Summary

  • AFRINIC's published material identifies legal fees of USD 633,807 for 2021, USD 1,250,527 for 2022, USD 1,133,630 for 2023, USD 27,322 for 2024 and USD 877,929 for 2025: USD 3,923,215 across five years, without a case-by-case account of scope, outcome, fee basis or recoveries.
  • Those totals do not prove waste, corruption or improper billing. A registry in litigation must defend lawful authority, protect members, comply with courts and obtain advice; the unanswered question is whether each mandate was necessary, proportionate and cheaper than credible alternatives.
  • The accounts make opportunity cost visible but not provable. In several years legal fees rivalled or exceeded broad published expense groupings, while the same institution faced account restraint, board paralysis, receivership, election failure and unresolved continuity design.
  • Members need matter-level disclosure that protects privilege: opening authority, issue, budget, actual and forecast cost, phase, court-cost exposure, insurer or third-party funding, alternatives considered, outcome, lessons and any benefit or liability avoided.
  • Registry resilience should have a protected, measurable budget of its own. Legal defence and continuity are not opposites, but legal spending should not be permitted to consume the money, attention and authority needed to preserve records, payment rails, skilled staff, security services and a lawful route back to member control.

The number that needs an explanation, not a verdict

Legal expenditure is an unusually easy line on which to build a morality play. A large figure can be made to imply institutional recklessness; a small one can be presented as prudent restraint. Neither conclusion follows from the total alone. A regional Internet registry may need counsel to answer injunctions, preserve assets, interpret its constitution, comply with court orders, defend staff, recover records or determine who is entitled to act for the company. Refusing to obtain advice can be more expensive than obtaining it.

AFRINIC's figures nonetheless demand scrutiny because the legal bill ceased to be marginal. The organisation's later finance pages identify USD 633,807 for 2021, USD 1,250,527 for 2022, USD 1,133,630 for 2023, USD 27,322 for 2024 and USD 877,929 for 2025. The five-year total is USD 3,923,215. The disclosed pattern is not smooth: it rises sharply, remains above one million dollars for two consecutive years, falls to a single named payment in 2024, and climbs again in 2025.

That pattern can support questions. It cannot support an accusation about a particular invoice. The public pages do not provide engagement letters, hours, rates, phase budgets, case allocation, write-offs, fee disputes, indemnity decisions, insurer reimbursements or a complete schedule of adverse and recovered costs. Some descriptions name a lawyer or firm; they do not explain what matter was handled or whether work crossed several proceedings. A year of low recognised expense may reflect reduced activity, timing, accrual treatment, a different payer or a narrow published classification.

It should not be interpreted as proof that litigation stopped.

The right starting point is therefore a bounded one. AFRINIC spent a material sum under the legal-cost headings it later published. Members cannot use those totals to determine value for money. That information gap is itself a governance fact because members funded the institution, elected its directors when ordinary governance existed, and bore the consequences when disputes reached bank accounts, budgets and board authority.

The public series begins later than the controversy

The assigned period begins in 2018, before the litigation crisis dominated AFRINIC's public life. That is useful because it prevents every expense from being read backward through later conflict. The 2018, 2019 and 2020 annual reports describe a functioning institution building reserves, running technical and membership services, supporting meetings and reporting annual financial results. They do not provide the same easily comparable, counsel-by-counsel legal table now published for 2022 through 2025.

Absence of a separately visible line is not evidence of zero legal cost. Legal advice may sit within professional fees, consultancy, another expense class or an immaterial aggregate. It may include ordinary corporate work unrelated to contentious proceedings. The older reports are therefore a baseline for institutional scale and reporting practice, not a complete pre-crisis legal series.

The evidentiary break matters. A chart beginning with the detailed 2021 figure can show what AFRINIC later disclosed, but it cannot honestly claim to measure the change from 2018 unless earlier classifications are reconciled from ledgers. Nor can it identify the marginal cost caused by one party, one court order or one management decision. Litigation is cumulative. Advice in one year may concern an order made earlier; an appeal may revisit work already done; a costs order may be recognised after the underlying hearing.

A credible historical account would restate each year under common categories. It would distinguish routine corporate advice, employment work, resource disputes, governance and election proceedings, receiver advice, appeals, compliance, recovery matters and costs paid to other parties. It would identify whether tax is included and whether the figure is cash paid or expense recognised. Until that reconciliation exists, the correct public series is explicitly partial.

This restraint is not excessive caution. It protects the analysis from the very defect it is examining: taking a broad accounting label and assigning it a story that the underlying evidence does not establish.

What the five disclosed years actually show

The official figures can be set out without pretending that the labels are more precise than they are.

Financial year Published legal-fee or legal-cost total What the public breakdown adds
2021 USD 633,807 A later member-facing reconciliation identifies the audited total but does not allocate it by matter or counsel.
2022 USD 1,250,527 Four lines are listed, including two firms or practitioners, a contempt-related line and a large payment to C&A Law.
2023 USD 1,133,630 Four lines are listed, dominated by C&A Law, with separate solicitor, adviser and former-director-lawyer descriptions.
2024 USD 27,322 One legal-adviser line is published.
2025 USD 877,929 Five practitioners or firms are listed; no case allocation is supplied on the finance page.

Adding these values produces USD 3,923,215. Arithmetic is the easy part. Interpretation requires at least four additional distinctions.

First, an expense is not necessarily a loss. Advice may prevent a larger liability, preserve a right, obtain clarity or satisfy a court. Second, a court win does not necessarily establish value. A technically successful application can be strategically unnecessary or leave the underlying conflict unresolved. Third, an adverse outcome does not necessarily establish waste. A responsible institution sometimes must defend a position even when the result is uncertain. Fourth, prolonged proceedings can be partly outside AFRINIC's control. Opposing parties, interveners, courts and procedural calendars affect cost.

The totals nevertheless reveal concentration. Legal spending was not a rounding error within an ordinary administrative year. In 2022, the published legal total of USD 1,250,527 was close to the separately published USD 1,322,887 total for a broad set of “other expenses”. In 2023, USD 1,133,630 in legal fees exceeded the USD 601,508 “other expenses” total. In 2025, USD 877,929 in legal costs slightly exceeded the USD 854,266 “other expenses” total. These comparisons do not pit law against every technical cost: salaries, depreciation and other substantial expenses sit elsewhere.

They do show that legal work occupied the scale of an entire major expense grouping.

That scale changes the burden of explanation. A board can reasonably delegate a modest, recurring legal retainer under an approved budget. Spending above one million dollars during constitutional crisis requires stronger evidence of mandate, procurement, monitoring and continuing necessity. The same is true for a receiver or other temporary authority. Emergency power may justify speed; it does not eliminate the duty to preserve a record from which members and a future board can evaluate the choice.

The 2021 accounts contained both capacity and fragility

AFRINIC's 2021 annual report creates the most useful bridge between legal cost and resilience. It recorded closing cash of USD 11.91 million, reserves of roughly USD 9.997 million and an annual surplus of about USD 1.917 million. A later AFRINIC member-facing page identifies USD 633,807 in legal fees for the year. On those figures, the organisation was not a company whose litigation burden had exhausted its balance sheet.

Yet the report also said that all bank accounts were frozen for a period and that AFRINIC could not honour financial commitments. It recorded USD 504,000 in support from stakeholders, while board material authorised recognition of up to USD 600,000 in third-party debt to keep the company operating. The problem was not simply how much money existed. It was whether money remained legally and operationally available for the functions that had to continue.

This distinction complicates any call to “spend less on lawyers”. During an account restraint, legal work may be the means by which payment capacity is restored or a court-approved operating allowance is obtained. Legal spending can therefore be a continuity expense. At the same time, the incident showed that a large reserve and a large legal budget did not amount to a prearranged continuity mechanism. AFRINIC still needed improvised support and third-party payment arrangements.

The governance question is not whether legal and resilience spending can coexist. It is why the institution had to improvise one while already paying materially for the other. A mature litigation plan should include the operational consequences of the relief sought or resisted. Counsel should know the minimum payroll, critical vendors, bank channels, authority documents and service map before an attachment is argued. Finance should know how a court order would affect payments before it arrives. Technical staff should know which systems and people cannot wait for a later hearing.

The 2021 episode thus establishes a practical test for value. Legal work should not be assessed only by orders obtained. It should also be assessed by whether the institution remained able to pay lawful obligations, preserve records, maintain services and explain to members what was protected. A legal strategy that treats continuity as someone else's budget is incomplete even when its legal analysis is sound.

Court costs are evidence, but not a complete invoice

Judgments and minutes sometimes state that an application was dismissed, set aside or allowed “with costs”. Those words matter. They identify potential cost consequences and can show which party prevailed on a procedural step. They do not disclose the amount ultimately taxed, agreed, paid, stayed, set off or recovered. Nor do they identify each party's own solicitor and counsel fees.

The distinction prevents two opposite errors. One is to add the face value of every legal-fee table to every costs order as though the latter were a second known bill. The other is to ignore costs orders because the amount is not printed in the judgment. A responsible account records them as contingent or realised items when the applicable accounting and legal tests are met, while public reporting states the status.

AFRINIC's 2021 financial material reportedly assessed litigation cases as not probable to result in a financial outflow for contingent-liability purposes. That is an accounting judgment at a date, not a guarantee that no case could later create cost. It should be paired with the legal-spending figure and updated risk range, especially where appeals, receiver proceedings or repeated election disputes continue.

A member dashboard need not publish privileged prospects of success. It can state that a matter has an adverse-cost exposure, whether an order has been made, whether amount determination is pending, and whether payment has occurred. It can also identify costs awarded in AFRINIC's favour and the amount actually recovered. Without both directions, the public sees gross legal expense but not the recoveries or avoided liabilities needed to assess net cost.

This is also why named-payee tables are limited public evidence. Knowing that a firm received an aggregate amount does not reveal whether it handled one complex proceeding, several connected matters, an appeal, urgent applications or routine advice. The invoice-level record belongs with authorised reviewers and auditors. The member-level record should aggregate by matter and purpose while preserving privilege, personal data and tactical confidentiality.

Not every legal dollar has the same institutional purpose

The phrase “legal fees” collapses functions that members should judge differently. At minimum, AFRINIC's spending should be divided into six portfolios.

The first is ordinary corporate maintenance: filings, contracts, employment advice, governance formalities and routine interpretation. This cost should be predictable, competitively procured and budgeted like any professional service.

The second is defensive litigation. The company may need to answer claims, preserve its assets or comply with orders. The relevant test is necessity, proportionality and control of exposure, not whether the institution chose the dispute.

The third is affirmative enforcement. When AFRINIC initiates or escalates action to terminate membership, recover resources or seek coercive relief, its burden is higher. Decision-makers should receive a documented merits assessment, operational impact analysis, range of cost, alternatives and clear authority before committing member funds.

The fourth is constitutional restoration. Advice concerning quorum, director authority, member meetings, receivership and elections may be unavoidable after governance collapses. Such work should have a defined exit objective. A restoration mandate that continues without measurable milestones risks financing an indefinite substitute for member government.

The fifth is election dispute and administration. The 2025 finance page separately reports USD 1,043,425 in election-related cost, including receiver, legal-service, platform, verification and logistical entries. That sum should not be relabelled as legal fees. Its existence shows how governance failure creates a broader professional-cost perimeter around litigation. The June election subtotal was USD 931,849 and the September election subtotal USD 111,576. The public needs to know which costs produced reusable controls and which were lost when a process was annulled.

The sixth is independent assurance: external legal opinions, investigations, forensic preservation and advice to an auditor or member body. Independence has value only if the commissioning and reporting arrangement prevents the subject of review from suppressing the conclusion.

These categories make trade-offs visible. A dollar spent defending an emergency order is not equivalent to a dollar spent pursuing a discretionary appeal. A dollar spent restoring a lawful election is not equivalent to a standing retainer. A single table can still protect confidential details while letting members see whether the institution is buying protection, enforcement, delay, restoration or assurance.

Opportunity cost must be demonstrated rather than invented

It is tempting to say that every dollar paid to lawyers was a dollar not paid for resilience. Accounting is not that simple. AFRINIC held reserves and generated surpluses in some relevant years. A project may have been delayed for lack of authority rather than lack of money. Technical spending can rise alongside legal spending. Some resilience measures require governance reform more than cash.

The published tables nonetheless permit a disciplined opportunity-cost inquiry. In 2022, legal fees were more than five times the combined published “computer expenses” and “remote site expenses” lines. In 2023, they were almost six times those two lines. In 2025, legal costs were more than two and a half times their combined amount. These ratios are not a comparison between all law and all technology: staff, infrastructure, depreciation and service costs appear elsewhere. They show why members need a specifically defined continuity portfolio rather than being told that ordinary information-technology expenditure covers resilience.

The missing portfolio would include tested backups of critical registration state, independent custody arrangements, alternative payment rails, vendor cure periods, cross-trained operational staff, documented authority succession, secure credential transfer, court-ready service mapping and periodic recovery exercises. It would include legal work needed to make those arrangements enforceable. It would also include a member-roll and election-continuity plan, because a registry that can restore servers but cannot restore legitimate authority remains fragile.

No public evidence reviewed here proves that a named resilience project was cancelled to pay a named invoice. The correct disclosure would make that relationship testable. Each major unplanned legal increase should identify the budget line that funded it. If it came from contingency, members should see the remaining contingency. If it drew from reserves, the resolution and reserve purpose should be clear. If planned work was deferred, the project, new date and risk acceptance should be recorded.

Opportunity cost then becomes an institutional decision rather than a rhetorical allegation. Members can decide whether an urgent defence justified delaying a recovery exercise. Auditors can test whether the transfer followed authority. A future board can reverse the priority. Without that record, legal expenditure becomes an unchallengeable fact after the money has gone.

Resilience needs its own protected definition

“Registry resilience” can become as vague as “legal services”. A protected budget should therefore begin with functions and service levels. It should state which records must remain accurate, which publication and security services must remain available, which staff roles are essential, which payments are time-critical, and which governance acts are needed to maintain lawful authority.

The budget should have a survival floor and a safe operating level. The survival floor covers only what prevents irreversible loss or unsafe operation during a short crisis. The safe operating level covers sustainable staffing, security monitoring, member support and controlled changes. Normal strategy, outreach and discretionary programmes sit outside both. This prevents management from sheltering every preference under continuity.

Funds alone do not create protection. A separate account in the same legal entity may remain subject to the same court order. A cross-border account may create evasion or governance concerns. A third-party payer may acquire influence or create an unreconciled liability. The legal form must be established prospectively, approved under valid authority, disclosed to auditors and capable of court supervision.

The same discipline applies to a legal reserve. Money set aside for cases should not be presented as available continuity runway. A case reserve estimates exposure and advice needs. A continuity reserve preserves defined functions. Combining them lets whichever crisis arrives first consume the entire pool.

Members should receive both figures in months as well as dollars. How many months of minimum services remain after committed legal budgets and known liabilities? How many months if fee income falls? How much remains immediately accessible if ordinary accounts are restrained? A reserve of millions can look reassuring while answering none of these questions.

This is where legal and technical design meet. Lawyers should help construct a lawful continuity reserve that does not prejudice creditors. Technicians should define the actual service floor. Finance should model timing. Members should approve the policy. Independent reviewers should test it. Resilience is not what remains after litigation; it is an obligation budgeted before litigation.

Authority to retain counsel is part of the cost question

An invoice can be reasonable in amount and still expose a governance problem if no valid organ authorised the engagement. AFRINIC's institutional vacuum made this question unavoidable. When the board lacked quorum, who could approve a new firm, expand a mandate, initiate an appeal or settle? When a receiver acted, which powers came from the appointment and which required court direction? When former directors' lawyers appeared as a published category, whose interests were represented and under what indemnity or company authority?

The 2024 Court of Civil Appeal judgment concerning AFRINIC treated authority to bring an appeal in the company's name as a decisive issue and restored the receiver order. That does not establish that every other engagement was unauthorised. It demonstrates why authority cannot be inferred merely because a lawyer appeared for the institution.

Every matter file should therefore begin with an authority instrument: a quorate resolution, valid delegation, receiver power or court order. The instrument should define scope, spending ceiling, reporting recipient and persons represented. If a director, employee or former officer has separate interests, the file should state whether the company funds that representation and why. Conflict advice should be independent where the approving person may benefit.

Emergency instructions need a short clock. Counsel may have to act before a full meeting can occur. The temporary authorisation should state the immediate step and expire unless ratified by a competent authority. Ratification should not be assumed from silence or from later payment.

This protects lawyers as well as members. Counsel should not have to decide from contested corporate correspondence who speaks for the client. A clear authority record reduces later fee disputes, privilege confusion and challenges to proceedings. It also lets auditors separate company expense from costs that belong to another person.

The legal-spend debate is therefore partly a corporate-identity debate. Before asking whether AFRINIC paid too much, members are entitled to know who lawfully chose the objective for which payment was made.

Procurement should test expertise without manufacturing dependence

Complex proceedings often justify specialist counsel and continuity of representation. Changing lawyers at a critical stage can waste money and knowledge. Those facts should not turn a longstanding engagement into an unreviewable monopoly.

AFRINIC needs procurement rules designed for contentious work. The institution should define the required expertise, seek more than one proposal where time permits, compare staffing and rates, set phase budgets, require periodic forecasts and identify when senior counsel is necessary. For urgent appointments, a retrospective independent review should occur after the immediate risk passes.

Fee arrangements deserve particular care. Hourly billing can be appropriate when scope is uncertain, but it moves duration risk to the client. Fixed or capped phases improve predictability but may encourage narrow definitions and change requests. Success fees can create conflicts or be legally unsuitable. The choice should match the matter and be recorded.

Invoices should show date, professional level, time, task and matter code in sufficient detail for authorised review, without placing privileged content in public accounts. Duplicate attendance, administrative work billed at senior rates, repeated research, unmanaged conferencing and work outside scope should be challenged. A person independent of the day-to-day litigation lead should approve material bills.

None of this implies that the disclosed AFRINIC invoices contained defects. The invoices are not in the public record reviewed here. These are controls made necessary by the scale and concentration of spending, not findings about a firm or practitioner.

Dependence also arises from knowledge custody. The institution should own a complete, indexed case record and receive periodic matter summaries. If only outside counsel can reconstruct orders, deadlines and strategy, changing counsel becomes prohibitively expensive and temporary officeholders become reliant on one adviser. Legal continuity, like technical continuity, requires portable institutional memory.

Alternatives are part of competent legal advice

A litigation budget is incomplete if it prices only the path already chosen. Before major affirmative action or appeal, decision-makers should see at least one credible alternative and the cost of doing nothing. Alternatives may include a standstill, preservation agreement, narrow declaratory application, mediation, expert determination, independent membership review, supervised election remedy, contractual amendment or settlement with public reusable principles.

Alternatives are not signs of weakness. They help isolate the question that actually requires a court. A broad resource dispute may contain a narrow issue of authority that can be resolved first. A contested election may require preservation and a targeted rerun rather than years of general proceedings. A payment restraint may be addressed through supervised operating allowances without deciding the underlying claim.

Each alternative has risks. Settlement can hide precedent and favour repeat players. Mediation may fail. A standstill can preserve an unfair status. Declaratory relief may not address urgent conduct. The point is not that compromise is always cheaper or better. It is that a responsible instruction records why the chosen path offered superior expected value after legal, operational and governance effects were considered.

This record should be refreshed at milestones. A case that was rational at filing may become irrational after an interlocutory ruling, board collapse or change in the evidence. Sunk cost is not a reason to continue. Nor should a temporary procedural success be confused with resolution of the merits.

Members need not receive confidential settlement ranges. They can be told that alternatives were considered, by whom, at what stage and why the chosen class of action remained necessary. That minimal disclosure disciplines decision-makers without prejudicing the case.

A win rate would be the wrong performance measure

Counting orders won and lost would produce a simple dashboard and a misleading one. Litigation outcomes differ in importance, finality and scope. An interim order can preserve a position without resolving rights. A procedural dismissal can be reversed. An appeal may clarify authority while prolonging institutional uncertainty. A settlement may save money while leaving members without a reusable standard.

Legal performance should instead be assessed across five dimensions. The first is legal protection: liability avoided, rights preserved, compliance achieved and finality obtained. The second is continuity: whether records, services, staff, payment capacity and member rights remained protected. The third is governance restoration: whether the matter moved AFRINIC toward a lawfully elected and accountable board. The fourth is economic control: budget accuracy, fee discipline, recoveries and opportunity cost. The fifth is institutional learning: whether contracts, bylaws, controls or dispute procedures were improved.

A matter can score well on one dimension and poorly on another. Urgent counsel may obtain release of funds while the institution fails to repair treasury design. A court may restore authority while election rules remain vulnerable. A settlement may control cost while confidentiality prevents future members from understanding the governing standard.

This multidimensional review should occur after each material phase, not only after final judgment. The legal lead provides facts; finance reconciles cost; operations reports continuity effects; an independent governance reviewer examines authority and member impact. No single entity should grade the work it commissioned.

The final public note should be short but specific. It can state what was decided, what remains open, total cost to date, recoveries, operational effect and control changes. Silence after a case ends allows the same institutional uncertainty to generate the next case.

The 2025 election bill is a warning about false savings

AFRINIC's 2025 finance page reports USD 1,043,425 in election-related cost. The first election subtotal was USD 931,849; the later election subtotal was USD 111,576. The lines include receiver fees, legal-service providers, an election company, a voting platform, identity checks, travel and logistics. The classification is AFRINIC's own and should be preserved: these are election costs, not all legal fees.

The contrast illustrates why resilience cannot be judged by the cheapest immediate option. A process that fails and must be replaced can make the first expenditure largely non-recoverable while extending the receiver's mandate and creating more legal exposure. A more expensive independent verification step at the start may therefore save money, but only if it addresses the actual failure mode.

The public record does not permit a finding that a particular provider caused the June process to fail, that every listed amount was wasted, or that the September process was superior in every respect. The receiver publicly referred to suspected irregularities involving voter documentation and said investigations had not reached final conclusions when annulment was announced. That uncertainty makes a cost-and-control review more important, not less.

Members should receive a post-election reconciliation separating reusable assets from abandoned work. Did identity-verification design carry into the second election? Were vendor credits obtained? Were legal opinions reusable? Which documents, data and controls now support future votes? What costs arose solely because the first process was annulled? The official totals do not answer these questions.

This is the broader lesson for legal fees. The cheapest legal step can be expensive if it produces no durable rule. The most expensive engagement can be justified if it resolves a fundamental authority issue and prevents recurrence. Cost becomes intelligible only when linked to an outcome and a control improvement.

Members need a matter register they can govern

The appropriate disclosure is a legal-matter register, not publication of invoices or privileged advice. Each material matter should have a stable identifier and a concise set of fields: forum, parties or issue class, date opened, decision authority, institutional objective, current phase, lead counsel, approved budget, actual cost, forecast to next milestone, adverse-cost status, recoveries, funding source, operational risk, next decision date and closure outcome.

Sensitive fields can remain restricted. Public reporting can aggregate party names where law or safety requires it. Merits probabilities, strategy, settlement ranges and privileged analysis need not appear. The point is to let members connect money to an authorised institutional purpose.

The register should reconcile to audited accounts. If several firms work on one matter, their totals should roll up to the matter. If one firm works on several matters, the public payee total should reconcile across them. Differences between cash paid, accrued expense and court costs should be explained. Currency conversion and tax treatment should be consistent.

Governance over the register must survive a vacancy. Under an elected board, an audit or finance committee should review it. During receivership or quorum failure, an independent accountant or court-approved reviewer should receive the same information and report within a bounded mandate. A future board should inherit the complete record at handover.

Member reporting should be quarterly during crisis and annual in ordinary periods. A material deviation should trigger an explanation before the next annual report. Waiting years to publish historic legal tables denies members any chance to influence the choice while it remains reversible.

The register would also protect fair criticism. Members could challenge an expanding matter without alleging invoice misconduct. Lawyers could be evaluated on scope and outcomes rather than insinuation. Decision-makers could explain why a costly defence remained necessary. Evidence would replace theatre on all sides.

A priority rule for the next crisis

AFRINIC needs an explicit order in which money and authority are protected when legal conflict intensifies.

First, preserve the verified registry state, security material, essential staff, critical vendors and payment capacity needed to prevent irreversible harm. Second, comply with court orders and obtain the advice necessary to understand them. Third, isolate the disputed act so that litigation does not spread into unrelated member services. Fourth, fund a lawful route back to ordinary member government. Fifth, pursue or defend broader merits issues under an approved, reviewed budget. Discretionary programmes come after those obligations.

This order does not place the institution above the court or creditors. It requires a transparent request for supervised continuity rather than secret asset movement. It does not privilege management over members. It preserves only the functions needed until members can again exercise lawful control. It does not prohibit forceful litigation. It demands that force be targeted and affordable.

The rule should be written before the identity of the next claimant or officeholder is known. That is how it avoids becoming a device for one side. Activation should produce a dated public notice, authority map, minimum-service budget and legal-spend forecast. Extensions should require fresh reasons.

AFRINIC's crisis showed that courts, lawyers, banks, directors, staff and members are all part of registry continuity. Treating legal cost as external to operations made the institution less prepared for the point at which legal process reached operations. Treating technical service as a reason to avoid legal accountability would be equally mistaken.

Legal fees became governance because the choices stayed opaque

The public figures do not prove that AFRINIC bought the wrong advice. They prove that members lack the information needed to determine what was bought. Over five disclosed years, almost USD 3.93 million was recognised under legal headings. In separate 2025 reporting, more than USD 1.04 million was attributed to two election efforts. During the wider period, the institution experienced account restraint, a board without ordinary quorum, a receiver, repeated court supervision and contested elections.

Those facts cannot be collapsed into a claim that litigation alone caused institutional failure. The disputes arose from contested contracts, resource decisions, governance defects and competing assertions of right. Nor can the figures be dismissed as the unavoidable price of defending a critical registry. Necessity must be shown matter by matter, especially when the institution itself chooses escalation.

The missing comparison is resilience. Members need to see not a hypothetical server purchased for every lawyer paid, but a real allocation of money and authority between legal protection, technical continuity and democratic restoration. They need to know which risks were accepted, which projects moved, which court costs remain, and what institutional change followed each major engagement.

Legal spending becomes accountable when it leaves behind more than orders and invoices. It should leave a clearer contract, a safer treasury, a lawful authority chain, a verified electorate, a narrower dispute mechanism or a durable precedent. If it leaves only another hearing date, decision-makers should have to explain why continuation remains the best use of member funds.

Registry resilience is not what AFRINIC can afford after paying lawyers. It is the protected purpose against which every legal mandate should be tested.

Sources and analytical limits

AFRINIC's finance index describes its budget process, including board approval, publication and quarterly monitoring. The detailed official pages for 2022, 2023, 2024 and 2025 provide the legal-fee totals and payee descriptions used here, as well as the published “other expenses” and 2025 election-cost tables. These pages establish AFRINIC's later disclosure; they do not authenticate individual invoices or explain case allocation.

The AFRINIC 2026 special general meeting page supplies the later member-facing reconciliation of the USD 633,807 legal-fee figure for 2021 and the institution's answers about reserves, third-party payment support and litigation assessment. It is an institutional response, not an independent adjudication of value for money.

The 2021 annual report is used for audited cash, reserve, surplus and account-restraint disclosures. The 2018, 2019 and 2020 reports provide the earlier institutional baseline but are not treated as a comparable case-level legal series.

The authority discussion is bounded by the Supreme Court of Mauritius Court of Civil Appeal judgment in African Network Information Centre (AfriNIC) Ltd v Cloud Innovation Ltd and another, 2024 SCJ 473. It supports the significance of authority to act in AFRINIC's name and receivership; it does not establish the validity or value of every legal engagement.

No reviewed public record supplies a complete 2018-present matter ledger, engagement letters, invoice set, hourly rates, insurer recoveries, settlement payments, taxed costs or budget-to-actual history. The article therefore does not allege overbilling, corruption, bad faith or professional fault by any named practitioner, firm, director, employee, receiver or litigant. Its cost ratios are deliberately limited to published categories and are not presented as a complete comparison of legal and technical expenditure.