Summary

  • AFRINIC’s Board announced Prichard Chakadenga, Solomon Dindi, Ganesh Ramalingum and Ikibah Ebi Benjamin as its four-person 2026 Nomination Committee under Article 9. The committee later screened candidates for three Governance Committee seats, two NRO Number Council or ICANN ASO Address Council seats, and two PDP Co-Chair positions.
  • The rules provide a public volunteer call, neutrality and impartiality duties, a ban on a NomCom member becoming a candidate in a covered election, conflict disclosure, unpaid service, eligibility checks and functional separation from ballot administration. Yet the reviewed public record does not reveal the volunteer denominator, the Board’s comparative rationale or vote, the four appointees’ composition categories, individual conflict or recusal records, or the denominator and reasoning behind the final eligible slate.
  • Those gaps do not prove illegality or misconduct. They show why written independence and evidenced independence are different things. A narrow process record could make gatekeeper selection and screening auditable without exposing private applicant files.

Analysis

At 09:37:16 UTC on 1 April 2026, AFRINIC’s announcement list carried a short institutional act with long consequences. The Board had appointed four people to the Nomination Committee for the year: Prichard Chakadenga, Solomon Dindi, Ganesh Ramalingum and Ikibah Ebi Benjamin. The notice was signed by Board Chairman Prof. Adewale Adedokun and invoked Article 9 of AFRINIC’s Constitution or Bylaws. A corresponding web article appeared on 2 April. The two versions differ on the honorific attached to Prichard Chakadenga, so the responsible course is to use the name without inferring gender. They do not differ on the appointment itself.

The notice did more than fill four volunteer positions. It installed the screeners for three different institutional contests covering seven places: three seats on the Governance Committee, two seats on the NRO Number Council or ICANN ASO Address Council, and two PDP Co-Chair positions. These offices do not all lead to the same final selection mechanism. Eligible Resource Members vote in the member-elected contests; the Policy Development Working Group selects PDP Co-Chairs through consensus.

Yet all three routes first passed through a four-person committee empowered to process nominations, verify eligibility, consider suitability, finalise a slate and supervise the process without itself administering the ballots.

This is why the appointment deserves more attention than its brief announcement received. Elections are usually described from the voter’s end: who stood, who voted and who won. The more revealing institutional moment can occur earlier, when someone decides who is entitled to stand at all. A ballot can offer a choice only among the people admitted to it. A consensus floor can select only among the people found eligible to reach it. The architecture of nomination is therefore not clerical decoration. It determines the boundary of the later choice.

That proposition must be kept within its proper limits. AFRINIC is a thin bookkeeper and coordinator for number-resource uniqueness and accurate records. It is not a sovereign ruler over networks, operators or Internet number assets. A committee appointment made under a corporate bylaw is a corporate procedural fact. It is not a popular mandate over Africa, a title to number resources, or authority to punish a network. Courts, regulators and governments acting within their lawful powers can punish. A registry may coordinate, keep records accurate and protect uniqueness.

Its internal election machinery may choose corporate and policy officeholders. It must never convert candidate-screening discretion into a route for resource punishment or a self-authenticating claim of sovereignty.

The appointment chain

The formal chain begins with AFRINIC’s members, who adopted and amended the Bylaws. Article 9 then places the appointment of NomCom in the Board’s hands. It also places two further links around the committee: NomCom operates under guidelines prescribed by the Board and reports to the Board. In other words, appointment, rule prescription and reporting are all Board-centred. Within that chain, NomCom receives substantive screening functions. The Election Committee later handles voter verification, ballot issuance, vote collection, counting and tabulation.

The final selectors—Resource Members in the elected contests and the PDWG in the Co-Chair process—act only after screening.

The annual timetable began before the appointment notice. On 4 February 2026, the Board approved the Election Guidelines 2026. Seven calendar days later, on 11 February, it issued a call for volunteers for four vacant NomCom positions. The call asked for an expression of interest and a short biography. It described neutrality, no direct or indirect interest in an election’s outcome, trustworthiness, institutional knowledge, diligence and good faith as expected attributes. The deadline was 20 February at 23:59 UTC. From the stated call date to the deadline was nine elapsed calendar days, or ten calendar dates if counted inclusively.

The appointment announcement came forty calendar days after that deadline. The public record reviewed through 9 August does not show a separate decision date. It does not include a Board resolution, minutes, a vote record or a selection memorandum. Nor does it disclose how many people answered the call, who they were, whether the Board interviewed them, which comparative rubric it used, how individual applicants scored, or why the four appointees were preferred. Those are public-record absences, not evidence that no internal process existed.

The distinction is essential. There may have been applications, deliberation, checking and conflict handling that were not published. A missing public record cannot responsibly be turned into an allegation about a named person. Equally, an announcement that expresses confidence in the appointees’ impartiality cannot substitute for evidence of the process by which that confidence was formed. Institutional trust is strongest when readers need not choose between blind belief and reckless suspicion. A compact audit trail can occupy the space between them.

By 8 May, the committee was operational. A call for candidates identified Ganesh Ramalingum as Chair, although the appointment notice had named four members without allocating offices. The reviewed record does not show the 2026 chair-selection step. AFRINIC’s older standing NomCom process page says community representatives select the Chair, but that page was last modified in 2021 and cannot simply be assumed to be the controlling 2026 record. The Guidelines also make the Chair one of the Election Trustees, alongside AFRINIC’s external legal adviser and the CEO or a representative of an Interim Managing Committee.

The missing chair-selection record therefore concerns more than a ceremonial label.

The 8 May call opened nominations for the seven positions and restated support, eligibility and malpractice-screening conditions. The stated original nomination deadline was 29 May. On 5 June, NomCom extended it to 8 June, a ten-day movement. That extension overlapped the screening timetable originally due to begin on 30 May; the public notices reviewed do not fully explain the revised internal vetting schedule. This does not establish a defect. It identifies a practical point that a post-process report could clarify.

On 17 June, NomCom published its final eligible slate. It contained eight names for seven places: five eligible people for three Governance Committee seats, two for the two NRO NC or ASO AC seats, and one for two PDP Co-Chair positions. The arithmetic is clear. The acceptance rate is not. The publication gives the numerator but not the nomination denominator. It does not identify excluded nominees, the reasons for any exclusion, whether interviews were used, or a nominee-level route by which a screening decision could be reviewed. It is therefore impossible, from the sealed evidence, to say how many nominations were excluded or why.

Screening is not ballot administration

The division of labour in the 2026 Guidelines is real and valuable. NomCom supervises and screens. The Election Committee administers voting, verifies voters, issues ballots, collects votes and counts them. The Election Committee is described as functionally independent while remaining administratively accountable to the CEO or equivalent. Election Trustees oversee opening and closing. Members or the PDWG then make the ultimate selections according to the relevant contest.

This separation reduces the risk that the same people decide candidate eligibility, control voter access and count the result. It is one of the strongest points in defence of the design. But functional separation at the ballot stage does not answer the independence question at the screening stage. The Board appoints NomCom, prescribes its annual guidelines and receives its report. NomCom decides which eligible people reach the later decision. The relevant dependency is therefore not total control by a single actor; it is a chain in which the appointing Board surrounds a short-lived gatekeeper that exercises consequential discretion.

The gatekeeper’s powers extend beyond checking whether boxes have been ticked. Under Article 9 and the Guidelines, NomCom may call for candidates, process nominations, prescribe or apply criteria and qualifications, interview nominees, verify eligibility, reassess suitability where malpractice is in issue, finalise candidate lists and supervise the polls conducted by the Election Committee. Some of these functions are objective. A candidate either has the required member support or does not. Others require judgment. Suitability and malpractice screening cannot be reduced to a database lookup.

Judgment is not itself improper. Every serious nomination process needs people to distinguish complete applications from incomplete ones, verify supporters, handle conflicts, protect personal information and stop objectively ineligible or abusive nominations from consuming a ballot. A process without any gatekeeping can become unmanageable or unfair. The question is not whether discretion exists. It is whether discretion is confined, evidenced and reviewable at the points where it matters.

The thin-bookkeeper boundary sharpens that question. AFRINIC may need corporate officeholders to oversee an organisation that performs number-registry services. It may need a functioning policy-development process. It may coordinate elections for those defined offices under its corporate instruments. Yet the legitimacy generated by those procedures is strictly internal and procedural. It does not elevate NomCom, the Board, the NRO, an RIR document or a claimed community into a sovereign source of power over operators’ assets. The most an official document proves is what the issuing institution wrote, announced or did.

It does not prove that the institution’s self-description is legitimate merely because the description appears in an official publication.

This prevents a subtle but serious circularity. If the Board selects the people who screen candidates, writes their rules and receives their report, and if the resulting slate later helps populate institutions that endorse the system, the completed process cannot retroactively prove the appointing authority’s wider legitimacy. Procedure can establish that a procedure occurred. It cannot enlarge the institution’s mandate beyond the function the procedure was authorised to serve. Candidate screening must never become self-authenticating authority.

The safeguards on paper

The 2026 structure does contain meaningful protections. The Board made a public call for volunteers. The Bylaws prohibit a NomCom member from standing as a candidate in an election for which that NomCom must find candidates. The Guidelines require impartiality, fairness, consistency and equal treatment. They require actual or potential conflicts of interest to be disclosed. The volunteer call expected no direct or indirect interest in an election outcome. Service is unpaid, though logistical and reasonable support may be provided. Eligibility is to be checked against authoritative AFRINIC records.

Candidate information, calls, criteria, timelines, slates, results and statistics are contemplated for publication. Ballot administration is separated from NomCom’s screening role.

One safeguard carries a particularly clear institutional memory. On 30 September 2020, AFRINIC’s members approved a special resolution adding Article 9.1A. It prevents a NomCom member from being a candidate in an election for which the committee must find candidates. The explanatory material described the problem as one person being both judge and party. This is a narrow but sound prohibition. It recognises that even a well-intentioned person should not control access to a contest in which that person seeks office.

The rule is also a useful illustration of what a safeguard can and cannot do. It removes the most direct conflict: screener as candidate. It does not by itself address every indirect interest, institutional relationship or appearance of bias. That is why the Guidelines’ broader conflict-disclosure duty matters. Yet the reviewed text does not say that disclosures must be public, identify the recipient, prescribe a form or time, or set out recusal and replacement mechanics. No individual declaration or recusal register for the four appointees was located in the reviewed public record.

Again, this is not proof that disclosures were not made privately. Some conflict information may properly remain confidential. It is a finding about evidence available to members, nominees and operators assessing the process from outside. Written duties are necessary. A visible account of how the duties were applied is what makes them auditable.

Unpaid service is another real protection, but not a complete theory of independence. Removing remuneration can reduce one direct incentive. It cannot eliminate reputational, institutional or relationship incentives. Nor should the absence of pay be romanticised into proof of neutrality. Volunteer systems still need selection criteria, disclosure, recusal, replacement and accountability. Good faith and careful design are complements, not substitutes.

Authoritative records likewise help. Using AFRINIC’s records to verify membership standing and formal eligibility can make similar cases more consistent. But a registry record answers only the question the record is designed to answer. It can show status in the system. It cannot decide every judgment about suitability. Where a committee moves from verification to evaluative screening, the evidentiary standard and review route become more important.

The unresolved composition question

The most visible documentary tension concerns the composition of the four-person committee. AFRINIC’s Bylaws interpretation section describes NomCom as comprising at most three community volunteers with substantial interest in number-resource management and AFRINIC’s mission, plus one Board representative drawn from a seat not open in the relevant election. Article 9 separately refers to a Chair and three other qualified and experienced people appointed by the Board after a public call to the African Internet community.

The 11 February 2026 call advertised four vacant volunteer positions. The 1 April notice named four people without identifying a Board-representative place. At the evidence cutoff, none of the four names matched the directors shown on AFRINIC’s public Board page. Read together, these documents create a facial public-record inconsistency: the definition describes three community volunteers plus one Board representative, while the call described four volunteer vacancies and the appointment did not classify the four people.

That comparison does not prove a legal breach. The public Board page may not answer every relevant question about designation. The Bylaws provisions may require interpretation. A representative’s status or selection could have been documented elsewhere but not located in the reviewed set. The proper conclusion is narrower: the public record does not reconcile the advertised vacancies and announced appointments with the composition categories described in the Bylaws.

This could be resolved without publishing a biography or making any insinuation about a named appointee. AFRINIC could identify which member occupied which composition category, state how the category was satisfied and publish the relevant resolution or minute. Classification is process information. Keeping it private creates uncertainty without producing an obvious confidentiality benefit.

The same discipline applies to the later identification of Ganesh Ramalingum as Chair. The record establishes that the 8 May call used that office. It does not establish how the office was assigned. A short statement of the selection step would close the gap. It would not invite a personality contest.

Independence has two layers

The appointment can therefore be assessed on two planes. The first is formal independence: what do the written rules require? On that plane, the design is more substantial than a dismissive account would suggest. It includes a public call, a direct-candidacy ban, conflict disclosure, impartiality and equal-treatment principles, unpaid service, authoritative record checking and separation from the mechanics of voting.

The second is evidenced independence: what public material shows that the safeguards operated in the appointment and screening at issue? On that plane, the record is thin. It lacks the number of volunteer applicants, a published selection rubric, comparative reasons, a Board vote or resolution, composition labels, individual conflict or recusal reporting, the nomination denominator, exclusion totals and reasons, and a clearly specified nominee-review mechanism.

Formal independence without evidence can still exist. A person can act impartially even when the public cannot audit the basis for believing so. But an institution concerned with durable confidence should not make that private virtue carry the entire load. The point of procedural evidence is to reduce the need to guess at character.

This is where the agency problem becomes concrete. Delegated technical actors can exercise meaningful procedural power without owning the networks or bearing the operators’ downside. A short-lived committee may be conscientious, yet the design still creates agency risk if its appointment and screening choices are not sufficiently visible to the people who must live with the resulting slate. Recurring committee rotation can also become procedural exclusion when the machinery is controlled by a small, self-reproducing circle. No evidence in the reviewed public record permits any of the four appointees to be labelled an insider.

The institutional design must not depend on proving or disproving personal belonging.

The remedy is to make the power surface smaller and the record clearer. NomCom should verify defined candidate conditions, document any evaluative criteria before applying them, record conflicts and recusals, explain process-level outcomes and provide an independent route for a nominee to challenge a material procedural error. None of that gives the committee authority over number resources. None creates a continental mandate. It simply disciplines a corporate gatekeeping function.

The strongest defence

The strongest defence of AFRINIC’s arrangement begins with corporate responsibility. Article 9 assigns appointment to the Board. A Board responsible for the organisation must be able to constitute a committee on a workable timetable. A public volunteer call invites participation. Objective eligibility rules protect the ballot from deficient nominations. The candidacy ban, conflict duty, unpaid status and separation from ballot administration address obvious independence risks. A temporary committee needs enough confidentiality to discuss suitability candidly and protect volunteers and candidates from unnecessary exposure.

This defence has force. Publishing applicant biographies without consent could chill volunteer participation. Publishing every conflict detail could reveal personal or professional information beyond what is needed to assess a recusal. Publishing interview notes or rejection reasons wholesale could expose candidates to reputational harm, privacy risk or disputes over candid judgments. A short annual process can be burdened into paralysis if every intermediate exchange becomes a public hearing.

Transparency and confidentiality are not opposites, however. The choice is not between a secret file and a public dump. Process-level disclosure can preserve the substance of privacy. AFRINIC could publish the number of volunteers, the selection criteria, the composition category for each selected member, whether a recorded Board vote occurred, aggregate conflict and recusal counts, the number of nominations received by contest, the number found ineligible, broad non-identifying reason categories, and the existence of an independent review route. Sensitive files could remain private.

The older standing NomCom page itself shows that procedural detail can be separated from candidate merits. It describes one Board member and three community representatives, chair selection from among community representatives, simple-majority decision-making, confidentiality around the merits reported by the Board representative, and staff non-participation in merits. Because the page dates from 2021, it should not be substituted for the 2026 Guidelines. Its value is narrower: it demonstrates that AFRINIC has previously described process architecture without publishing private candidate evaluations.

What the final slate proves—and what it does not

The final slate publication proves that NomCom completed a screening output by 17 June. It proves the number of people found eligible in each contest: five, two and one. It proves that the four-person committee’s work affected who could proceed toward the seven open places. It does not reveal how competitive the intake was. Eight eligible names could have emerged from eight nominations or from a much larger field. The evidence does not say.

That missing denominator matters for assessing gatekeeping intensity. If almost everyone was found eligible, evaluative discretion may have had limited practical effect. If many were excluded, the screening criteria and review path become much more consequential. Because the denominator is absent, neither story can responsibly be asserted. The right response is not to select the narrative that suits one’s institutional preference. It is to publish the denominator.

The contest with one eligible person for two PDP Co-Chair positions illustrates the same boundary. The numbers show a shortage of eligible names relative to positions. They do not show why. The causes might involve the nomination intake, eligibility rules, supporter requirements, candidate willingness, timing or other factors. No unsealed fact permits a conclusion. A post-process report could state what happened without blaming a person.

The Guidelines’ residual dispute provision does not fully close this accountability question. It sends issues not expressly covered to collective consensus of members or the PDWG present at the meeting and says those issues do not interrupt the process. That may allow the meeting to continue when an unexpected question arises. It is not a detailed appeal right for an excluded nominee. A meeting-consensus mechanism and an individual procedural-review mechanism serve different purposes.

The wider legitimacy dispute

NRS has publicly disputed the authority and legitimacy of AFRINIC’s current Board in connection with the 2026 AGMM and urged member action. That is NRS’s institutional position. NRS advocates, researches, convenes and represents members who expressly authorise it. It does not operate AFRINIC’s registry, RPKI, WHOIS or RDAP; it does not run this NomCom, administer appeals or elections, hold settlement authority, or provide custody and continuity merely by speaking. Its claim cannot be converted into independent proof that the April appointment was unlawful.

AFRINIC’s own publications take the other institutional position: the Board acted under Article 9 and the committee proceeded under the 2026 Guidelines. Those documents establish what AFRINIC said and did. They do not, merely by being official, conclusively establish the wider legitimacy of the Board, the election or the RIR model. No final legal determination specific to this April appointment appears in the sealed record.

The article’s subject is not the general legitimacy of the 2025 or 2026 elections. It is not a replay of the receiver-era 2025 NomCom controversy. It is not an account of candidate biographies or vote totals. The wider Board-authority dispute matters here only because it concerns the actor making the appointment. It remains attributed, unresolved and outside the narrow thesis.

That disciplined boundary prevents two symmetrical errors. The first would treat AFRINIC’s invocation of Article 9 as conclusive proof of a mandate over networks and resources. The second would treat NRS’s challenge to the Board as conclusive proof that four named appointees acted improperly. Neither inference is supported. The institutional design can be criticised without converting uncertainty into accusation.

The operator’s stake

NomCom does not allocate IP addresses. It does not operate AFRINIC’s registry, routing infrastructure, RPKI, WHOIS or RDAP. It does not directly change an operator’s registration or existing resource rights. Its effect is indirect: it controls admission to offices that advise on corporate governance, represent the region in global number-resource coordination, or administer the policy-development process.

For operators and resource holders, that indirect effect still matters. A narrow or weak candidate slate can increase the cost of evaluating choices. Opaque criteria and invisible conflict handling can reduce confidence and increase participation or dispute expense. The people who eventually fill governance and policy offices can shape the predictability of institutional decision-making. But every effect is mediated through later officeholders and later processes. It would be wrong to write as though the four screeners directly controlled routes or address blocks.

The distinction is more than caution. It is the correct theory of institutional power. A nomination committee controls access to a corporate decision surface. It must be accountable for that access. The registry institution around it remains a coordinator and recordkeeper, not a government. Where later policy touches commercial conduct, legal rights or punishment, authority must come from competent law and the parties actually authorised to decide—not from the fact that a committee once screened the officeholder.

This is the danger of mandate laundering in miniature. A public call is described as community participation. Board appointment becomes constitution under Article 9. NomCom produces a slate. Members or a working group select from that slate. Officeholders later act. At each step, a valid procedural fact can be rhetorically enlarged until the completed chain is made to sound like authority over an entire region’s operators. The enlargement is false. The process supports only the bounded corporate and policy function it was created to perform.

A proportionate accountability record

The most useful reform is not to expose personal files. It is to establish a standard annual process record. Before appointment, the Board could publish the selection rubric and composition categories. After appointment, it could publish the number of applicants, the category assigned to each selected member, the date and form of the decision, and a non-personal statement of how conflicts were checked. NomCom could publish its chair-selection method, aggregate conflicts and recusals, the nomination denominator by contest, aggregate eligibility outcomes and broad reasons.

A separate reviewer could examine nominee challenges without disclosing confidential merits.

Such a record would answer the questions the 2026 publications leave open while preserving legitimate confidentiality. It would let members test whether the public call was meaningfully competitive. It would reconcile the four-volunteer advertisement with the Bylaws’ three-community-plus-one-Board definition. It would show whether the written conflict duty operated in practice. It would reveal how much screening discretion affected the slate without turning rejected candidates into public case files.

The counterfactual makes the trade-off visible. Imagine the same four people selected through a membership-elected or independently convened panel, using a published rubric, composition designations, a conflict register and a nominee-review mechanism. The committee could exercise substantially the same screening powers and keep sensitive records confidential. The Board-centred dependency would be weaker and the chain more auditable. But the alternative would add another election or selection process, cost time and money, risk politicising a short annual committee and still require conflict controls.

It is a design comparison, not proof that Article 9’s present allocation is void.

The more immediate improvement is therefore evidentiary. Even without changing who appoints, AFRINIC can reveal more about how appointment and recusal worked. If the record shows a fair process, disclosure strengthens confidence. If it exposes a correctable mismatch, early correction is cheaper than a later legitimacy dispute. In either case, the system becomes less dependent on institutional assertion.

A gate that must remain a gate

The 2026 NomCom was publicly presented as an Article 9 gatekeeper. Its four members were appointed, the committee issued a candidate call and it published a final eligible slate. The written design contains genuine safeguards and a meaningful separation between screening and ballot administration. Those facts should not be erased in pursuit of a simple indictment.

The public evidence is nonetheless insufficient to audit how the Board selected the screeners, how the four positions matched the Bylaws’ composition categories, how conflicts and recusals operated, or how much screening reduced the candidate field. Those are accountability gaps. They are not proof of illegality, bad faith or misconduct by Prichard Chakadenga, Solomon Dindi, Ganesh Ramalingum or Ikibah Ebi Benjamin.

AFRINIC should remain what the Internet needs from a registry: a thin coordinator of uniqueness and a keeper of accurate records. Corporate governance can make that service accountable. It cannot transform the service into sovereignty. A NomCom may keep an election gate under defined rules. It may not turn that gate into a throne, and no later officeholder should be able to use a screened corporate mandate as authority to punish networks or seize control of assets.

The decisive question is therefore not whether readers trust four names. It is whether the institution publishes enough process evidence that personal trust becomes unnecessary. The answer in the 2026 record is incomplete. That incompleteness can be repaired with narrow disclosures, independent review and a firm boundary between corporate procedure and power over the running Internet.