Summary

  • Resolution 201608.298 created a standing invitation for AFRINIC’s Legal Adviser to attend every Board meeting; it did not prove that the adviser attended that meeting, attended later meetings, voted, held a veto or acquired any decision-making power.
  • The strongest case for the arrangement is practical: recurring access can help directors identify the right corporate instrument, authority limit and procedural risk before a choice hardens into action.
  • The governance risk lies less in the presence of counsel than in an incomplete record. Protected legal substance can remain protected while the Board records non-sensitive facts about the matter, decision-maker, governing instrument, conflicts, reliance and resulting act.
  • Neither unanimity nor legal participation could enlarge AFRINIC’s powers. It remained a Mauritius private company performing a narrow technical bookkeeping and coordination function, not a regulator, police force, punishment authority, confiscator or court.
  • The proportionate answer is a bounded legal-advice protocol that distinguishes invitation, attendance, participation and reliance, and that separates ordinary minutes from any protected record rather than treating every lawyer-related exchange as automatically secret.

A consequential sequence in a two-hour call

The recorded Board meeting began over Webex at 15:12 UTC on 10 August 2016 and adjourned at 17:17 UTC. During the review of the agenda, chief executive Alan Barrett proposed adding a discussion under Any Other Business about the Legal Adviser’s attendance at Board meetings. That proposal establishes how the subject entered the recorded agenda. It does not establish Barrett’s motive, reveal who drafted the later resolution or tell us whether a recurring-attendance arrangement had been debated elsewhere.

The placement of the discussion matters. Immediately before the attendance item, the Board took note of advice from the Legal Adviser concerning a possible special general members’ meeting. The minutes summarize the recommendation: if members were to vote to constitute the Governance Committee, an SGMM should be called at AFRINIC-25. The merits of that recommendation belong to a different question. For present purposes, the important fact is structural. The record first shows advice attached to a particular matter; it then shows the Board considering whether the adviser should be available across Board meetings as a recurring feature.

Under item 7.3, the Board discussed and agreed that the Legal Adviser should be invited to attend Board meetings to provide advice. Sunday Folayan proposed Resolution 201608.298. Haitham El Nakhal seconded it. The resolution passed unanimously, with Alan Barrett and Krishna Seeburn recorded as not present for that vote. The archived Board register describes the invitation as applying to all Board meetings; the minutes use every Board meeting. Those formulations carry the same practical breadth: the resolution was universal in invitation, not confined to a named dispute, a single corporate instrument or one anticipated agenda item.

That is the institutional act. Its precision is important because almost every larger conclusion must be resisted. The vote did not record that the adviser was present on 10 August. It did not guarantee later attendance. It did not say whether the invitation covered an entire meeting or only selected items. It did not confer a vote, veto, ruling power or executive office. It did not identify the adviser, define the client, reproduce a retainer, state a fee, establish a conflict protocol or prescribe how advice would appear in the minutes. It did not declare any communication privileged. It did not report a later benefit or harm.

The resolution nevertheless matters. A standing invitation changes the architecture through which a Board encounters legal advice. A lawyer summoned for one defined question enters a decision at a visible point. A lawyer who may be invited to every meeting can become part of the recurring environment in which issues are framed, instruments are chosen and risks are identified. That can make a private company more disciplined.

It can also make it harder for members to understand the basis of decisions if the existence of legal advice becomes a reason to record less rather than a reason to distinguish carefully between protected substance and accountable corporate facts.

This tension has practical stakes because AFRINIC’s records and coordination services matter to organisations that rely on Internet number resources. Dependence on a registry service can make corporate error expensive. Delayed decisions, uncertain records or mistaken use of corporate powers can travel outward into planning, financing and network operations. Yet practical dependence does not transform the registry into government. AFRINIC was and remains, for this analysis, a Mauritius private company carrying out a useful but bounded technical bookkeeping and coordination function.

A Board resolution and a lawyer’s participation cannot manufacture sovereignty. They cannot turn corporate preference into legislation, a registry entry into an adjudication of ownership or operational leverage into a public power to punish.

The resolution changed access, not authority

The cleanest way to understand Resolution 201608.298 is to separate four concepts that minutes often blur: invitation, attendance, participation and reliance.

Invitation is the only one that the resolution directly establishes. The Board decided that the Legal Adviser should be invited to every meeting. That is an instruction about access and expected availability. It sets a recurring default for the meeting organiser and gives the adviser a pathway into the Board’s deliberative setting.

Attendance is a separate historical fact. An invitee may accept or decline; a person may join for one item, remain for a whole call or be absent despite a standing invitation. The 10 August minutes list nine people as present at some point—the CEO and eight directors—but do not name the Legal Adviser in that Present list. That omission supports only a narrow statement: the adviser was not named there. The minutes’ summary of advice on the preceding item does not resolve the question. Advice could have been delivered before the meeting, in writing, through an officer or during the call without the adviser appearing in the named list.

The sealed record does not choose among those possibilities.

Participation concerns what an attendee did. A lawyer might listen silently, answer a question, identify an instrument, suggest a process, flag a risk or withdraw from a matter because of a conflict. None of those modes is specified by the resolution beyond the general purpose of providing advice. Participation also does not imply decision power. The Board remained the corporate decision-maker. Counsel could inform the directors’ understanding but could not cast a director’s vote, substitute a legal opinion for the required corporate act or convert advice into a binding judgment.

Reliance asks whether the authorised decision-maker used the advice in reaching a choice. A Board can receive advice and reject it, follow it, adopt only part of it or postpone action. A minute that says advice was received is not necessarily a minute that says it was relied upon. Conversely, recording reliance does not require publishing the adviser’s analysis. A useful record might state that the Board considered legal advice on the applicable company instrument and relied on it in choosing a specified procedure, while reserving the legal merits.

That kind of notation preserves the chain from question to authority to act without revealing the protected reasoning itself.

These distinctions expose why the resolution was more consequential than its few words suggest. By making invitation recurring, the Board made it likely that questions about attendance, participation and reliance would also recur. A one-off arrangement can be reconstructed from a single agenda. A standing interface requires a repeatable record discipline. Otherwise a universal invitation becomes a universal ambiguity: members can see that counsel had access in principle but cannot tell when advice entered a decision, what corporate authority it concerned or whether the directors acted on it.

The distinction also protects counsel from exaggerated claims. Presence at a meeting does not make the adviser the author of every decision taken there. Advice does not imply control. A recorded legal concern does not imply that the Board agreed. If minutes do not identify which item involved legal participation, later readers may attribute too much influence to counsel or too little responsibility to directors. Better metadata protects both sides of the relationship by locating decision responsibility where it belongs.

The strongest case for a standing chair

There is a straightforward, serious case for inviting legal counsel to every Board meeting. Boards do not encounter law only when an agenda uses the word legal. Questions of notice, quorum, delegation, conflicts, contracting authority, member rights, record correction and the choice between Board action and member action can surface unexpectedly. If counsel must be arranged after the problem appears, directors may proceed without timely advice or defer a matter that could have been handled properly during the meeting.

Recurring access can also preserve context. An adviser who follows successive meetings may understand which corporate instrument is under discussion, what earlier decision created the present question and which factual assumptions the directors are making. Context can reduce the cost of repeatedly reconstructing a matter. It can help counsel intervene early, when changing the process is still cheaper than repairing a defective act. A precise warning before a vote may avoid the need for corrected minutes, repeated notices, a reconvened meeting or a dispute over whether the right decision-maker acted.

The arrangement may be especially valuable in a technically consequential organisation. Registry operations sit within contracts, corporate rules and operational dependencies. A seemingly narrow Board choice can affect how members plan, how staff implement a decision and how resource holders assess continuity. Timely legal advice can help directors distinguish what the company may do from what would require member action, contractual agreement or no action at all. It can encourage written reasons, clearer resolutions and more accurate descriptions of the instruments on which a decision depends.

A standing invitation can be administratively efficient as well. The Board need not renegotiate access each time an issue arises. Meeting organisers know that advice may be available. Directors can ask a question while the facts are fresh. The adviser can hear how a proposal is framed rather than receiving a compressed account after the event. This lowers the friction of seeking help and can make legal review preventive rather than merely defensive.

Confidentiality may also be legitimate. Candid advice can include assessments of vulnerabilities, negotiating positions, legal exposure or possible responses that should not be broadcast. Directors should be able to test assumptions and hear uncomfortable answers without every sentence becoming immediate public copy. Concise minutes are not inherently evasive. Full transcripts can make meetings less candid, obscure the actual decisions beneath conversation and expose sensitive interests without improving accountability.

Nor does the brevity of the 2016 record prove that fuller private controls were absent. An appointment instrument, retainer, conflict check, attendance protocol or more detailed record may have existed outside the archived register and minutes. The official material establishes what it contains; silence cannot be inverted into evidence that no other document existed. A fair analysis therefore treats prudence and efficiency as genuine possible reasons for the resolution, not as cover for a presumed misconduct that the record does not show.

This strongest case should be accepted before asking what is missing. The point is not that legal advice should have been kept outside the room, nor that all legal thinking should have been published. The point is that recurring advice needs recurring boundaries. Good access and good accountability are complements. A Board can preserve candid merits while identifying the authority chain around them. Indeed, a standing arrangement makes such identification more important because counsel’s availability becomes ordinary rather than exceptional.

A private company cannot borrow public power from counsel

The first boundary is substantive, not documentary. AFRINIC’s technical importance does not make it a sovereign institution. It maintains and coordinates number-resource records within a private organisational and contractual setting. It is not a state, legislature or regulator. It is not a police force or prosecutor. It possesses no public power to punish, confiscate or adjudicate merely because its database and services are operationally important.

Legal advice can clarify a company’s lawful options. Counsel can point directors toward the constitution, bylaws, contracts, resolutions or procedures relevant to a matter. Advice can show that a proposed act belongs to members rather than directors, that notice is required, that a conflict should be managed or that a resolution needs different wording. It can identify risk and explain uncertainty. These are valuable corporate functions.

But advice cannot enlarge the source of power. If a Board lacks authority to perform an act, a lawyer’s presence cannot supply it. If a registry record does not create ownership, a legal explanation cannot turn the record into a confiscatory title system. If a private organisation cannot declare guilt or impose public punishment, counsel cannot transform a corporate allegation into a judicial finding. A standing invitation is an interface for advice, not a jurisdictional grant.

This boundary matters because operational dependence can distort language. An organisation that can affect access to an important service may begin to describe its preferences in the vocabulary of mandate, enforcement or stewardship. Unanimity can intensify that impression. Yet Resolution 201608.298 was unanimous only among the directors participating in that private-company vote. It was not legislation, a public referendum or proof of a continental principal.

The absence of Alan Barrett and Krishna Seeburn for the vote further identifies the decision as a particular corporate act at a particular moment; it does not undermine the vote’s recorded unanimity, but it prevents the word unanimous from floating free of who actually participated.

The same discipline applies to the word community. Members and resource holders have concrete interests in AFRINIC’s governance. They pay fees, hold corporate rights in defined ways and rely on the service. Their interests deserve accurate records and accountable decisions. They do not merge into a sovereign public merely because an institution invokes a broad community. A lawyer advising the Board owes duties defined by the relevant private legal relationship, not an invented office as judge for an entire region.

Properly bounded, counsel can reinforce this modest understanding of power. The adviser can ask: Which company instrument authorises the Board? Is the decision reserved to members? What contractual right is actually engaged? What process is required? What remedy exists if the record is wrong? Those questions make legal advice a restraint on institutional overreach. Poorly recorded, however, the same advice can become an opaque invocation—“legal advised”—that closes discussion without showing the source of corporate authority. The difference lies not in publishing every legal conclusion but in recording the decision architecture around it.

What the minutes show, and what they leave open

The archived 2016 Board register and the minutes perform different but complementary evidentiary roles. The register preserves the resolution’s identifier and summarizes the universal invitation. The minutes place that act within the meeting: a Webex session, an agenda amendment, the preceding summary of advice, a discussion under item 7.3, the proposer, the seconder, the vote and the two people not present for it. Together they establish the decision with unusual clarity on some dimensions.

They are much less informative on the design of the relationship. The adviser is unnamed. The client is not defined. That matters because “the organisation,” “the Board,” individual directors and an officer are not interchangeable clients. Without a client definition, a later reader cannot tell whose interests the advice was meant to serve or who was authorised to instruct the adviser.

The appointment authority is also not stated in this closed record. We do not know from these documents who selected the adviser, under what instrument, for what term or with what reporting line. The resolution itself is an attendance arrangement, not a complete appointment instrument. It says nothing about whether the adviser was an employee, outside counsel, a contractor or someone serving under another form of engagement. It says nothing about fees.

Scope is similarly open. “To provide advice” may sound self-explanatory, but a recurring Board setting generates practical questions. Was advice available on every agenda item or only when requested? Could the adviser raise issues without a request? Did attendance continue during executive sessions? Who decided when counsel should leave? Could directors seek separate advice? How were disagreements between legal and operational judgments handled? Resolution 201608.298 answers none of these questions in the public record.

Conflict handling is not visible either. This is not evidence that a conflict existed or that checks were absent. It is a description of the record boundary. A standing adviser may encounter matters involving the company, Board members, officers, members or previous advice. A protocol can specify how possible conflicts are identified, who decides on recusal and when separate counsel may be needed. Without published non-sensitive metadata, members cannot distinguish a managed conflict from a matter in which conflict was irrelevant or a matter for which no public account was given.

The minutes do not state a rule for privilege, redaction or waiver. No conclusion about Mauritian privilege follows from that silence. More fundamentally, not every exchange with a lawyer is automatically privileged. Administrative scheduling, a bare statement of who attended, the identity of the decision-maker, the corporate instrument invoked and the fact that a resolution passed are different kinds of information from confidential legal analysis. Treating all counsel-adjacent information as one indivisible secret would sacrifice accountability without necessarily protecting legal substance more effectively.

The record also does not tell us what happened later. It contains no attendance ledger for subsequent meetings, no series of advice requests, no evidence of reliance, no cost account and no evaluation of decision quality. It does not prove that the standing arrangement worked or failed. It cannot support claims of improper influence, hidden control, inflated expenditure, operational harm or misconduct. Those would require evidence outside this subject.

This is why the correct question is architectural rather than accusatory. Given what the Board resolved, what record would allow members to understand corporate authority without demanding disclosure of genuinely sensitive advice? The answer begins by refusing two extremes: radical publication of legal merits and blanket opacity around everything counsel touched.

Privilege is not a black curtain

Legal confidentiality is often discussed as if a meeting must be either open in substance or entirely shielded. That binary is poorly suited to Board records. A single agenda item can contain several layers of information with different sensitivity.

At the outer layer are ordinary corporate facts: the item considered, who convened the meeting, who was present, who was absent, who proposed and seconded a resolution, who recused, which decision-maker had authority and what action followed. These are the coordinates of corporate action. Recording them ordinarily reveals no legal theory.

A second layer concerns advice metadata: whether legal advice was requested or received, the general corporate matter to which it related, the adviser’s role, the client, whether a conflict check was required, whether the Board relied on advice and whether separate counsel participated. These facts can often be expressed at a useful level without exposing the analysis. They tell members how the Board governed itself.

A third layer contains the legal substance: the adviser’s reasoning, assessment of exposure, discussion of legal weaknesses, negotiation positions, possible claims or strategic alternatives. Some of this may warrant protection. The exact boundary depends on the applicable relationship and law, which Resolution 201608.298 does not establish. The governance point is narrower: protecting this layer does not require erasing the first two.

A fourth layer is the corporate act itself. A Board cannot make its decision unknowable merely because advice informed it. The resolution, instrument used, authorised decision-maker and implementation step remain part of the company’s accountable conduct. If the act later changes, is corrected or is reviewed, the record should allow readers to follow that change without reconstructing the legal opinion.

Separating these layers prevents privilege from becoming a rhetorical shortcut. A minute can say that the Board considered advice about its authority under a named company instrument and resolved to take a specified action, while omitting the protected analysis. It can record that a director recused because of a disclosed category of conflict without publishing private detail. It can note that counsel joined for one item and left before the vote.

None of these examples claims that the underlying communication was legally privileged; they show how a Board can preserve confidentiality where justified and still produce an intelligible corporate record.

The separation also improves later correction. If a decision proves defective, reviewers need to know which body acted, what instrument it believed empowered the act, whether advice was relied on and what record must be amended. They do not necessarily need the adviser’s full reasoning. Authority metadata provides an audit trail that remains useful even when legal substance stays protected.

The 10 August minutes themselves contain three separate examples of record handling elsewhere in the meeting: an executive session, a redacted name associated with a memorandum of understanding and a decision to replace previously published May minutes after corrections were approved. These acts should not be attributed to Resolution 201608.298. They concern distinct items and may have independent reasons. They do, however, demonstrate that the meeting record could distinguish session type, redact a specific detail and correct an earlier public record.

In other words, the available choices were not limited to publishing everything or publishing nothing.

That comparator is valuable because it grounds a proportionate approach. Sensitive material can be placed in an executive-session record. A precise detail can be redacted without erasing the entire item. A mistaken public minute can be corrected and replaced. None of those techniques answers who may authorise protection or waiver, but each illustrates a tool from which a coherent protocol could be built.

The risk lies in untracked reliance

Standing access moves legal analysis earlier in the decision cycle. That is its chief benefit and also the source of its accountability risk.

When advice arrives early, it may shape the problem before directors see a formal proposal. Counsel may identify which instrument governs, suggest that an issue be framed as a contract question rather than a policy question or warn that a decision belongs to a different corporate body. This can prevent error. It can also make the advice less visible because its influence appears in the wording of the agenda, the chosen procedure or the absence of an option that was discarded before minutes were drafted.

Influence is not itself improper. Boards are supposed to use competent advice. The problem arises when the corporate record preserves only the conclusion. If members see that the Board “agreed” but not which instrument empowered it, they cannot assess whether the right body acted. If a minute invokes legal advice but does not state whether the Board relied on it, readers may assume either that counsel controlled the decision or that the reference was decorative. Both interpretations can be wrong.

Untracked reliance also weakens institutional learning. Suppose a later Board encounters the same kind of corporate question. It needs to know whether the earlier action rested on a particular instrument, whether the advice was limited to a factual assumption and whether a later correction changed the position. A complete legal opinion may be unnecessary and unavailable. A modest reliance note could still prevent the next Board from treating a context-specific choice as a general rule.

The risk compounds when attendance is universal in principle. Counsel may hear matters before their legal significance is apparent. Directors may begin to treat informal remarks as definitive advice. Staff may implement an action without knowing whether the Board relied on counsel, merely consulted counsel or proceeded despite a caution. A protocol can require that material reliance be confirmed in a form suitable for the record, while leaving spontaneous discussion outside that category until the Board chooses to act.

This is also where responsibility can become blurred. Directors, not advisers, hold the Board vote. If a decision causes delay or requires correction, saying that lawyers were involved does not answer who authorised the act. Conversely, if directors reject advice, counsel should not be treated as having endorsed the result merely because the adviser attended. Minute metadata can protect the integrity of both roles: advisers advise; authorised corporate bodies decide.

The distinction has economic consequences. Preventable corporate-process errors consume staff time and professional fees. They can delay member decisions, unsettle contracts and force repeated meetings. An unclear authority record raises the cost of review because each later reader must reconstruct the chain from scattered documents. Recurring legal access may lower those costs if it produces better instruments. It may raise them if the arrangement creates dependency without defined scope, review or fee control. The 2016 record proves neither outcome, but it identifies the control surface on which either could occur.

Why registry governance reaches infrastructure without becoming regulation

AFRINIC’s function is narrow in legal character but significant in practical effect. Organisations use registry records and coordination services when managing number resources and network plans. Confidence in those records supports continuity. Governance failures can therefore produce costs beyond the Boardroom even when they remain private-company failures.

A defective notice or uncertain resolution can delay a corporate decision. An ambiguous delegation can leave staff unsure who may act. Inconsistent records can complicate a member’s assessment of its rights. A dispute over authority can consume fees and management attention. These pathways do not require a sovereign power to be consequential. They arise from contracts, organisational dependence, technical coordination and the cost of uncertainty.

That combination makes accurate language essential. Calling AFRINIC a regulator would exaggerate its legal position. Calling it merely irrelevant bookkeeping would understate the operational reliance attached to the function. It is a private technical bookkeeper and coordinator whose services can matter greatly. Governance should therefore be judged with seriousness appropriate to its impact and restraint appropriate to its authority.

Legal advice can help maintain that balance. A careful adviser can distinguish a recordkeeping function from ownership, a contractual remedy from punishment and Board discretion from legislation. Counsel can remind directors that a technically powerful action still needs a valid private-law basis. This is perhaps the best version of a standing meeting interface: law acts as a boundary around operational temptation.

But the same practical importance increases the need for members to audit the corporate chain. If a decision affects how the service is administered, members should be able to identify the responsible body, governing instrument and resulting action. They need not inspect candid legal analysis to ask whether a director had a conflict, whether the Board or membership held the relevant power or whether a correction was made. These are accountability questions, not demands to turn legal advice into public performance.

Fee control belongs in the same category. The record before us gives no evidence of what the adviser cost or whether a fee was charged for meeting attendance. It would be wrong to infer waste. Yet a standing arrangement creates the possibility of recurring cost and dependency. Aggregate reporting can show members the scale of legal support without disclosing matter strategy, hourly narratives or sensitive advice. Periodic review can ask whether the arrangement remains proportionate and whether the adviser’s scope still matches the company’s needs.

The broader economic lesson is that governance controls should follow the mechanism of impact. The mechanism here is not a regulator issuing public rules. It is a private Board making choices within a technically relied-upon organisation. The relevant safeguards are therefore corporate: clear instruments, defined authority, conflicts management, accurate minutes, reviewable reliance, cost oversight and correction. Those safeguards respect AFRINIC’s bounded role while taking the effects of poor administration seriously.

Nine names, two absences and the meaning of unanimity

Numbers in minutes can look more conclusive than they are. The 10 August record names nine people as present: the chief executive and eight directors. That count does not prove that all nine remained throughout the two-hour-and-five-minute meeting. The vote record specifically says Alan Barrett and Krishna Seeburn were not present for Resolution 201608.298. The resolution was nevertheless recorded as unanimous.

There is no contradiction. Unanimity describes the votes of those participating in the decision, not the continuous attendance of everyone named at the meeting’s opening or elsewhere in the minutes. The proposer and seconder are identified, and no dissent is recorded among those voting. This is a clear procedural fact.

Its meaning should remain corporate. A unanimous Board vote can show agreement and make implementation easier. It cannot create a public mandate or cure a lack of substantive power. If every participating director agrees that the company should do something beyond its authority, the agreement does not make the act sovereign. If every director accepts legal advice, the advice does not become a judgment binding the public.

The two recorded absences also reinforce why reliance metadata matters. Alan Barrett introduced the attendance discussion onto the agenda but was not present for the resolution. It would be tempting to tell a story about motive, sponsorship or withdrawal. The record does not support one. Agenda initiation does not prove authorship, and absence does not prove disagreement. The sound account stops at the documented sequence.

Similarly, Sunday Folayan’s proposal and Haitham El Nakhal’s seconding show procedural sponsorship. They do not disclose the resolution’s private drafting history or prove that either man formed a particular legal conclusion. Named roles should be used to locate accountability, not to invent psychology.

A mature minute protocol makes this restraint easier. It records when people join or leave for material votes, distinguishes a proposer from an author, notes recusals separately from ordinary absence and records the decision without embellishment. Such precision reduces the need for speculation. Resolution 201608.298 offers enough detail to identify the vote, but not enough to describe the adviser’s own position within it.

A proportionate protocol for recurring legal advice

The best counterfactual is not to reverse the invitation. It is to surround it with a bounded protocol that captures authority and reliance while preserving candid substance.

First, identify the adviser and the client. The public record need not include private contact details or the content of the retainer. It should state whether counsel advises AFRINIC as a company, the Board in its corporate capacity or another defined client. If circumstances require separate advice for a director or officer, that distinction should be explicit. Clear client identity prevents a standing presence from becoming an ambiguous relationship in which everyone assumes counsel represents them individually.

Second, publish the appointment authority, term and reporting line. Members should know which corporate body appointed the adviser, under what instrument and for what period. They should know who may instruct counsel on behalf of the client and who reviews the engagement. This does not reveal legal advice. It shows that the relationship itself rests on an authorised corporate act.

Third, define scope. The protocol should distinguish a standing invitation from automatic attendance. It can state whether the adviser receives every agenda, who decides which items require attendance, whether counsel may remain for all items and whether a separate invitation applies to executive sessions. It should distinguish listening, responding to a request, proactively flagging a legal concern and delivering a formal opinion. These categories prevent casual remarks from silently acquiring the status of Board-relied advice.

Fourth, establish conflict handling. A recurring adviser should have a defined channel for identifying a possible conflict. The protocol should state who decides whether recusal, limited participation or separate counsel is required. Public minutes can record that a conflict procedure was applied without exposing the confidential facts behind it. If a director has a related conflict, that recusal belongs in the decision record as well.

Fifth, map advice to agenda-level metadata. For each material item, the minute or associated register can state whether advice was sought, whether it was oral or written, the general corporate matter, the governing instrument considered and whether the Board relied on it. The record need not summarize the merits. A compact entry—advice sought on Board authority under a specified company instrument; Board relied on advice in selecting a member-vote procedure—would make the chain intelligible without publishing the analysis.

Sixth, separate records by function. Ordinary minutes should record attendance, authority, decisions, recusals and action owners. An executive-session record can preserve sensitive deliberation at a controlled level. If a privilege log is appropriate, it can identify date, matter, participants, client and record holder without disclosing substance. The existence and precise legal effect of privilege should not be assumed; the point is to keep any claimed protection bounded, reviewable and distinct from ordinary corporate facts.

Seventh, assign waiver and disclosure authority. If material is protected, the protocol should say who may decide whether that protection is asserted, reviewed or waived. Otherwise the organisation risks a one-way system in which secrecy expands automatically but no accountable body can revisit it. The decision to disclose or continue withholding can itself be recorded without publishing protected merits.

Eighth, capture the corporate act and follow-through. Minutes should connect advice, where materially relied upon, to the resolution or action that followed. An action owner and review date can be identified. If the Board later corrects the act, the public record should link the correction to the earlier entry. This turns advice from an untraceable influence into one input within a visible authority chain.

Ninth, report aggregate cost and dependency. Members do not need invoices or strategic matter descriptions to know the overall scale of recurring legal support. Aggregate expenditure, reviewed over a suitable period, can show whether a standing arrangement remains proportionate. The Board can also review concentration risk: whether too much institutional knowledge sits with one adviser, whether handover records are adequate and whether separate expertise is sometimes needed.

Tenth, review the arrangement periodically. A universal invitation should not become permanent by inertia. A scheduled review can assess attendance patterns, the number and kind of matters supported, conflicts, cost, timeliness, usefulness and whether the protocol still protects both candour and accountability. The review need not grade individual legal opinions in public. It should confirm that the company’s need, the adviser’s role and the Board’s recordkeeping remain aligned.

This protocol is proportionate because it does not demand radical transparency. It accepts that legal substance may need protection. It also rejects the idea that the presence of counsel makes the surrounding corporate act unknowable. Identity, appointment authority, client, matter category, governing instrument, conflict handling, reliance, decision, aggregate cost and review are governance facts. Recording them makes a standing interface safer and more useful.

What an accountable minute could have looked like

It is possible to imagine a stronger record for item 7.3 without inventing facts or disclosing advice. The minute could have identified the purpose of the recurring arrangement, the appointment authority under which the adviser served and the distinction between an invitation and actual attendance. It could have stated who would determine the items for which counsel’s participation was needed. It could have assigned responsibility for documenting attendance and any material reliance.

For later meetings, an attendance line could identify when the adviser joined and left. An agenda entry could state that the Board sought advice on its authority under a named company instrument. If the Board relied on that advice, the minute could record the fact of reliance and the action taken. If a sensitive discussion occurred, the ordinary minute could note the transition to an executive session and preserve the protected substance separately. If no advice was sought, the mere existence of the standing invitation need not colour the record.

This approach would reduce several predictable errors. It would prevent invitation from being mistaken for attendance. It would prevent attendance from being mistaken for participation. It would prevent participation from being mistaken for control. It would prevent receipt of advice from being mistaken for reliance. And it would prevent reliance from being mistaken for an enlargement of corporate authority.

The method would also improve public understanding of corrections. The same 10 August minutes approved corrected May minutes and instructed that the previously published version be replaced. That act shows that a Board record can be maintained rather than frozen. A legal-advice protocol could use the same principle: if later review shows that an authority citation, attendance notation or action record was wrong, correct the public metadata while preserving an audit trail. Correction should be a sign of responsible record stewardship, not an admission that every underlying discussion must be disclosed.

An accountable minute is therefore neither a transcript nor a press release. It is a compact map of corporate action. It tells the reader what matter reached the Board, which body could decide it, who participated in the act, what instrument governed, whether advice materially informed the choice, what was resolved and how the action can later be reviewed. The legal merits can remain in the appropriate protected record.

What cannot be concluded from Resolution 201608.298

A bounded conclusion is as important as a strong thesis. The official record does not identify the Legal Adviser. It does not establish whether the adviser attended the 10 August Webex call or any later Board meeting. It does not show a written opinion on the preceding item. It does not prove that advice was delivered live.

The record does not define the client, appointment, retainer, term, scope, fee or conflict procedure. It does not state a rule of privilege or confidentiality. It does not show who controlled redaction or waiver. It does not describe a reliance protocol. It does not establish what further private records existed.

Nothing in the resolution proves secrecy, improper influence, capture, collusion, misconduct or illegality. Nothing proves that counsel caused a later decision, legal expense, dispute or operational loss. Nothing proves that the arrangement improved outcomes. The invitation cannot be used as a shortcut into later litigation, later legal governance structures or an assessment of counsel’s performance.

The immediately preceding advice about an SGMM should likewise remain bounded. It helps explain the sequence by showing legal input followed by discussion of recurring access. It does not make this article an assessment of whether the SGMM recommendation was correct. The distinction preserves the actual subject: the move to a standing meeting interface and the record design it required.

Even the same-meeting examples of executive session, redaction and minute correction have a limited role. They show record-handling techniques available within the meeting record. They do not prove that Resolution 201608.298 caused those techniques, that counsel directed them or that they were part of a privilege system.

These limits do not make the resolution trivial. They make the analysis reliable. The act can be significant without becoming evidence of every later controversy. It can reveal an institutional design question without supporting an allegation. It can justify a better protocol without proving that no protocol existed privately.

The durable lesson: separate counsel’s merits from the Board’s act

Resolution 201608.298 placed legal advice inside AFRINIC’s recurring Board environment. In the best case, that brought timely, context-rich guidance to directors before corporate mistakes became expensive. It could help the Board identify the proper instrument, respect member rights, manage conflicts and keep the company inside its limited private powers.

The same arrangement carried a predictable record risk. If every contact with counsel were treated as an indivisible confidential event, members might see decisions without the authority chain that produced them. If the Board instead recorded all legal substance, it could impair candour and expose genuinely sensitive interests. Neither extreme is necessary.

The workable line runs between merits and metadata. Legal reasoning, exposure assessments and negotiating positions may deserve protection where the applicable relationship supports it. The identity of the corporate decision-maker, appointment authority, client, matter, governing instrument, conflict procedure, fact of reliance, resulting act, aggregate cost and route to correction are different. They are the coordinates by which members can audit a private company.

That audit does not turn members into a sovereign public or AFRINIC into a regulator. It keeps responsibility within the correct private-corporate chain. Counsel advises. Directors decide within powers granted by company instruments and contracts. Staff implement authorised acts. Members exercise the rights they actually hold. Technical records coordinate resource administration; they do not become judgments of a court.

The formal achievement of the August 2016 vote was simple: a universal invitation. Its governance consequence was to make legal access ordinary. Once access becomes ordinary, the record around it must become systematic. The Board needed a way to distinguish when the chair was merely available from when the adviser attended, participated and materially informed a decision. It needed a way to preserve candid substance without allowing legal involvement to erase ordinary accountability.

The archived record does not tell us whether AFRINIC built that system or how the arrangement performed. It proves the invitation and the vote, and no more. But the design standard is clear. A technically consequential private registry benefits from legal advice when that advice keeps the company disciplined and bounded. It loses accountability when the label legal makes authority, reliance and correction disappear. The better institution protects the opinion while preserving the map of who could act, why the act belonged to that body and what followed.