Summary

  • Resolution 201511.254 had set the May/June 2016 Annual General Members’ Meeting as the point for finalising the first Governance Committee election and appointment. Resolution 201602.270, passed unanimously by email on 28 March 2016 at NomCom’s request, moved that work to the AFRINIC meeting in November/December—an extension of roughly five to six months.
  • The committee envisaged in the then-current terms was advisory and its advice was non-binding. It nevertheless mattered because formal advice was meant to be considered and a Board decision not to follow it was expected to produce published reasons. Until the committee existed, neither that channel nor that explanation mechanism was available.
  • The delay left the Board, management and membership organs carrying their pre-existing corporate functions through the extended interval. It did not grant AFRINIC sovereignty or governmental jurisdiction, and the record does not establish bad faith, illegality, a suppressed complaint or a manipulated election.
  • Calendar-year terms offer a credible benign reason for moving initial selection closer to year-end. The public record, however, does not reproduce NomCom’s request or explain the operational problem, rejected alternatives, interim arrangement or milestones. Good postponement design would make those elements visible without pretending that a private advisory committee was a public regulator.

A small resolution with a measurable interval

The institutional act is both narrower and more consequential than the language of political controversy would suggest. In November 2015, Resolution 201511.254 approved version 2 of terms for a Governance Committee and set the election and appointment of its initial members to be finalised at the May/June 2016 Annual General Members’ Meeting. On 28 March 2016, before that meeting window arrived, the Board passed Resolution 201602.270 by email. The resolution attributed the request for delay to NomCom and shifted the first election and appointment to the AFRINIC meeting in November/December 2016.

The two meeting windows, rather than two precise polling dates, define the change. The defensible measure of the extension is therefore roughly five to six months.

The Board minutes dated 13 April add three particulars: the resolution had passed by email on 28 March, it was unanimous, and it was moved by KS and seconded by AB. Those initials should remain initials; the record supplied here does not justify expanding them. The same minutes place the item beneath a heading concerning resolutions to be ratified, while their operative description says it had already passed by email. That presentation can be reported as a documentary tension, but it cannot safely be turned into a second adoption, a hidden ballot or a separate legal event. The fixed point is the recorded email passage on 28 March.

Within that compact chronology lies the governing limit. AFRINIC is a private, membership-based technical bookkeeper and coordinator. It maintains registry records and operates processes on which network operators and other organisations rely. That reliance can make its choices practically important. It does not make AFRINIC a state, a legislature, a court or a regulator. The March act was an internal corporate scheduling decision. It neither transferred sovereignty nor expanded public power, and the committee whose birth it postponed was not designed as a governmental authority.

That distinction is not a reason to dismiss the resolution. It is the reason to examine it accurately. Private institutions can allocate information, attention and timing in ways that affect confidence and operational planning without wielding police, prosecutorial, punitive, confiscatory or adjudicative power. The relevant question is not who ruled a territory. It is who could ask organised questions, deliver formal advice, require a publicly visible explanation under the committee’s design, and decide when that added process would begin.

What the proposed committee would have added

Version 2 of the Governance Committee’s terms described an advisory body. Its advice would not bind the Board. That negative proposition matters: the committee was not given a veto, appellate jurisdiction or power to substitute its judgment for an existing corporate organ. A recommendation could be declined. The Board would remain the Board, management would remain responsible for management, and members would retain the functions available to them within the organisation.

But non-binding does not mean institutionally empty. The terms contemplated that formal advice would be considered and that, when the Board did not follow it, reasons would be published. This creates a two-part interface. First comes structured input: an identified forum can frame a question, gather views and state advice in a form that reaches the decision-maker. Second comes reason-giving: departure from formal advice becomes an occasion for explanation in public. Neither stage compels an outcome. Together, however, they can improve the information around a decision and make the decision-maker’s reasoning easier to inspect.

The difference can be seen without imagining any particular controversy. Before the committee exists, a governance concern may reach the Board through ordinary correspondence, meetings, member interventions, staff channels or other routes that happen to be available. The record does not show that such routes disappeared. What is missing is the specific, pre-announced path that the new terms were designed to supply. There is no constituted body able to turn a concern into formal committee advice, and consequently no committee advice from which a Board departure could trigger the contemplated publication of reasons.

Once that design has been approved and scheduled, its start date becomes part of the institution’s accountability architecture. Moving the date does not erase existing mechanisms, and it does not prove that a particular question went unanswered. It delays a marginal addition: another forum, another route for information, and another reason-giving expectation. The impact is real precisely because it is bounded. The organisation continues to operate, but for the extension it operates without the extra interface it had said it would create.

This is the temporary accountability gap at the centre of Resolution 201602.270. It is not a claim that advice was suppressed. A body that did not yet exist could not have issued the relevant formal advice, and the available record identifies no suppressed recommendation or foregone outcome. Nor is it a claim that the committee would have changed any Board decision. Counterfactual influence cannot be established. The narrower inference is enough: during the added five-to-six-month interval, the contemplated mechanism for advice and public reasons was unavailable because its institutional carrier remained unconstituted.

Timetable control and incumbent position

Timetables are often treated as neutral administration. Sometimes they are. Yet control of a start date also controls when a new participant enters an existing sequence of decisions. Resolution 201602.270 changed the point at which the Governance Committee could begin to occupy its proposed place in AFRINIC’s internal arrangements. Until then, the existing Board, management and membership organs continued through their familiar corporate paths.

This continuation should be described with care. The resolution did not confer new powers on those organs. It did not remove a proven legal right from another actor. It did not create a new jurisdiction, authorise punishment or decide a dispute. Existing decision-makers simply remained in their pre-existing positions while an additional advisory forum stayed absent. Their practical advantage was one of timing and institutional configuration: the current routes continued, while the planned route that might have supplied structured questions and a public explanation had not begun.

“Incumbent” in this setting therefore means the people and organs already performing corporate functions, not office-holders exercising sovereign rule. Their relevant capacity was practical agenda-setting, timetable control and pre-existing corporate decision authority. Postponing the committee preserved that configuration for longer. It did not demonstrate that anyone sought preservation for improper reasons. The minutes record unanimity, but unanimity proves agreement on the recorded resolution, not a common unrecorded motive.

Equally, the attribution to NomCom proves that the resolution said NomCom requested the delay; it does not disclose what NomCom wrote, who framed the request, what options it considered or why every Board member agreed.

That separation between effect and motive is essential. An institutional effect may be observable even when subjective reasons are not. If a meeting is moved, the interval changes whether or not the motives are known. If a committee remains unconstituted, its procedures cannot operate whether or not anyone intended to avoid them. A rigorous account can therefore say that the delay preserved existing decision paths and postponed a new accountability interface. It cannot leap from that effect to an allegation of entrenchment, fear of a candidate, suppression of dissent or election manipulation.

The distinction also improves evaluation. Motive-focused analysis invites speculation about personalities that the record cannot resolve. Process-focused analysis asks answerable questions. How long was the extension? What function had been promised but was not yet available? What reason was stated? What milestones were made public? Was there a temporary channel? What would automatically end the postponement? Those questions identify the institutional quality of the decision without pretending to know private emails or private intentions.

The strongest benign account

There is a coherent administrative case for the change. Version 2 of the terms stated that ordinary committee terms would begin on 1 January and end on 31 December. An initial selection finalised in May or June would therefore sit awkwardly against a calendar-year design. Selection in November or December could make it easier to prepare members for a January start, reduce the ambiguity of partial terms and align appointment records with a clean annual cycle. More preparation time could also give NomCom room to organise work that had not been completed before the earlier meeting window.

A decision by email in late March could settle the calendar before the May/June meeting, rather than let participants proceed on an obsolete assumption. Unanimity may indicate that the Board regarded the scheduling solution as straightforward. It is also possible, in general institutional terms, that transitional arrangements would have created more confusion than a single move towards year-end. These considerations do not require suspicion to explain why reasonable administrators might prefer the later meeting.

That is the strongest benign case, but it must remain an interpretation rather than an attributed rationale. The published resolution says that NomCom requested the delay. It does not reproduce the request or state that calendar alignment, onboarding, workload or preparation time was the reason. The ordinary 1 January term clause makes those possibilities coherent; it does not prove them. The record also does not establish that the Board weighed early constitution with a short initial term against later constitution with a standard term.

The unanswered question is consequently not whether an administratively respectable explanation can be invented. One can. The question is whether the public record connected a concrete problem to the scale and design of the remedy. It did not publish the underlying request, a detailed operational reason, an alternative timetable, a set of delivery milestones, a member response or an interim substitute for the committee. These omissions narrow what readers can evaluate. They do not prove that internal discussion, consultation or mitigation never happened; they show only that the identified public instruments do not disclose them.

Calendar alignment is especially useful as a test because it points towards less disruptive alternatives. If the difficulty was that normal terms began on 1 January, the institution might have considered a transitional first term, a staged appointment, an interim advisory group or an early constitution followed by formal term alignment. None can be asserted as suitable without knowing operational constraints, and none is shown to have been considered.

They are counterfactual design options that expose the missing part of the explanation: why did the selected five-to-six-month delay better serve the stated architecture than a shorter or transitional solution?

A published answer would not have needed to be long. A bounded explanation could have said what scheduling conflict existed, why the May/June target was no longer workable, which milestone would replace it, who owned delivery and what temporary channel would cover the gap. Such an account would not transform administrative preference into a legal obligation. It would make the relationship between cause, duration and safeguard legible.

Why the missing explanation matters

Reason-giving is sometimes confused with surrendering discretion. In fact, a reason can preserve discretion while making its use intelligible. The Board remained entitled, within its private corporate structure, to make the scheduling decision recorded in the resolution. Explaining the operational constraint would not have handed that decision to the future committee or to members at large. It would have allowed readers to assess whether the chosen interval was proportionate to the problem.

The subject is unusually apt because the delayed body itself was designed partly around reasons. Its advice was non-binding, but a Board departure from formal advice was expected to be accompanied by published reasons. The act of postponing that body therefore postponed not only an advisory voice but a particular pattern of explanation. That does not mean the resolution violated the future mechanism; the mechanism was not yet functioning. It means the manner of postponement should be assessed against the accountability value the institution had already recognised in approving the terms.

An adequate public account might have answered four distinct questions. First, what obstacle made the original May/June window impracticable? Second, why did that obstacle require an extension to November/December rather than a shorter adjustment? Third, what visible steps would show that the later date remained attainable? Fourth, how could a governance question be raised in a structured way while the new committee was absent? The resolution answered the broad timing question and attributed the request. It did not answer the other three in detail.

The absence of those answers increases dependence on trust in the existing decision path. Members and operators can see that the timetable moved, but not how the length of the move was derived or what would happen if preparation slipped again. They can see that NomCom requested the change, but not the problem it was trying to solve. They can see a later target window, but not intermediate checks. The result is not lawlessness or operational breakdown. It is a thinner basis for evaluating whether the planned accountability mechanism is genuinely approaching.

This is why a temporary gap should be measured by more than elapsed days. Duration matters, but so do information and reversibility. A five-month extension accompanied by a stated constraint, monthly milestones, an interim questions channel and a firm long-stop date is institutionally different from the same five months accompanied only by a new end window. Both postpone the committee. Only the first tells affected readers how the postponement is controlled.

A registry’s influence without sovereignty

AFRINIC’s narrow role sharpens rather than weakens the need for precision. A regional Internet registry keeps records and coordinates administrative processes associated with number resources. Operators may organise plans, contractual expectations and technical work around the reliability of those records and processes. Consequently, governance uncertainty at the registry can affect confidence, predictability and the willingness to make plans. It can impose attention costs even when no packet moves differently as a direct result of the resolution.

Practical dependence, however, is not governmental jurisdiction. AFRINIC cannot acquire sovereignty because others rely on its database. It has no police or prosecutorial power over networks or Internet users. It cannot turn an internal timetable resolution into a public punishment, confiscation or adjudication. The March delay did not itself reroute traffic, revoke resources, regulate network conduct or determine legal rights. Claims of that kind would confuse operational reliance with public authority and would exaggerate the act under review.

The better operator question is about the quality of coordination. Can an organisation on which others rely describe when a new process will start? Can it show whether preparations are on track? Can it explain a delay in terms that let an operator distinguish ordinary administration from an open-ended institutional risk? Those are practical questions. A network planner does not need to believe AFRINIC is a government to care whether its internal procedures are predictable.

The same point explains why the Governance Committee’s non-binding character is no answer to the accountability concern. Information can affect a decision without commanding it. A forum can improve institutional memory, expose assumptions and make a decision-maker articulate reasons without exercising superior power. For operators, those functions can make the environment easier to understand. They supply signals about whether the registry identifies problems, listens through known channels and follows a calendar it has announced.

There is also an important limit on causation. The closed record does not identify an outage, route failure, revoked resource or delayed deployment caused by Resolution 201602.270. General operational relevance cannot substitute for event-specific proof. The proper claim is that reliable registry governance supports predictability because operators rely on records and processes, and that a poorly explained postponement can reduce visibility into those processes. It is not that this delay broke infrastructure.

An accountability gap, not an institutional vacuum

The committee’s absence did not leave AFRINIC without organisation. The Board continued, management continued and members retained their existing place in the private membership structure. The source material does not establish that ordinary correspondence, meeting participation or other channels ceased. Calling the interval an institutional vacuum would therefore be inaccurate.

“Gap” has a more exact meaning. A particular mechanism had been approved for future operation; its initial constitution was scheduled; that schedule was moved; and during the extension the mechanism was not available. The gap existed at the margin between the old configuration and the promised addition. Its practical content was the absence of structured committee advice and the associated expectation of published reasons when the Board declined formal advice.

That framing avoids two opposite errors. The first is melodrama: presenting a private scheduling decision as a seizure of state power or the destruction of lawful government. The second is trivialisation: assuming that because the committee was advisory, postponing it changed nothing. The evidence supports neither. The resolution was modest in legal character and meaningful in procedural effect.

It also prevents retrospective reasoning from contaminating the March event. Later developments cannot be used to pretend the committee was already available during the extension. Nor can its later composition or conduct be projected backwards into version 2 of the terms. The question is what the approved design was meant to add at the time and what remained unavailable when its first election and appointment were deferred.

In that sense, the resolution is a useful study in institutional latency: the period between approving an accountability design and making it operational. Approval creates expectation, but only constitution creates capacity. During latency, the organisation may speak as though a reform exists because its terms have been accepted, while those affected cannot yet use it. A timetable decision determines how long that discrepancy persists.