can therefore be stated precisely. A private company confronted a genuine conflict in its own election apparatus. It selected an emergency gap-filling design that was not expressly described in its bylaws. That design was potentially defensible because it isolated the conflict, preserved other election work and returned the choice to members. It was potentially dangerous because incumbents retained extensive influence over the body that determined which candidates reached the ballot. Everything that follows turns on whether the process kept that influence bounded and visible.