Judicial independence, judicial delay and the missing evidence behind the retirement age argument

By Mohammed Ismail Mohammed Sadaath

Sri Lanka’s justice system is burdened by serious delays, and reforms are unquestionably necessary. The question is not whether reform is needed. The question is whether extending the retirement age of judges of the Supreme Court and Court of Appeal is demonstrably connected to the causes of judicial delay and whether it is likely to produce a measurable reduction in the backlog.

An assessment must distinguish a legitimate policy objective from evidence that a particular intervention will achieve that objective. The government’s reasoning appears to be that longer tenure retains experienced judges; retained experience increases judicial productivity; greater productivity accelerates case disposal; and faster disposal reduces the backlog. The first proposition is intuitive. The remaining links are empirical propositions and require evidence.

The scale and distribution of the backlog matter

Official parliamentary records place the total number of pending cases at 1,131,818. The reported distribution included 5,785 Supreme Court cases, 4,572 Court of Appeal cases, 27,324 High Court cases, 262,665 District Court cases and 813,726 District Magistrate Court and Magistrates’ Court criminal cases, alongside other categories. This distribution is analytically important: only a small fraction of the national backlog sits in the Supreme Court and Court of Appeal. A reform centred on the tenure of superior court judges therefore requires a quantified explanation of how it will materially reduce congestion concentrated elsewhere.

Judicial delay is a system problem 

Criminal justice is a sequential institutional process. A case may depend on police investigation, collection and preservation of evidence, forensic and medico-legal examination, prosecutorial review, Attorney-General advice or indictment, court administration, trial, judgment and appeal. Delay at an upstream stage can prevent the downstream stage from functioning, regardless of the experience of the judge. This is why a credible backlog diagnosis must separate investigative delay, forensic delay, prosecutorial delay, administrative delay and delay arising within adjudication itself.

Government Analyst reports: a documented upstream bottleneck

Parliament’s Committee on Public Accounts documented that legal action and bail decisions can be affected by the non-receipt of Government Analyst reports. The department attributed delays to high demand, staff shortages and resource constraints, and the committee linked delayed reports with suspects remaining in remand for longer periods. This is a concrete causal mechanism: a judge cannot replace a laboratory result with experience, and longer judicial tenure does not make an outstanding scientific report arrive faster. If scientific evidence required for prosecution or bail assessment is unavailable, keeping the judge in office for an additional period does not make that report arrive faster. The appropriate reform questions concern laboratory throughput, staffing, equipment, turnaround times and prioritization of time-sensitive cases.

Forensic medical reports and adjournments

The National Authority for the Protection of Victims of Crime and Witnesses recorded discussions with police officers in which court adjournments and delays in reports from Forensic Medical Officers were identified as matters requiring attention. In cases involving injury, sexual violence, death or other medico-legal questions, delayed reports can postpone evidentiary preparation. Describing the resulting interval simply as court delay risks attributing to judges’ time actually lost elsewhere in the justice chain.

The Attorney-General’s Department: capacity and prosecutorial delay

Parliament has also identified significant delays and capacity constraints in the Attorney-General’s Department. Official 2024 cadre data show 218 approved State Counsel posts but only 144 officers in position, leaving 74 vacancies. A parliamentary committee report also recorded that only about 70 State Counsel were working in the Criminal Division. Serious criminal proceedings depend on prosecutorial review, advice and indictment. If a file is awaiting review or must be returned for further investigation because material is incomplete, the eventual trial judge cannot compensate for that lost time.

These figures matter because serious criminal proceedings depend on prosecutorial review, advice and, where appropriate, indictment. If a file is waiting for legal review, or must be returned for further investigation because material is incomplete, the eventual trial judge cannot compensate for that lost time. The relevant performance indicators therefore include the median time for Attorney-General advice, the number and age of pending files, the proportion returned for further investigation, and the interval between completion of investigation and indictment.

Court staffing and judicial administration

Court performance is an organizational output, not simply an individual attribute of judges. A 2026 parliamentary committee report recorded that shortages of stenographers in Jaffna were contributing to delays in delivering judgments. The Ministry reported requests to fill 753 vacancies in institutions under the Judicial Service Commission and stated that the justice sector alone had a staff shortage of 6,026. Registrars, stenographers, interpreters, translators, clerks, fiscal officers, research staff and information systems all affect the rate at which judicial work can be processed.

International diagnosis: resources, caseload and case management

The IMF’s 2023 Governance Diagnostic Assessment independently described the court system as suffering from poor resources, an overwhelming backlog, limited capacity among legal personnel, antiquated processes and insufficient use of technology for case management. It also noted that judges are consumed by administrative responsibilities that reduce the time available for adjudication. These findings point toward structural capacity and case flow problems rather than a single retirement age variable.

The causal claim behind retirement age extension remains unproven

Retaining experienced judges may preserve institutional knowledge, but retention should not be confused with expansion of judicial capacity. If a judge who would retire is immediately replaced by a qualified successor, extending the incumbent’s tenure does not create an additional judicial position; it changes who occupies that position. The empirically relevant comparison is therefore experienced incumbent versus qualified successor, not experienced incumbent versus an empty bench.

If the government wishes to rely on backlog reduction as a principal justification, it should demonstrate that judges retained beyond the present retirement age dispose of materially more cases or dispose of them materially faster than likely successors after controlling for jurisdiction, case complexity, staff support and caseload. Without such evidence, retaining experience is a plausible administrative preference but not yet a demonstrated backlog policy.

What have previous reforms actually expedited?

The same evidentiary standard should apply to reforms already cited by the government: additional judges, digitalization, case management initiatives, travelling courts and expanded High Court capacity. Public policy evaluation distinguishes outputs from outcomes. Appointing judges or installing a digital platform is an output; a reduction in median disposition time, an improvement in clearance rate or a decline in old pending cases is an outcome. The government should therefore publish before and after performance data showing how many additional cases these reforms have actually expedited.

The missing performance indicator: clearance rate

A backlog cannot be understood from pending case totals alone. A core indicator is the clearance rate: cases disposed of during a period divided by new cases received during that period multiplied by 100. A court receiving 10,000 new cases and disposing of 8,000 has an 80% clearance rate and is accumulating backlog. If it disposes of 11,000, the clearance rate is 110%, and the court is reducing the stock of old cases. If retirement age extension is a backlog intervention, the government should estimate the additional disposals and change in clearance rate attributable specifically to that reform.

Judicial independence and security of tenure

The constitutional question should also be framed carefully. Judicial independence is broader than the absence of direct political instructions in an individual case. It includes institutional safeguards governing appointment, tenure, remuneration, removal and retirement. The UN Basic Principles on the Independence of the Judiciary state that judges’ term of office, security, conditions of service, pensions and retirement age should be secured by law with guaranteed tenure until the applicable mandatory retirement age or expiry of term.

This does not mean that changing a retirement age is inherently incompatible with judicial independence. The more precise question concerns design, predictability and application. A prospective rule applying to future appointments raises different institutional considerations from altering the tenure of judges already holding office. Constitutional design should minimize both actual dependence and reasonable perceptions that judicial tenure may become contingent upon political decisions made during a judge’s term.

Evidence that should precede constitutional amendment

A credible Judicial Backlog Impact Assessment should publish by court and jurisdiction, annual case intake and disposals, clearance rates, median processing time, the age profile of pending cases, case load per judge, judicial vacancies, waiting times for Government Analyst and forensic medical reports, Attorney-General processing times, files returned for further investigation, causes of adjournments, support staff vacancies, measured outcomes of digitalization and case management reforms and the projected additional cases disposed of specifically because of retirement age extension.

The central empirical questions 

Of Sri Lanka’s more than 1.1 million pending cases, how many additional cases does the government project will actually be disposed of because judges remain in office longer? What proportion of total delay is attributable to the retirement of judges compared with delays caused by forensic reports, Government Analyst reports, police investigation, Attorney-General processing, administrative vacancies, adjournments and weak case management?

If credible estimates exist, they should be published and subjected to professional and academic scrutiny. If they do not, the case backlog should not be presented as though it already constitutes an empirically demonstrated justification for altering constitutional tenure.

Does extending tenure raise a judicial independence question?

The constitutional issue requires a more precise formulation than the proposition that any legislative policy concerning courts necessarily constitutes interference. Parliament unquestionably possesses authority to legislate on the structure and administration of the judicial system within the constitutional framework. Establishing courts, determining jurisdiction, providing resources, increasing judicial numbers, improving salaries and adopting procedural or technological reforms are not, merely by reason of being legislative choices, violations of judicial independence. The relevant question is narrower and more demanding: whether a particular amendment to constitutionally secured conditions of judicial tenure is institutionally justified, transparently designed and compatible with the safeguards required to preserve both actual independence and public confidence.

Article 107(5) of the Constitution presently fixes the retirement age of judges of the Supreme Court and Court of Appeal. This does not mean those ages must remain unchanged forever; constitutions can be amended. It means retirement age forms part of the legally predetermined conditions of judicial tenure. The Commonwealth Lawyers Association has similarly observed that any change requires constitutional amendment and should be preceded by adequate public and stakeholder consultation.

Extension and reduction of tenure are different

It is correct to distinguish a reduction of the retirement age from an extension. Lowering the retirement age of serving judges could prematurely terminate judicial office and would therefore raise an obvious and direct security of tenure problem. An extension does not operate in the same way; it enlarges rather than shortens the period during which judicial office may be held. That distinction should be acknowledged.

However, it does not follow that an extension can never engage judicial-independence concerns. The UN Basic Principles on the Independence of the Judiciary treat the term of office, security, conditions of service, pensions and retirement age as matters that should be adequately secured by law and state that judges should enjoy guaranteed tenure until a mandatory retirement age or expiry of their term where one exists. The purpose of such predictability is not limited to preventing dismissal. It is also to ensure that the conditions governing judicial service are established by general rules rather than becoming contingent upon discretionary political choices affecting serving office holders.

The analytically important distinction is therefore between a general, prospective amendment to the judicial service applicable according to predetermined rules and an amendment whose practical effect changes the tenure expectations of judges already occupying office. The latter does not prove corruption, inducement or improper influence. Nevertheless, it legitimately raises questions of institutional design, appearance and public confidence. Constitutional safeguards should seek not only to prevent actual interference but also to minimize institutional arrangements capable of generating a reasonable perception of dependence.

Legislative power is not the end of the inquiry

The separation of powers does not require parliament to remain passive in relation to the justice system. Judicial power must operate through institutions supported by legislation, budgets, personnel, procedure and infrastructure. Parliament may therefore establish and re-organize courts, enact procedural laws, increase the number of judges, improve conditions of service and allocate resources. A serious academic response should expressly recognize this point.

Yet the existence of legislative competence does not settle the constitutional merits of every particular measure. The fact that parliament can constitutionally initiate a reform does not answer whether the reform is necessary, proportionate, well designed or compatible with broader principles of institutional independence. Formal power and substantive justification are different questions. This distinction is especially important where the reform does not merely regulate court administration but alters a constitutionally specified condition of tenure for holders of superior judicial office.

The relevant inquiry is not whether parliament may legislate concerning the judiciary – it plainly may – but whether altering the tenure conditions of serving judges is supported by a compelling and objectively demonstrated institutional rationale, whether the rule is general rather than personalized, whether it is accompanied by adequate safeguards and consultation and whether the same policy objective could be achieved through less constitutionally sensitive interventions.

The debate should not be personalized as a judicial bribe 

The constitutional debate is weakened when it is reduced either to allegations that an extension amounts to a bribe to judges or to the counterclaim that any concern about tenure necessarily insults the integrity of individual judges. Neither position is necessary for serious analysis. There is no need to allege that a Chief Justice or any other judge would alter a judicial decision in anticipation of extended tenure. Such an allegation would require evidence relating to the individual concerned and should not be inferred from institutional criticism alone.

The stronger concern is structural rather than personal. Sound constitutional design seeks to organize judicial tenure so that judges do not need to depend – and cannot reasonably be perceived as depending – on favourable political decisions concerning continuation in office. The objective is to protect judges themselves as much as it is to protect litigants and the public. A rule can therefore deserve institutional scrutiny even when every judge affected by it is personally beyond reproach.

For the same reason, claims that recent judicial appointments followed seniority or that particular appointments were not politically contested do not by themselves determine the merits of retirement age reform. Even if those claims are fully accepted, they relate to a different institutional question. A government may act appropriately in one aspect of judicial appointments while a separate constitutional proposal still requires independent assessment on its own evidence and safeguards.

Historical comparisons and the preventive function of judicial independence

Past controversies involving the removal or impeachment of judges are not exact parallels to an extension of retirement age, and they should not be treated as though they were. A forced removal directly threatens security of tenure in a manner that an automatic extension does not. Acknowledging this difference improves rather than weakens the constitutional analysis.

But judicial independence is a preventive constitutional principle. Its safeguards are intended to structure relations between institutions before improper pressure occurs, not merely to provide a remedy after an obvious attack on a judge. The absence of circumstances comparable to removal, impeachment or overt intimidation therefore does not establish that no institutional independence question exists. Different mechanisms can create different levels and forms of institutional risk.

This is also why the distinction between decisional independence and institutional independence is essential. Decisional independence protects the freedom of judges to decide individual cases without improper interference. Institutional independence concerns the architecture within which that decision-making occurs, including appointment, tenure, removal, remuneration, resources and the relationship between the judiciary and political branches. A reform need not dictate the result of a single case before it becomes appropriate to examine its effects on institutional independence.

Comparative experience shows that extension is possible but design matters

Comparative experience confirms that extending judicial retirement ages is not inherently incompatible with judicial independence. Jurisdictions have increased retirement ages in response to demographic change, judicial manpower shortages and the desire to retain experienced jurists. The comparative lesson, however, is not simply that an extension has occurred elsewhere; it is that the design and context of the reform matter.

Comparative experience also shows that extension is not inherently incompatible with judicial independence. In a recent comparative discussion, the Commonwealth Lawyers Association referred to Hong Kong’s 2019 extension of statutory judicial retirement ages in the context of manpower and retention concerns. The relevant lesson is that design, transitional arrangements, demonstrated manpower needs and safeguards for institutional confidence matter.

Sri Lanka should therefore avoid two absolute positions: that extending retirement age is necessarily an attack on judicial independence or that an extension can never raise an independence concern because it benefits judges rather than removing them. The more defensible position is contextual. Legitimacy depends upon the objective pursued, the empirical necessity for the intervention, the generality and prospectivity of its application, the transparency of the process and the safeguards surrounding implementation.

Bringing the constitutional and empirical questions together

The constitutional and backlog dimensions of the debate should ultimately be analyzed together. If the principal public interest justification for changing judicial tenure is the reduction of case delay, then the strength of the constitutional justification depends partly upon whether the reform is actually capable of delivering that benefit. A weak or untested causal connection between tenure extension and case disposal makes it more difficult to justify altering a constitutionally secured condition of judicial office on efficiency grounds.

This is where the earlier evidence on Government Analyst delays, forensic medical reports, police file quality, Attorney-General capacity, court staffing, adjournments and case management becomes constitutionally relevant as well as administratively relevant. If these documented bottlenecks explain a much larger proportion of delay than judicial retirement, less constitutionally sensitive reforms directed at those bottlenecks may offer a more direct and measurable route to reducing the backlog.

Conversely, if a Judicial Backlog Impact Assessment demonstrates that impending retirements create substantial loss of specialized capacity, that replacement cannot reasonably occur without a measurable reduction in throughput, and that a general prospective extension would materially improve clearance rates, that evidence would strengthen the case for reform. The point is not to pre-judge the empirical result. It is to require that constitutional engineering follow evidence rather than assumption.

Diagnose before prescribing

Sri Lanka urgently needs faster and more accessible justice. But a case awaiting a Government Analyst report will not move merely because the judge is permitted to remain in office longer. A prosecution awaiting forensic medical evidence does not become trial-ready because retirement age changes. An incomplete investigation remains incomplete regardless of judicial seniority. An understaffed Attorney-General’s Department cannot process files faster simply because the eventual trial judge serves an additional period. And a court without adequate stenographic, administrative and case management support remains constrained regardless of the experience of the person sitting on the bench.

The evidence therefore supports a systems diagnosis: Sri Lanka’s delays arise from multiple interacting bottlenecks. The academically defensible position is neither that retirement age extension must automatically be rejected nor that it should automatically be accepted as a backlog solution. First, identify the causes of delay; second, quantify their contribution; third, evaluate reforms already implemented; and fourth, establish the marginal effect of retirement age extension and only then determine whether constitutional amendment is justified by the evidence. Such evidence would also allow future reforms to be assessed against transparent benchmarks rather than broad assumptions about institutional performance.

The unanswered question remains: How many cases will actually be cleared because judges retire later, and is that effect greater than what could be achieved by fixing the documented bottlenecks elsewhere in the justice system?

-This article was originally featured on groundviews.org