Sri Lanka’s Central Cultural Fund Investigation: Audits, Allegations and the LKR 11 Billion Claim

The Central Cultural Fund record contains two different stories. The first is documented in statutory audits: approvals were bypassed, records were incomplete, procurement rules were not followed, and auditors could not obtain enough evidence to express an opinion on the 2018 and 2019 financial statements. The second is a politically explosive allegation, announced in 2020, that LKR 11.059 billion had been misappropriated between 2016 and 2019.

Those stories overlap, but they are not the same. The audits establish serious failures of public financial control. They do not establish that LKR 11.059 billion was stolen, that any named person received the money, or that criminal liability has been proved.

This distinction matters because the original version of this investigation reproduced a broken table. It overstated the itemised allegation by LKR 450 million, used citation numbers that did not match its bibliography, described an audit disclaimer too broadly, and treated several interpretations of political motive as though they were findings. This corrected edition returns to the documents.

Five labels are used throughout:

• Audit finding means a matter stated in a report of the Auditor General or National Audit Office.
• Committee allegation means a claim attributed to the Edirisinghe Committee through a public statement or media report; the full committee report was not available for this review.
• Public response means the position publicly attributed to Sajith Premadasa, the SJB or CCF management.
• Author interpretation means an inference drawn from the record. It is not an official finding.
• Current status means the latest position located in official or otherwise identified public material as of 14 August 2026, not a claim that no unpublished action exists.

Evidence cut-off and current-status check: 14 August 2026. No fresh approach to the named people or institutions was made for this migration review; their published responses are presented below. The right of reply remains open. Anyone holding a committee report, enforcement decision, court record or material response not cited here is invited to provide it through the corrections channel.

1. Background: The Central Cultural Fund and the Political Milieu

The controversy cannot be assessed by starting with the LKR 11 billion headline. The Fund’s legal design, revenue model and governing structure explain both its public importance and the risks created when formal controls fail.

1.1. Mandate and Financial Structure of the CCF

The Central Cultural Fund was established under Act No. 57 of 1980. Its institutional account describes a body responsible for conservation, archaeological research, heritage presentation and visitor infrastructure across major sites. It can receive state grants, donations and revenue connected with heritage management. Its Board of Governors is chaired by the Prime Minister and includes ministers and senior officials. The CCF’s own institutional overview sets out this mandate and structure.

That model gives the Fund a degree of revenue autonomy unusual for an ordinary government department. Entrance income from heavily visited heritage sites can finance conservation without relying exclusively on an annual Treasury allocation. Autonomy, however, does not remove public-law controls. The Fund remains accountable for approvals, procurement, accounting, audit evidence and the purposes for which its money is used.

Author analysis. The relevant governance risk is concentration of discretion. A large self-generated revenue stream sits under a board composed mainly of political office-holders and senior state officials. That does not prove abuse. It does mean that board procedure, documentary evidence and independent audit carry unusual weight. When those safeguards weaken, it becomes difficult to separate authorised policy, irregular administration and alleged misuse after the event.

1.2. The Political Landscape (2016-2019)

The period covered by the 2020 allegation crossed four calendar years under the coalition commonly described as the Yahapalanaya government. It was not coterminous with Sajith Premadasa’s responsibility for the CCF. The Ministry of Mass Media records that Premadasa was sworn in as Minister of Housing, Construction and Cultural Affairs on 20 December 2018, and the later gazette allocation placed the CCF under that ministry. His direct ministerial responsibility for the Fund therefore covered only the final part of the 2016–2019 period, not all four years. Political competition inside the governing coalition intensified before the November 2019 presidential election, when Premadasa became the UNP-led alliance’s candidate.

An earlier controversy had already emerged. In September 2019, Daily FT reported Premadasa’s parliamentary denial of allegations concerning withdrawals from Fund deposits. The later Edirisinghe inquiry was therefore not the first time CCF spending had entered an electoral contest.

The political setting is relevant, but it cannot decide the evidence. A claim made during an election is not false merely because the timing is advantageous. An official report is not conclusive merely because its authors held formal titles. The appropriate test is narrower: what document supports each claim, what did that document actually find, and what remains unproved?

2. The Edirisinghe Committee Report: Allegations of an Rs. 11 Billion Misappropriation

The LKR 11.059 billion figure came from a committee appointed after the 2019 change of government. Its public impact has exceeded the accessibility of its evidence. This review found extensive reporting about the report, but not the full report itself.

2.1. The Committee’s Mandate and Composition

Contemporary reporting describes a three-member committee chaired by retired High Court Judge Gamini Sarath Edirisinghe, with former ministry secretary Gotabaya Jayaratne and attorney Harigupta Rohanadeera. It examined CCF activities from 2016 to 2019 and reportedly submitted a 142-page report to Prime Minister Mahinda Rajapaksa on 28 July 2020. Daily FT attributed the announced findings to a statement from the Prime Minister’s Office.

The committee’s composition and handover are observable events. Its underlying interviews, schedules, transaction testing and legal reasoning were not available for this review. The numerical findings must therefore be described as committee allegations reported through official statements and media, not findings independently verified by Research Mind.

2.2. A Reconciled Breakdown of the Allegation

The public announcement used the language of loss and misappropriation. The following table is an arithmetic reconstruction of amounts attributed to the committee. It is not a verified loss schedule or an audit finding.

Reported category	LKR million	Evidentiary status
Recruitment above the approved cadre, including salary and allowance costs	3,060	Committee allegation reported by media
Withdrawal of fixed deposits without approval and associated interest loss	2,608	Committee allegation reported by media
Amount used for other purposes instead of contribution to the Archaeological Trust	2,266	Committee allegation reported by media
Cultural and religious donations or contributions without approval	2,316	Committee allegation reported by media
Sisu Daham Sevana construction spending without the stated approval	753	Committee allegation; related procurement failures appear in statutory audits
Loss associated with conversion of tourist-ticket foreign-currency income	48	Committee allegation reported with differing terminology
Loss associated with transfer of Ape Gama project assets	8	Committee allegation reported by media
Total of reported components	11,059	Arithmetic sum of seven reported amounts; not a verified headline loss

The arithmetic is exact:

3,060 + 2,608 + 2,266 + 2,316 + 753 + 48 + 8 = 11,059

The official 2023 COPE report introduces a second numerical conflict. It describes the committee’s four-year total as LKR 10,500 million, then lists the same seven categories shown above. Those listed figures sum to LKR 11,059 million, not LKR 10,500 million—a difference of LKR 559 million. The report does not reconcile that difference. LKR 11,059 million is therefore the arithmetic total of the seven COPE-listed amounts and the total used in the contemporary public announcement; it is not independent verification that a loss of that value occurred.

The legacy article instead used LKR 2,316 million for both the Archaeological Trust and donation categories. It then added a separate LKR 400 million election-period withdrawal. That produced LKR 11,509 million—LKR 450 million above the announced total. Contemporary accounts distinguish the Archaeological Trust amount as LKR 2,266 million. They also report the LKR 400 million allegation separately rather than including it as an eighth additive component. The reported breakdown and its LKR 11.059 billion total can be checked against the contemporary account.

The separate LKR 400 million allegation concerned an allegedly unauthorised withdrawal during the 2019 presidential-election period. Reports differ over whether the money came from a savings account or a fixed deposit. Without the full committee schedule, the relationship between that amount and the LKR 2.608 billion fixed-deposit category cannot be established safely. It is therefore not added to the table.

The headline also combines unlike concepts. Excess staffing costs, foregone interest, an unpaid institutional contribution, allegedly unapproved donations and asset-transfer losses are not one accounting category. Some may describe expenditure made without proper authority; others may describe opportunity cost, a liability to another body, or an asserted loss. Calling the combined amount “stolen” would go beyond the public evidence.

2.3. Findings on Administrative and Procedural Malpractice

The committee announcement also alleged that 25 current accounts had been maintained without Treasury approval, equipment had been purchased contrary to financial rules, money had been released from a dollar account, and documents had been created after transactions. Some of these themes overlap with statutory audit observations, but overlap does not independently validate every committee characterisation.

One politically important allegation concerned a Board of Governors meeting said to have occurred on 15 November 2019, the day before the presidential election. The committee reportedly said five members attended while signatures of absent members appeared on the record. It recommended verification of the signatures.

That allegation should be kept separate from COPE’s later account of the 209th governing-council meeting. In 2023, COPE reported that seven of 11 members had agreed to decisions, four had not signed, signatures differed from earlier records, and a senior official said he had not participated. The two accounts raise a related authenticity question, but their numbers are not interchangeable.

3. The Auditor General’s Perspective: A Pattern of Systemic Weakness

The strongest evidence in the record is not the committee headline. It is the sequence of statutory audit reports. These documents establish recurring failures of approval, record-keeping, financial reporting, procurement and parliamentary accountability. They also contain management responses, which matter when assessing whether an issue was disputed, accepted or said to be under correction.

The Auditor General’s reports are independent statutory evidence. They are not criminal judgments, and their observations must be stated in audit terms.

3.1. Early Warnings: Key Findings from the 2016-2017 Audits

The 2016 Auditor General’s report documented several failures later echoed in the public controversy:

• Twenty-five current accounts for area offices had been opened without Treasury approval.
• LKR 79,074,872 recorded under other cultural activities had been given to temples, peraheras, viharas and religious festivals without the governing board approval required by the cited provision.
• Fixed-deposit investments of LKR 1.6 billion had been made without governing-board approval.
• A Cabinet decision required 25% of ticket income to be transferred to the Archaeological Management Trust. On 2016 ticket income of LKR 3,445,925,208, the report calculated LKR 861,481,302 as due, while only LKR 104,490,691 had been paid. The resulting shortfall was LKR 756,990,611.
• Annual reports for 2013, 2014 and 2015 had not been tabled in Parliament by 30 October 2017.

These are concrete compliance findings. They do not depend on the later Edirisinghe Committee. They also show why the legacy phrasing required correction: LKR 861.481 million was the calculated contribution due, not the entire unpaid amount.

The 2017 audit continued to identify accounting weaknesses and old unrecovered balances. Its age analysis recorded housing loans of LKR 18,368,291, festival advances of LKR 433,010 and special advances of LKR 87,991 outstanding for more than five years without effective recovery action. It also identified unclear accounting policies and delayed corrective work.

The significance is cumulative. A single missing approval might be an isolated control breach. Repeated unauthorised accounts, unapproved investments, missing board approvals, delayed annual reports and unresolved balances point to a control environment that could not reliably demonstrate that decisions were authorised and recorded when made.

3.2. A Red Flag: The “Disclaimer of Opinion” on the 2018 & 2019 Financials

The Auditor General issued a disclaimer of opinion on the CCF financial statements for both 2018 and 2019. The 2018 report states that sufficient appropriate audit evidence could not be obtained to provide a basis for an opinion. The 2019 report uses the same formal conclusion.

A disclaimer is serious because users cannot rely on the auditor to say whether the financial statements present a true and fair view. Its meaning must nevertheless be kept precise. It is not an adverse judicial finding. It does not prove that every balance was wrong, that money was stolen, or that a named official committed fraud. It means the evidence limitation was so significant that the auditor did not express an opinion.

The 2019 report shows the scale of the documentation problem. The audit schedule identifies LKR 6,033,058,973 in fixed-deposit investments for which fixed-deposit documents, certificates, renewal notices and bank confirmations were requested. It also identifies missing or insufficient schedules for fixed assets and depreciation, receivables, refundable deposits and investment income. Management responses state that some files and schedules had been or would be prepared and submitted later.

Author analysis. These findings support a conclusion of severe reporting and control failure. They do not independently corroborate the committee’s entire LKR 11.059 billion allegation. An inability to verify records increases uncertainty; it does not determine which allegation within that uncertainty is true.

3.3. Case Study in Irregularity: The Audit of the ‘Sisu Daham Sevana’ Program

Sisu Daham Sevana provides the clearest project-level example because the statutory audit sets out its design, expansion and procurement treatment. The 2020 operational audit records an estimated programme value of LKR 1.14 billion in the 2019 budget, originally intended for 332 Dhamma-school buildings.

The audit states that the programme was not implemented through the Cabinet Appointed Procurement Committee required for a procurement of that scale. Construction was assigned directly to three state entities for 45, 112 and 211 buildings respectively. The expanded total of 368 buildings required LKR 1.2822 billion, exceeding the approved allocation by LKR 142.2 million.

The amount commonly presented as LKR 753 million “lost” requires a further correction. The 2022 special audit prints LKR 753.5 million as paid to the construction institutions by 30 October 2019. Its own institutional line items, however, are LKR 121.5 million, LKR 278.7 million and LKR 363.3 million, which total LKR 763.5 million. The same table reports LKR 443.379378 million in certified work and a LKR 320.120622 million balance, which also total LKR 763.5 million. The report does not reconcile this LKR 10 million inconsistency. The audit criticised unsettled advances, incomplete work and failures to take buildings over; it did not determine that either aggregate was a loss or had been stolen.

The reports include management’s position that the Board of Governors had approved the programme and that a Cabinet memorandum had been submitted. The special audit records that the memorandum was returned for resubmission with missing requirements and states that the project began without Cabinet approval. That evidence qualifies Premadasa’s published claim that the projects had the necessary approvals. It does not turn the management response into an admission or decide criminal liability. The narrower audit point is that the required procurement and approval route was not demonstrated.

The numbers also expose a source conflict that should remain visible. The audit refers to 332 buildings planned and 368 allocated after expansion. Premadasa’s 2019 parliamentary account, as reported by Daily FT, referred to 361 Dhamma schools and a target of 1,000. These figures may describe different stages or measures, but the reviewed sources do not reconcile them.

4. The Minister’s Position: Defense and Counter-Allegations

A fair investigation must present the response at the same evidentiary level as the allegation. Premadasa consistently denied misuse and defended the projects as legitimate cultural and religious development. His position does not answer every audit finding, but omitting it would distort the record.

4.1. The Allegations Against Minister Sajith Premadasa

Premadasa was the minister politically responsible for the portfolio from 20 December 2018 during the period in which several of the most disputed 2019 programmes were implemented. The statutory audits do not make a finding about electoral motive or personal political benefit. They can show whether approval and procurement rules were followed, but they cannot, without additional evidence, prove why a project was selected or whether its political effect was its intended purpose. No reviewed court judgment establishes personal enrichment, party financing or criminal misappropriation by Premadasa in relation to the LKR 11.059 billion claim.

The safe conclusion is therefore limited. He held ministerial oversight; spending under the portfolio was subject to serious audit criticism; and a later committee publicly alleged misappropriation. Guilt, innocence and personal motive cannot be inferred from those facts alone.

4.2. The Parliamentary Defense: Justifying Expenditures on Religion, Culture, and National Security

In September 2019, before the Edirisinghe Committee was appointed, Premadasa gave Parliament an account of the spending. Daily FT reported his position that LKR 1.598 billion had been withdrawn from deposits after the Easter Sunday attacks sharply reduced tourism income. He said the money allowed cultural-development work to continue.

His reported examples included:

• the Sisu Daham Sevana construction programme, described as covering 361 Dhamma schools with a target of 1,000;
• LKR 384 million for work at Mirisawetiya and Mihintale;
• LKR 50 million for Thalawila Church and support for the Madhu Shrine; and
• assistance to mosques damaged after the Easter attacks.

Premadasa’s defence had two parts. First, he denied personal or improper use. Second, he argued that the spending served national, cultural and religious purposes across communities and had Cabinet approval.

The first proposition is a denial of corruption or private benefit. The second is a justification for the policy. Neither automatically answers the audit question. Expenditure may pursue a legitimate public purpose and still breach approval, procurement or accounting requirements. Conversely, a procedural breach is not proof that the underlying project had no public value or that its sponsor acted corruptly.

That distinction should replace the legacy article’s claim that the defence amounted to an “ends justify the means” argument. The record supports a clash between project justification and procedural compliance. It does not establish Premadasa’s private reasoning.

4.3. The Public Defense: Framing the Investigation as a Political Smear Campaign

The Samagi Jana Balawegaya characterised the 2020 report as a “political stunt” timed to damage Premadasa before the general election. That was the party’s public response, not a verified finding about the committee’s motive. The report was publicised eight days before polling, but timing alone does not prove why it was released then.

Premadasa later set a more specific challenge. In December 2023, Ada Derana reported that he offered to resign if written proof established that he had used state funds for himself or his party. That formulation addresses personal and party benefit. It does not resolve whether expenditure was properly approved, whether controls failed, or whether a programme produced political advantage without transferring money to a private account.

Both sides narrowed the dispute in ways useful to them. The committee announcement grouped different kinds of loss and non-compliance under one large number. The defence emphasised public purpose and absence of personal gain. The documentary task is to keep those propositions separate.

5. Aftermath and Accountability

The later record is defined by recurring scrutiny without a single public document resolving the headline allegation. COPE examined the Fund, Cabinet approved another committee, and Parliament continued to receive delayed institutional material. Each development has a different scope.

5.1. The Politics of Disclosure: The Report’s Release Before the 2020 General Election

The Edirisinghe Committee reportedly handed over its report on 28 July 2020. Sri Lanka held the parliamentary general election on 5 August. The eight-day interval is a verified feature of the chronology and a legitimate reason to scrutinise how the claims were presented.

What follows from that timing is limited. The release placed the allegation in the final campaign week while Premadasa led the principal opposition campaign. It does not follow that the audit failures were invented, that every committee claim was false, or that the committee was created solely for electoral use. Motive requires evidence beyond a calendar.

Author analysis. The responsible conclusion is that the timing increased the duty of precision. Instead, the public received a headline number assembled from unlike categories while the full report remained unavailable. The headline survived. The arithmetic did not.

5.2. Parliamentary Oversight and the COPE Investigation

COPE returned to the Fund using the Auditor General’s special audit rather than treating the Edirisinghe report as a final determination. Parliament’s published COPE report records the 24 November 2022 examination, while its 2023 follow-up account records concern about the 209th governing-council meeting.

The official report also makes the scale problem explicit. When COPE asked about LKR 10,500 million allegedly used without board approval, the Chief Accounting Officer said the internal audit had then identified LKR 1,298 million in capital expenditure: LKR 481 million for Sasunodaya, LKR 786 million for Sisu Daham Sevana and LKR 31 million for other activities. He added recurrent expenditure context and said the internal investigation had not concluded. The Ministry Secretary then attributed a LKR 10,500 million four-year total to the Edirisinghe Committee, despite listing seven figures that add to LKR 11,059 million. These are competing accounts inside the same parliamentary record, not a clean reconciliation.

According to that account, seven of 11 members had agreed to the meeting’s decisions and four, including the former Prime Minister, had not signed. The Auditor General identified differences between signatures on that list and signatures used at earlier board meetings. The decisions were also said not to have been adopted at the 210th meeting, and a senior official stated that he had not participated in the questioned meeting.

COPE recommended:

• a committee chaired by a retired judge to investigate the issues raised by the special audit for 2015–2019;
• dissolution of the existing single-person internal inquiry;
• referral of the 209th meeting signature list to the CID and Government Inspector;
• recall of the CCF to review progress; and
• presentation of the proposed committee report to COPE within three months.

These are oversight actions, not findings that signatures were forged. The referral shows that the authenticity question was sufficiently serious to warrant examination. The reviewed material did not contain a resulting forensic determination.

5.3. Published CID and CIABOC Outcomes

The legacy article said there was no public evidence of a comprehensive CID or CIABOC investigation into the full LKR 11.059 billion allegation. That formulation was too absolute. Public search cannot prove that an investigation does not exist, particularly where investigative files may lawfully remain confidential.

The narrower finding is this: the sources reviewed for this update did not provide a published CID conclusion, CIABOC decision or court judgment determining the headline allegation. Parliament’s documented referral concerned the signature list. That is not the same as a criminal investigation of every category in the committee total.

The official COPE report records one limited CIABOC status. At COPE’s request, the Chief Accounting Officer quoted a CIABOC letter stating that the Commission had decided not to proceed with file No. 15/28/2018 because sufficient information to act under the Bribery Act had not been revealed. That recorded decision concerns the identified file. It should not be enlarged into a determination of every CCF transaction, every category in the Edirisinghe total, or any later inquiry.

Absence of a published outcome leaves the legal status unresolved. It does not acquit the controls, prove the allegation, or establish that authorities abandoned a case for political reasons.

5.4. The Unresolved Legacy: A New Committee for a New Era

On 7 July 2025, Cabinet approved another three-member inquiry into reported irregularities in financial assistance to religious places and other CCF activities from 2017 to 2020. The official Cabinet decision says the proposal followed the 223rd Board of Governors meeting held on 17 May 2025. It identifies the chair, G.M.W. Pradeep Jayathilaka, as a retired High Court Judge. The legacy article incorrectly called him a retired Supreme Court Judge.

In September 2025, Daily Mirror reported a deputy minister’s statement that the committee had identified more than LKR 8 billion in spending without board approval. The same account said legal action would follow once the final report was submitted. That was a reported preliminary political statement, not the committee’s published final report. No official final report was located by the evidence cut-off, so the LKR 8 billion figure is not adopted as a finding here.

Parliament continued to receive delayed institutional records. The 2016 CCF annual report was presented on 5 February 2025 and taken up again on 9 April 2025. On 20 May 2026, the 2017 CCF annual report was presented. These developments correct the legacy impression that nothing later reached Parliament, but tabling historic annual reports does not resolve the questioned transactions or committee allegations.

The control problem also continued beyond the original period. The National Audit Office’s transactions report for 2021–2023 says the financial statements for 2020–2023 had not been submitted by the report date and records further weaknesses in documentation, investment authority, estimates and board frequency. This later evidence supports an institutional diagnosis; it does not prove the 2016–2019 committee allegation.

The Archaeological Trust issue also needs present-day qualification. The 2016 audit calculated a shortfall against the 25% Cabinet formula. In March 2026, however, COPA recorded that the Department of Archaeology had sometimes requested less than 25% and that some allocated funds remained unused. That later explanation does not retrospectively cure a failure to follow the governing decision, but it complicates any claim that every unpaid rupee was an economic loss to conservation.

Institutional reform remained in progress. A 2023 oversight committee recommended statutory controls on the Fund’s use of money. On 16 March 2026, Cabinet granted policy approval to amend the Central Cultural Fund Act. Policy approval is not enactment and does not adjudicate past conduct. The Parliament Acts listing and government-bills listing reviewed for the 14 August 2026 current-status cut-off contained no CCF amendment.

The continuing pattern is clear: a statutory audit, an internal inquiry, COPE recommendations and another committee have addressed overlapping periods under different mandates. Accountability cannot be measured by the number of committees appointed. It depends on published evidence, reasoned findings, corrective action and, where justified, proceedings conducted under law.

6. Media Framing and Analysis

The public controversy was built largely through mediated fragments: a Prime Minister’s Office statement, headlines, parliamentary speeches, party responses and later committee announcements. This investigation does not claim to have conducted a quantitative media-content analysis. It can still identify recurring framing choices in the cited material.

6.1. Attribution and Competing Frames

The reviewed Daily FT committee account retained attribution in its headline—“Committee alleges”—but its opening paragraph stated that LKR 11 billion “was misappropriated” before returning to attribution to the Prime Minister’s Office. The SJB response article framed the same dispute around transparency and electoral timing. These examples show how the evidentiary status of a claim can shift within coverage.

Attribution is not a stylistic detail in a case like this. “The committee alleged LKR 11.059 billion” accurately identifies the source and status of the proposition. “LKR 11.059 billion was misappropriated” presents the proposition as established fact. The second construction is not justified by the documents reviewed here.

The cited coverage foregrounded different propositions. The 2020 committee account led with the attributed LKR 11 billion allegation; the 2019 parliamentary account foregrounded Premadasa’s explanation of the projects and withdrawals; and the SJB response foregrounded electoral timing. These are competing frames in the reviewed examples, not a classification of Sri Lanka’s media landscape. None replaces the underlying audit documents.

6.2. The Impact of the Electoral Cycle on Journalistic Framing

The sources reviewed here cluster around the July 2020 handover, COPE activity in 2023 and the new inquiry in 2025. This is a description of this investigation’s source chronology, not a quantitative measurement of all coverage. The legacy article went too far by treating that limited pattern as proof that journalists lacked commitment to investigation.

There are more restrained explanations. News coverage follows official events because those events produce documents, statements and public conflict. Long-running financial investigations also require access to records that may not be released. The evidentiary criticism remains valid: repeated republication of the same official statement can create the appearance of independent confirmation when every article ultimately relies on one unavailable report.

For readers, the practical rule is simple. Count original documents, not headlines. Ten articles repeating one statement remain one underlying source.

7. Conclusion: An Assessment of Evidence, Politics, and Institutional Failure

The CCF controversy is neither a simple exoneration nor a proved LKR 11 billion theft. It is a documented failure of financial control surrounded by an unresolved, politically charged allegation. Precision does not weaken that conclusion. It makes the accountability question harder to evade.

7.1. Synthesizing Verified Irregularities and Unproven Allegations

The statutory evidence establishes that:

• accounts, investments and donations proceeded without required approvals;
• a substantial Archaeological Management Trust contribution was not paid in full;
• annual reports were delayed;
• old advances remained unrecovered;
• auditors could not obtain sufficient evidence to express opinions on the 2018 and 2019 financial statements;
• major balances lacked adequate schedules or supporting records; and
• Sisu Daham Sevana expanded beyond its approved allocation while the required procurement route was not followed.

The statutory evidence does not establish that LKR 11.059 billion was stolen, that the committee’s aggregation used legally comparable categories, or that Premadasa personally or criminally benefited.

The committee allegation remains material because its individual categories overlap with real audit concerns. It remains unproved because the full report and underlying evidence were unavailable, the public arithmetic was repeatedly mistranscribed, and no reviewed judicial decision determined the headline claim.

A disclaimer of opinion captures the central problem. The records were insufficient for the auditor to express an opinion. That uncertainty should prompt investigation and correction. It should not be filled with whichever political conclusion a reader already prefers.

7.2. The Interplay of Political Patronage and Public Finance Management

Public expenditure can serve a lawful cultural or religious purpose and still be administered improperly. This case shows why purpose and process must be tested separately.

Premadasa’s examples describe visible projects with beneficiaries across religious communities. The audit record describes missing approvals, procurement failures and weak documentation. Both can be true. A building does not prove that its contract was lawful. A procurement breach does not prove that the building was worthless or that its sponsor stole the money.

Author analysis. Politically visible grant programmes create patronage risk because the minister associated with them may receive reputational benefit even where no money enters a private account. That is a governance risk, not a finding about Premadasa’s intent. The proper controls are transparent criteria, recorded board decisions, competitive or lawfully exempt procurement, public beneficiary schedules and timely audit evidence.

The answer is not to prohibit cultural support. It is to make the decision trail strong enough that a later government cannot plausibly convert administrative opacity into an unanswerable corruption headline.

7.3. Enduring Weaknesses in Sri Lanka’s Governance and Accountability Mechanisms

The failure spans more than one office. CCF management did not maintain a record sufficient for timely audit assurance. The governing system did not prevent repeated approval and procurement failures. Annual reports reached Parliament years late. Committees revisited overlapping periods without producing a single, readily inspectable public resolution.

The political system then converted weak records into competing certainties. One side treated an announced committee total as proved theft. The other treated public purpose and absence of demonstrated personal gain as a complete answer. Neither position addresses the full record.

Three actions would materially improve accountability:

1. Publish the complete reports of the 2020 and 2025 committees, subject only to lawful redactions, with their schedules and management responses.
2. Publish the status and outcome of COPE referrals and any resulting corrective or enforcement action without implying guilt before a lawful determination.
3. Require the CCF to maintain a public register of governing-board approvals, major grants, procurement method, recipient institution, project status and audit follow-up.

The most defensible conclusion is direct. The CCF’s controls failed seriously. The LKR 11.059 billion allegation has not been established by the public evidence reviewed here. Both propositions demand attention.

Readers, institutions and people named in this report may submit a documented correction, missing primary record or substantive response through the site correction process. A material response will be assessed against the same evidence standard and, where warranted, added with a dated correction entry.

Sources

The source list in the legacy article contained 56 numbered references whose numbers did not correspond to its bibliography; one repeatedly cited number had no bibliography entry. Those references have been replaced by direct claim-level links. The structured source list attached to this work records the principal documents and reports reviewed.

Official Reports and Government Publications

The central evidence consists of the CCF institutional record, Auditor General and National Audit Office reports for 2016–2020, Parliament’s COPE material, the July 2025 Cabinet decision and the May 2026 parliamentary record. These sources establish what an institution reported or decided. They do not establish facts outside the scope of that document.

Media Articles

Media reports are used where the full committee report was unavailable, to record contemporary publication of the allegation, and to present Premadasa’s and the SJB’s responses. Repetition across outlets is not treated as independent corroboration where the reports rely on the same government statement.

Research transparency

Methodology

Documentary investigation with an evidence cut-off of 14 August 2026. The review gives greatest weight to the Central Cultural Fund Act and the Fund’s institutional record, statutory reports of Sri Lanka’s Auditor General and National Audit Office for 2016 to 2020, the 2022 special audit, the transactions audit for 2021 to 2023, Parliament’s 2023 COPE report and later official records, and Cabinet decisions. Media reporting is used to reconstruct the publicly announced claims of the 2020 Edirisinghe Committee because its full report was not available among the reviewed public records, and to record the responses of people and parties concerned. Each proposition is classified in the text as an official audit finding, a committee allegation reported through the media, a public response, author interpretation, or current status as of the cut-off. The legacy arithmetic and citations were independently reconciled against the source documents. This is a documentary review, not a forensic audit or determination of civil or criminal liability.

Key findings

• Statutory audits establish repeated failures in approvals, accounting records, procurement, reporting and supporting documentation at the Central Cultural Fund. They do not establish that LKR 11.059 billion was stolen or identify criminal liability.
• The publicly reported components of the Edirisinghe Committee allegation total exactly LKR 11.059 billion when LKR 2.266 billion, rather than LKR 2.316 billion, is used for the Archaeological Trust category. A separate LKR 400 million election-period allegation is excluded because the public record does not establish whether it overlaps the LKR 2.608 billion category.
• The Auditor General’s disclaimers of opinion for 2018 and 2019 mean that sufficient appropriate audit evidence was unavailable to support an audit opinion. A disclaimer is not, by itself, a finding of fraud.
• The Sisu Daham Sevana audit provides a documented example of control failure: a programme estimated at LKR 1.14 billion for 332 buildings expanded to 368 buildings requiring LKR 1.2822 billion, while mandatory procurement procedures were not followed. The special audit prints LKR 753.5 million as paid, but its institutional line items and its certified-work-plus-balance figures each total LKR 763.5 million. Neither figure was itself labelled a loss by the audit.
• Sajith Premadasa denied misuse, described the spending as cultural and religious development, and disputed personal or party benefit. Those responses do not resolve the documented compliance questions, but they form an essential part of a fair account.
• Later COPE and Cabinet inquiries show continued institutional concern. The reviewed sources did not contain a published final report from the committee appointed in 2025 or a judicial determination of the headline LKR 11.059 billion allegation by the evidence cut-off.

Limitations

The full 2020 Edirisinghe Committee report, its working papers, witness material and underlying transaction files were not available in the reviewed public sources. The investigation did not have access to CCF bank records, procurement files, Cabinet memoranda, police files, CIABOC files or the complete record of later committees. Absence of a document from the reviewed public record is not proof that no investigation or document exists. Audit findings concern financial reporting, compliance and control; they should not be converted into findings of theft, corruption or personal enrichment. Media reports sometimes differ on terminology, the account from which LKR 400 million was withdrawn, and the status of later proceedings. The evidence cut-off should be updated if an official committee report, court record or enforcement decision becomes public.

Funding

This investigation received no funding.

Conflicts of interest and AI assistance

The author declares no relationships with the institutions discussed in this investigation or with any political party. This investigation concerns contested allegations involving public institutions and political figures. It does not allege criminal guilt or offer a legal opinion. AI assistance was used to extract the legacy record, locate sources, reconcile arithmetic and restructure the article, and to prepare the PDF edition. The cited records are provided for inspection, and the author remains responsible for the analysis and final publication.

PDF edition note

This PDF presents the corrected investigation with its evidence cut-off of 14 August 2026. Preparation of this edition does not represent a new search for subsequent developments. The author confirmed the funding and relationship declarations on 7 September 2026. The original publication date and existing correction history are retained.