Connect with us

Features

Working with Lalith A; and an Indian rope trick

Published

on

(Excerpted from the memoirs of Chandra Wicremesinghe, Rtd. Addl. Secy to the President)

I assumed duties as Additional Secretary in the newly created Ministry of National Security sometime in 1984 and was shortly afterwards appointed Additional Secretary Ministry of Defence as well. The Minister of National Security was the late Lalith Athulathmudali and the Minister of Defence was President, J.R.Jayewardene. The Secretary to the Ministry of National Security was the late Mr.DBIPS Siriwardhana while Secy/Defence. was Gen. Sepala Attygalla.

Mr. Athulathmudali having been at Oxford University, where he had been the President of the Oxford Union, was a brilliant speaker with a sharp intellect to match. Endowed abundantly with these twin attributes, he was able to represent SL at international fora with aplomb and finesse. Many will recall his scintillating performance at a BBC interview where the BBC interviewer tried his utmost to embarrass him and SL over human rights issues et al. Minister Athulathmudali stood up to the barrage of questions with admirable sangfroid, giving cogent answers which left the interogator nonplussed and at a loss for words.

Despite his intellectual brilliance Minister Athulathmudali had his weak moments when he used to quite inexplicably go off at a tangent. This was, I must say, a little known side of his colourful personality. He used to be suddenly obsessed with some pet scheme of his or by some sudden hunch which used to be pursued by him with extraordinary tenacity. One such ignominious episode was the sudden infatuation he took to a confidence trickster called Kelly Senanayake.This man had inveigled himself into the Minister’s confidence promising to get the JVP to give up their armed struggle and come into the political mainstream. It took a while for the minister to discover that KS was a fraud and a cheap crook who had succeeded in leading him up the garden path!

Again, I remember the Minister summoning me to his office one morning and saying that he had a two pronged strategy to bring about reconciliation and amity between the Sinhala and the Tamil people. He outlined his strategy as follows:

1) The settling of Tamils in the South and Sinhalese in the North. He elaborated further that he had already worked out a plan to settle Tamil people in Agalawatte and in Kalutara to start with. He seemed convinced that his plan would bring about amity between the two communities.

2) The Minister also proposed closing down all Universities for a period of two years and converting them into Rehabilitation Centres for the JVP and LTTE cadres who were in detention camps.

I made known to the Minister my own misgivings particularly regarding his second proposal on the grounds that there would be a violent uproar over the proposal by the local undergraduates and the academic staff; while in the International arena, we would be accused of running concentration camps. He however dismissed my apprehensions as being groundless and wanted me to immediately meet Dr. Stanley Kalpage, who was the UGC Chairman at the time and sound him on the proposal.

Accordingly, I went over to the UGC office and met Dr. Kalpage and conveyed to him the plan the Minister had in mind. Kalpage was simply aghast at the idea of closing down the Universities for two years and said that it was a ‘crazy plan’! He also rang up the President and made an early appointment to see him telling me that he was keen on meeting the President before the Minister met him. In the absence of further developments on the matter thereafter, I felt that the Minister’s plan had been shot down by the President.

A few days later Mr. Athulathmudali called me and said that there was a Seminar on the ‘Rehabilitation of Terrorists’ scheduled to be held in Bangkok and added that he was not sending me for it as I did not seem to believe in such rehabilitative approaches. It amazes me even now how a person with such a brilliant mind could get carried away to the point of pursuing schemes which many would consider ill-conceived and highly impractical. It could perhaps be attributed to the streak of intellectual arrogance he had, despite being an eminently likable and personable individual.

Secy/National Security ,Mr. DBIPS Siriwardhana was a person with a keen intellect and a razor sharp mind. As he had many years of experience in high positions in the Public Service, working immediately under him was indeed a rewarding experience immensely beneficial to me. He was quick in attending to files and was famed for having a clean table devoid of files. A literally clean table was an obsession with him and one got the impression that he was waiting for papers to be placed in the in tray to pounce on them and dispose of them almost with undisguised glee! He was indeed phenomenally quick while at the same time being intensely focused on studying the papers submitted to him for orders (which he did in double quick time), attending to them with remarkable facility and promptitude.

His orders were brief and clear and written in a beautiful, flowing hand. Brevity and crystal clear clarity of expression, were his singular forte. I have no doubt that many who had the fortune to work with him benefited immensely by their interaction with him. He was however, at times cynical, often making snide remarks (in rather loud whispers) during meetings even with Minister Athulathmudali, which were strangely enough ignored by the latter. I was serving two Secretaries at the time, the other being General Sepala Attygalla who was Secretary/Defence. I had no problems with either of them and despite the trying times the country went through at that time with the LTTE and the JVP. I attended to the duties entrusted to me diligently and to their satisfaction.

I was appointed a Council Member of the National Dangerous Drugs Control Board by H.E. the President in 1986 and continued to function as Council Member of NDDCB, till 1977. As a member of the NDDCB, I participated in several seminars overseas.

Bringing home Lankan refugees settled in India

It was in 1985, if I recall correctly, that President JR appointed me to Chair a Committee to arrange for the transportation and the resettlement of SL Tamils who had fled the island following the communal disturbances of 1983. The other members of the Committee were Mr. Nirupam Sen who was Deputy Indian High Commissioner in SL at the time, the Controller of Immigration and one or two other Senior Govt. officials.

The Committee had three sittings in all, which were held in the conference section of the room of Secy./Defence, Gen Attygalla. The meeting opened in a spirit of cordiality and candor, with the Deputy HC/India promising to extend all logistical support necessary to SL in having the Tamil refugees transported back to the island. Preliminarily, the Committee looked at the magnitude of the problem taking into account the numbers involved, the location of the refugee camps in South India, the transportation problems and finally, arrangements to be made at this end for their re-settlement. In the course of our discussions Mr. Sen ventured to say that as the Palk Straits were somewhat choppy at that time of the year and with the numbers to be transported being substantial, the crossing via the Straits may be quite risky.

Further, he suggested that rather than engaging many small boats for the purpose, it may be more convenient and advantageous to charter two big vessels to transport the refugees bypassing the Palk Straits. He suggested almost as a matter of course, going round the Southern coast and berthing the big vessels in Trincomalee and making Trinco the disembarkation point. This was the time High Commissioner Dixit was acting like a Satrap trying to treat SL like a colony of India. I for one, disliked Dixit’s overbearing demeanour and downright arrogance and whenever he walked into General Attygalla’s office, I made it a point to get up and leave the room abruptly, conveying in no uncertain terms my dislike of the man.

Having my own suspicions about Sen’s move to off load the refugees in Trinco, I immediately pointed out that we should use the traditional passage through the Palk Straits to bring the Tamil refugees back. If the weather was rough and the sea unruly I added, it would still be preferable to postpone their transportation till the weather improved and bring the refugees back via the Palk Straits. I also hastened to point out that the Tamil refugees were for the most part from villages in the Mannar and Vavuniya Districts and it would facilitate the logistics of their inland transportation and resettlement if they came through the Palk Straits and got off at Mannar.

The Indian Deputy HC thereupon requested me to fix the next Committee meeting giving about 10 days time for him to re-canvass the issues of the mode of transportation, the route to be taken and the point of disembarkation of refugees with his Govt. Accordingly, I requested him to inform me when he was ready to have the second round of talks so that I could convene another meeting of the Committee thereafter. He contacted me about a week later and said he was ready to have the next round of talks. Thereupon a date mutually agreed on was fixed.

To my surprise Sen arrived at the meeting accompanied by a couple of others, one of whom was introduced to me as the Dy/Secy. of the Ministry of Rehabilitation in Delhi and the other as a Senior official in that Ministry. This Dy./Secy who was a big made individual, sat in the chair next to mine and without any further ado tried to commandeer the meeting by saying authoritatively: “Gentlemen, Mr. Sen has been briefing us on certain issues that have arisen concerning the transportation of the SL Tamil refugees in India back to SL. The Indian Govt. has chartered two ships to transport the entire lot of refugees in the different refugee camps in India to SL in one operation. The ships will leave the Indian ports the day after tomorrow with the refugees and will go round the Southern coast of the island and anchor in Trincomalee harbour where they will disembark”.

Realizing that this unprepossessing gentleman was trying to bulldoze his way through with bludgeoning tactics, I maintained a straight face throughout this unexpected outburst. The moment he stopped his harangue, I looked him straight in the eye and said quietly that the SL Govt. was not agreeable to the arrangement which had not been even discussed nor mutually agreed upon by the two sides. This gentleman thereupon said that there was no going back on the arrangement as the two ships which were already chartered, would be leaving India in two days time. At this stage I told him that the only thing for the Indian Govt. to do was to cancel the charter as SL will not permit the disembarkation of the refugees in Trincomalee.

Realizing that we were not going to give in on the issue, he asked me whether any other alternative could be suggested. I conveyed to him that if the refugees could not be brought via the traditional route of the Palk Straits the only other alternative was for the two ships to circumnavigate the island and proceed to Kayts. He immediately said that the Kayts pier could not berth the two large vessels to which I replied that arrangements could be made for the ships to be anchored in mid–ocean so that the refugees could be ferried ashore in barges. Knowing that he would not be able to have his way, this gentleman whose name I have forgotten, got up abruptly saying tersely ” This will not do!” and stormed out of the room with the other Indians including Dy/HC Sen following close on his heels.

I brought what transpired at the meeting to Gen. Attygalle who said that it was good that a firm stand was taken by us not to permit the vessels to proceed to Trinco and disembark the refugees there. I fixed the final meeting of the Committee to take place five days later, inviting Sen for same. As expected, he failed to attend the meeting. In consultation with the other members of the Committee, I wrote the report and sent it to Mr. Sen for his signature. Expectedly, Sen refused to subscribe to the document (quoting a line from Rousseau) and returned it saying that he would not be signing it as he did not agree with the recommendations made in the report.

I submitted the Report to Gen. Attygalla who read it and said he agreed fully with the recommendations made as the Indian Intelligence arm RAW was up to tricks in SL and the insistence on Trincomalee as the port of disembarkation was one of their machinations to bring in Indians in droves to Trincomalee along with the SL refugees and set up a little Indian colony there. This was the time Dixit, who was acting like a Satrap, had prevailed on the SL Govt. on various dubious grounds, even citing SL’s own security interests, to permit Indian officials and even Indian Service personnel to enter SL sans visas.

Secretary Defence had, I was told, handed over the Report to the President at the weekly Security Council meeting. On being told by Gen Attygalla that Deputy HC Nirupen Sen had refused to sign it, the President had startd reading the document smiling to himself occasionally. This was told to me by Gen Nalin Seneviratne the Army Commander, who also told me that the President had spent a good 20 minutes reading the Report and had not proceeded with the meeting till he had finished it. (Nalin also told me jokingly, not to write such lengthy reports as the Service Commanders had been kept twiddling their thumbs till the President finished reading the report). He also said that President JR had given the Report back to Secy /Defence saying that it was a good report.

While in the Ministry of Defence, I was able to associate closely with Gen.Nalin Seneviratne and IGP Cyril Herat, two rare gentlemen who headed the Army and the Police Force respectively. They were officers who possessed outstanding leadership qualities and were widely respected for their unimpeachable integrity and the high principles they followed in the discharge of their official duties. IGP Herat in fact, took a scrupulously principled stand by opting to retire prematurely, rather than yield to the importunate insistence of President JR, to promote a certain Police Officer, whose promotion, the IGP felt strongly, would have been grossly unfair by certain other officers who were far ahead in seniority and who in many other respects, merited promotion much more.

(To be continued)



Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Features

Ditwah: A Country Tested, A People United

Published

on

Floods caused by Cyclone Ditwah

When Cyclone Ditwah roared across the island on November 27 and 28, 2025, it left behind a landscape scarcely recognisable to its own inhabitants—homes reduced to rubbles, vital infrastructure torn apart and entire communities engulfed by floodwaters that surged with terrifying speed. The storm’s ferocity carved deep scars into the island’s social and economic fabric, displacing thousands and severing lifelines that families had relied upon for generations. In its aftermath, the air hung heavy not only with the scent of mud and debris, but also with a palpable collective grief—a profound sense of loss etched on every face. As of December 9, the day of writing, the death toll had reached 635, with an additional 192 individuals reported missing. In Kandy alone, one of the most severely affected districts, 234 lives were lost. Island-wide, 12,123 families—amounting to 1,776,103 people—were displaced.

As a small island situated in the monsoon-fed waters of the Indian Ocean, Sri Lanka has long lived in intimate coexistence with hydro-meteorological hazards. For centuries, the monsoon winds that swept across the island brought not only life-giving rains to nourish paddy fields, forests, and communities, but also shaped the rhythms of daily life, agriculture, culture and even the island’s civilisation itself. Yet this same monsoon—when delayed, intensified, or disrupted—has had the power to unsettle entire ways of life and inflict widespread human suffering. Over generations, communities learned to read the sky and the sea, developing localised knowledge systems and adaptive skills to cope with the uncertainties of winds and waves. This reservoir of traditional wisdom fostered a form of social resilience deeply embedded in the island’s cultural fabric. At present, however, this traditional resilience is increasingly tested by the new realities of climate change and the growing frequency of severe cyclones.

When Cyclone Ditwah struck on November 27, 2025, it unleashed a force so violent that it reshaped many districts within hours, leaving behind a trail of destruction that stretched as far as the eye could see. Whole neighborhoods were crushed under winds that tore roofs from their foundations, while surging floodwaters swept through villages, carrying away homes, livelihoods, and the fragile sense of security people had built over generations. Roads lay fractured, communication lines collapsed, and families found themselves cut off in pockets of isolation marked by debris and despair. In the storm’s wake, the silence was haunting—broken only by the cries of survivors searching for loved ones and the distant hum of rescue teams navigating the ruins. The scale of the devastation was overwhelming, a human and infrastructural tragedy so profound that it demanded not just an emergency response, but a coordinated, compassionate, and deeply human-centered approach to crisis management.

The most devastating natural disaster Sri Lanka has experienced in recent history remains the 2004 Indian Ocean tsunami, which claimed over 35,000 lives and displaced nearly a million people. Sweeping across two-thirds of the nation’s coastline—more than 1,000 kilometers—it affected approximately 234,000 families and destroyed over a million houses. More than two-thirds of the country’s fishing fleet was obliterated. Beyond the immense human suffering, the tsunami exposed profound gaps in preparedness and underscored the urgent need for a systematic, coordinated approach to disaster risk management.

Over the last decade, Sri Lanka has increasingly confronted hydro-meteorological hazards driven by the accelerating impacts of climate change. Cyclones such as Roanu (2016), Mora (2017), Burevi and Amphan (2020), and Yaas (2021) highlight the growing frequency and severity of extreme weather events. According to the Sri Lanka – Disaster Management Reference Handbook, Cyclone Roanu brought the highest recorded rainfall in more than 18 years, triggering floods in 24 of the country’s 25 districts. Covering 1,400 square kilometers, the flooding affected nearly half a million people and inflicted damages estimated at US$600 million. Just a year later, Cyclone Mora caused severe flooding across 15 southern districts and unleashed landslides that further compounded human and infrastructural losses.

These climate-induced pressures have been accompanied by increasingly destructive monsoon-related disasters. In May 2016, the Aranayake landslide wiped an entire village off the map, killing 144 people, leaving 96 missing, and rendering hundreds homeless as their dwellings were buried under rubble. The following year, unprecedented monsoon rains caused flash floods and landslides that killed more than 210 people and displaced 630,000 across 15 districts. Subsequent monsoon seasons delivered similar devastation: in 2018, floods and landslides resulted in 24 deaths and affected 170,000 people; in 2019, heavy rains left 16 dead and displaced more than 7,000. Even in 2020, despite the successful evacuation of more than 75,000 residents ahead of Cyclone Burevi—an example of improved preparedness—post-cyclone flooding still affected over 100,000 people and destroyed or damaged nearly 4,000 homes.

Compounding this pattern of extreme rainfall and flooding is the paradoxical increase in drought conditions, another manifestation of climate variability. The worst drought in four decades struck between October 2016 and October 2017, affecting 2.2 million people across the North Western, North Central, Northern, and Eastern Provinces. From March to May 2020, another severe drought impacted more than 500,000 individuals in 14 districts, forcing the government to implement emergency drinking water distribution across six provinces. These cycles of excess and scarcity are further aggravated by the seasonal rise in vector and rodent-borne diseases—most notably dengue fever and leptospirosis—adding another layer of complexity to Sri Lanka’s disaster management landscape.

Societal Resilience in Disaster Management

As these converging crises demonstrate, Sri Lanka’s vulnerability to climate-driven disasters is no longer episodic but structural—woven into the lived reality of communities across the island. Yet amid repeated cycles of loss and recovery, what stands out most is not only the scale of devastation but the remarkable capacity of ordinary people to adapt, support one another, and rebuild their lives. This enduring strength points to a deeper truth: effective disaster management cannot rely solely on institutions or technologies; it must draw upon—and reinforce—the social resilience embedded within communities themselves.

Having lived under the influence of monsoons for generations, traditional communities developed sophisticated knowledge and skills to cope with nature’s unpredictability. Long before formal disaster management systems existed, villagers relied on environmental cues and collective action to prepare for seasonal threats. In the upstream and valley areas of the Kalu Ganga, for example, older generations still recall how communities repaired boats and rafts through shramadana well before the rainy season began. They observed the behavior of birds, animals, and changes in wind patterns to decode early warning signs that modern meteorology would later confirm.

Such practices demonstrate that traditional communities were not merely passive recipients of natural hazards; they were active interpreters of their environment. Their resilience stemmed from a deep ecological intimacy, a lived knowledge system refined through experience. Today, there is immense value in unpacking this traditional knowledge and synergising it with modern technology—not to romanticise the past, but to strengthen contemporary preparedness.

The Role of Community and the Political Domain

Building societal resilience requires more than cultural memory; it demands structured collaboration between communities and the political system. While communities are often the first responders in any disaster, the political domain plays a crucial role in mobilising, legitimising, and coordinating their efforts. Transforming political will into national will requires an organic articulation between civil society and political leadership—a partnership where both domains reinforce one another rather than operate in isolation. Within this broader framework, disaster management encompasses three equally critical components:

Disaster Risk Management

In each of these, the state has a vital role—from policy formulation to resource allocation, coordination, and accountability. Yet, the effectiveness of state-led initiatives ultimately hinges on the strength of the relationship between institutions and the communities they serve.

Beyond Culture: Technology and Institutions as Pillars of Resilience

While socio-cultural resilience forms an indispensable foundation, it is no longer sufficient on its own, given the scale and complexity of contemporary climate-induced hazards.

Modern disaster risk management relies on a robust interface between technology, institutional networks, and community participation. Advanced and accessible communication technologies—early-warning systems, mobile alerts, satellite data, and community-level dissemination platforms—play a crucial role in transforming timely information into effective action.

But technological tools reach their full potential only when supported by strong institutional structures, in both formal and informal, capable of mobilising people and resources rapidly and equitably. Thus, societal resilience can be understood as a system supported by three interdependent pillars.

Societal Resilience

When these elements function in harmony, the collective capacity to withstand and recover from disasters is significantly enhanced. Ultimately, social resilience is not merely the ability to endure shocks—it is the ability to recover with dignity. A humane disaster management system recognizes the agency, knowledge, and lived experiences of affected communities. It integrates cultural wisdom with modern capabilities, fosters trust between citizens and institutions, and ensures that every step of the disaster cycle reflects empathy, inclusion, and respect.

Immediate Community and Government Responses to the Crisis

Within ten days of the Ditwah disaster, the Sri Lankan government succeeded in rapidly mobilizing the security forces, key institutional structures, political leadership, and community organisations to confront the crisis. Given the scale and depth of the devastation, meeting the challenge and mitigating its effects seem to be a formidable task. The armed forces and government departments, supported by unaffected communities, provided exceptional assistance to meet the initial challenge. People in the South—often guided directly or indirectly by local political/community leadership—volunteered in large numbers, travelling to the hills to support recovery efforts. Much of the initial work of clearing debris and cleaning homes was carried out through community participation. Infrastructure repairs, particularly the restoration of roads, water supply, and electricity, were undertaken through coordinated action by relevant government agencies who worked tirelessly day and night. As a result, nearly 80 per cent of essential infrastructure was restored within ten days, with the exception of the severely damaged railway network, which requires longer-term reconstruction.

In the immediate aftermath, the government declared a nationwide state of emergency under the Public Security Ordinance, enabling the rapid deployment of resources across sectors. Through the Disaster Management Centre (DMC) and relevant ministries, authorities activated emergency operations: evacuation orders were issued in high-risk flood and landslide zones, shelters were established across the country, and search-and-rescue missions commenced immediately after landfall.

Concurrently, the government announced a comprehensive relief and recovery package. Affected households received allowances for cleaning and resettlement, support for temporary accommodation, and financial assistance for the repair or reconstruction of damaged homes. Immediate access to financial resources—including a Rs. 30 billion contingency allocation that did not require prior parliamentary approval—enabled swift implementation. The declaration of this extensive and unprecedented relief package played a key role in restoring hope and strengthening the self-confidence of affected communities.

Recognizing the magnitude of the crisis, the government established a special recovery fund that brings together public and private sector contributions to support long-term reconstruction, infrastructure repair, and livelihood restoration. Involving prominent private sector leaders—including those who are not aligned with the ruling administration—alongside government officials and key ministers is intended to build trust within the business community and reinforce transparency in the fund’s management. The substantial international assistance received and pledged reflects a renewed confidence among external partners in the government’s ability to manage funds transparently and ensure that aid reaches intended beneficiaries. Sri Lanka further collaborated closely with international and humanitarian agencies to scale up multi-sector support. Organizations such as the World Food Programme (WFP), International Organization for Migration (IOM), and World Health Organization (WHO) mobilized food, water, medical supplies, shelter materials, and rapid-response teams—often in coordination with government efforts—to reach displaced persons and vulnerable populations, particularly in remote and landslide-prone areas.

During this ten-day period, the President personally attended the district coordinating committee meetings in all cyclone- and flood-affected areas. These meetings brought together political leaders—both from the ruling party and the opposition—along with key administrative officers and representatives from the relevant line ministries to review disaster response, mitigation measures, and recovery needs. The manner in which the President raised issues, sought clarification, and directed action demonstrated a high level of preparation and a clear understanding of the scope and complexity of the damage. His engagement signaled a proactive and informed approach to crisis governance, contributing to more coordinated and timely interventions across affected districts.

Thus far, these measures largely pertain to confronting the immediate challenge and mitigating its impacts. Yet effective mitigation must ultimately lead into long-term recovery planning and strengthened preparedness for future climate-induced crises. Ditwah is not the first or the last. Climate change has altered the frequency, scale, and unpredictability of extreme weather events, making it clear that Sri Lanka must now learn to live with recurring climate hazards as a structural condition rather than an episodic disruption. This requires a sustained investment in resilient infrastructure, risk-sensitive development planning, and community-level adaptive capacity. In this sense, the response to Cyclone Ditwah should not only be understood as an emergency undertaking, but also as a critical moment to embed long-term climate resilience into national policy and institutional practice.

Lessons learned

The devastation wrought by Cyclone Ditwah has once again tested Sri Lanka’s institutional capacity, the NPP political leadership and peoples’ resilience. Since the 2004 Tsunami, the country has made significant progress in establishing organisational structures and policy frameworks for disaster management, making it a central domain of contemporary statecraft. Yet, the experience of Ditwah underscores the need for further strengthening in four key areas. First, given the multiplicity of ministries and agencies involved—from the Ministry of Disaster Management and the National Council for Disaster Management to the Disasters Management Center, the Meteorological Department and the National Disaster Relief Services Centre—clear mechanisms are essential to avoid overlap and ensure coherent, efficient action.

Second, disaster preparedness and response must harness the collective capacities of state institutions, NGOs, and community-based organisations, whose collaboration is indispensable for effective disaster risk governance. Third, the integration of traditional knowledge systems—rooted in long-standing practices of environmental stewardship and community resilience—should inform planning and implementation, complementing modern technology and institutional expertise. Finally, in a multi-ethnic, post-conflict society, sensitivity to ethno-political dynamics is imperative across all three phases of disaster management: preparedness, emergency response, and post-disaster recovery.

Ultimately, Cyclone Ditwah revealed both the vulnerabilities and strengths of the nation—demonstrating that while Sri Lanka’s systems were tested, its people were united in response, reaffirming the country’s capacity to confront adversity through collective resolve. The spontaneous networks of support that emerged in the cyclone’s aftermath demonstrated that unity is not merely an aspiration but an operational force in moments of crisis. In reaffirming the country’s capacity to confront adversity through collective resolve, the response to Ditwah offers a powerful reminder that the resilience of the people remains Sri Lanka’s most reliable foundation for future challenges.

by Prof. Gamini Keerawella ✍️

Continue Reading

Features

Rare Seahorse discovered in Sri Lankan waters sparks urgent conservation debate

Published

on

Sri Lankan marine researchers have formally documented the presence of the rare and Vulnerable Three-Spot Seahorse (Hippocampus trimaculatus) in Sri Lankan waters for the first time, an important milestone in the country’s marine biodiversity records.

The discovery was made through the examination of four dried specimens collected from fishermen operating off the southern coast near Madiha, nearly 150–200 km offshore. The evidence confirms that the island’s marine ecosystem hosts a greater diversity of seahorses than previously recognized.

Until now, only two species—Hippocampus kuda and Hippocampus spinosissimus—were scientifically confirmed in Sri Lanka, both largely linked to the northwestern lagoon systems. This discovery shifts that narrative southward.

Lead scientist Janamina Bandara emphasised the importance of the breakthrough, saying the identification not only verifies the species’ presence but also extends its known distribution range in the Indian Ocean.

He told The Island:”This is the first authentic record of Hippocampus trimaculatus from Sri Lankan waters. This species was assumed to occur here based on regional presence, but until now, we lacked verified scientific proof.”

Found in an Unexpected Habitat

While seahorses are typically associated with seagrass beds, shallow estuaries, or mangroves, the discovery revealed a surprising observation—these specimens were found attached to floating masses of marine debris.

Bandara described it as one of the most unusual natural behaviours documented in local marine fauna.

 “The specimens appear to have utilised drifting debris as habitat, which has not been explicitly recorded before,” he explained.

Photographs obtained from young field biologists show pieces of plastic waste, frayed fishing nets, fabric residues, and other floating refuse entangled into large drifting clusters.

Marine scientists say this phenomenon—informally referred to as “floating artificial reefs”—has been increasingly documented elsewhere in Asia and the Pacific. However, Sri Lanka has lacked records until now.

Bandara added that the drivers behind such habitat use remain unclear, raising questions about whether this behaviour reflects adaptation or desperation.

Specimens Documented, Sexed and Archived

Janamina Bandara

The research team collected four specimens—one male and three females—over two separate encounters, in March 2024 and June 2025. Measurements included head-to-snout ratios, ring counts, and coronet shape, all critical criteria in identifying seahorses.

 “All diagnostic features matched published descriptions, including distinct hook-shaped cheek and eye spines,” Bandara confirmed.

The specimens have since been deposited at the University of Ruhuna for long-term academic reference.

Illegal Trade Still Active

The finding has also shed light on the continuing illegal trade of dried seahorses in Sri Lanka—an industry long suspected, but seldom traced with scientific evidence.

The specimens originated from fishermen who admitted they sell dried seahorses to intermediaries and tourists. The team found that prices vary by size and buyer type.

“Smaller specimens sell for roughly Rs. 1,000 locally, while foreign buyers pay up to Rs. 5,000. Larger specimens fetch significantly more,” Bandara said.

Map. Known distribution of Hippocampus trimaculatus with the current study site indicated. Red dots: confirmed research-grade observations (n = 76) of the species from iNaturalist. Blue dot: study site location (Madiha coast, Southern Sri Lanka).

Many dried specimens are reportedly converted into gold-plated pendants, marketed under the claim of bringing luck and prosperity. In some tourist markets, dried seahorses are sold discreetly alongside shells and corals.

While enforcement exists, Bandara says it remains largely symbolic.

 “Raids happen, but are limited. Without awareness among fishermen and tour operators, the trade will continue,” she said.

Global Conservation Context

The Three-Spot Seahorse is listed as Vulnerable by the IUCN Red List and is protected under Appendix II of CITES, meaning its international trade requires permits. The species faces high risk from:

Bycatch in trawl fisheries

Rising demand from Asian traditional medicine markets

Rapid habitat decline due to marine pollution

Slow reproductive turnover

Seahorses exhibit monogamous pair bonding and unique male pregnancy, making their populations extremely fragile when harvested.

Sri Lanka, positioned at a central point in the Indian Ocean trade network, remains vulnerable to illegal wildlife trafficking routes.

Bandara emphasised that biodiversity verification has regulatory relevance.

 “Scientific records strengthen diplomatic and policy decisions. Without confirmed presence, enforcement remains weaker,” she explained.

Calls for Greater Action

Following the discovery, the research team is urging local authorities and NGOs to prioritise:

Awareness programmes for coastal communities

Monitoring of multi-day fishing vessels

Inclusion of seahorses in biodiversity assessments

Tourism-season enforcement in southern coastal markets

Bandara believes this new evidence allows Sri Lanka to become an active contributor to global seahorse conservation efforts.

A Turning Point for Marine Biodiversity Research

Beyond the immediate conservation implications, this finding marks one of the most scientifically significant marine records of recent years.

It suggests that Sri Lanka’s offshore ecosystems are both understudied and vulnerable to emerging human-driven pressures. Researchers now believe more undocumented marine species may inhabit local waters, awaiting formal identification.

 “This discovery is not only a scientific milestone but also a reminder that our oceans hold species that are disappearing faster than we are documenting them,” Bandara said.

As marine debris continues to accumulate and demand for illegal ornamental wildlife persists, researchers warn that scientific discovery alone will not ensure the species’ survival.

Bandara says what happens next will determine the fate of this newly confirmed marine icon.

“If we act now—educate, regulate and monitor—we stand a chance to protect these animals before they vanish unnoticed.”

By Ifham Nizam ✍️

Continue Reading

Features

Human-elephant conflict and housing needs of villagers

Published

on

Agbo is back home at Mahakanadarawa leading his harem (Source: YouTube).

During the recent Ditwah cyclone, elephants were seen floating in treeless floodplains that were once their forest habitats. On a good-weather day in 2017, near Kokilai, the Navy found a pair of elephants riding waves after a beach outing two kilometers offshore in the high seas. Divers guided them ashore after a 12-hour struggle.  Trains barrel through elephant herds regularly, decapitating half a dozen in one tank. A herd of elephants over 100 parades across a highway serenading motorists stuck in a kilometer-long traffic jam. Recently, adding insult to injury, a lone elephant was sitting deep inside a latrine pit behind a small house, and was dug out by a caterpillar tractor.  A speeding bus ran over and killed a baby elephant, and police shot dead the mother who stood crying over her baby’s body. The tusker named Sinharaja was still a baby when the Army pulled it out from an agri-well some years ago in Nuwarakalaviya.  He is now royalty tasked with carrying the sacred tooth relic at the Dalada Maligawa.   These extraordinary events, rubrics of a national drama, show that fates, ours and elephants, are inexorably linked.

Over 7,000 elephants and countless villagers in Sri Lanka are torn apart daily by myriads of unpleasant encounters. Our elephant population is multiplying alongside us, making these encounters even more remarkable.   As the government owns all elephants and writes laws for them, it also owns the product of these encounters. Since it has law books for the villagers, too, it cannot disregard the mess its protégé, this patrician in the wild, leaves on the villagers’ doorstep. Only the government can find a lasting solution to blunt the prickly edges of this national emergency, but not without contributions from the villagers.

George Orwell wrote in Animal Farm, “Some animals are more equal than others.” But the sentimentally charged public opinion about our cultural icon cannot outweigh the burden it placed on villagers living on the edges of elephant habitats.

As will be explained later, I propose a Gam Udawa-type house for each newly married couple who choose to remain within their village’s boundaries. If anyone edits this out as impractical, please come down from the ivory towers and visit a village bordering an elephant corridor to see for yourself the internecine damage elephants and villagers cause to each other.

There is no rich body of literature on the kinetics of village housing. But the volume of villagers’ experience is a safe guide to navigate it. I saw, over the span of three decades, how a major elephant corridor, one or two kilometers wide, adjoining my village above Mahakanadarawa reservoir, got swallowed up as villagers built (and still do) homes there.  Thus, one way to stop this is to contain the village where it is now. Halting home-building activities in elephants’ homes is a futuristic idea that the government has not tried. This experience also suggests that a study of the environmental impact of new village housing is in order.

Little parts that drive conflict

The government does not hear or see the little parts that drive the human-elephant conflict in the village. The only elephant problem it has is an 8 am to 5 pm thing, caged and tied to concrete stumps with steel chains at a compound in Dehiwala, minutes from Colombo’s urban universe. Together with Dehiwala, provincial compounds like Pinnawala, and a few national parks hold less than 1% of the Sri Lankan elephants, leaving the rest to roam around and harass villagers. Officials who have the power and know-how to resolve this tragedy do not feel it in real time. They do not live anywhere near where elephants live.

Indeed, it is a stretch one may suggest the government can find new space for the elephants like grandiose, unwieldy ideas like port-city-style landfills along the coast. However, we can work with existing landmasses more studiously using other methods. Driving elephants to the current Managed Elephant Ranges (MER) is not one of them. MER seems to lack sufficient food, as evidenced by the emaciated elephants we see in these ranges.    An elephant is a big animal and needs a bigger lunch setup.

HOW WE GOT HERE 

Until the mid-20th century, abundance of forest accommodated all villages and some more elephants; there was no reason to think villagers were taking elephants’ feeding ground any more than governments had any plans to reduce friction before it reached an unmanageable level.  Elephants’ feeding grounds occupied forest area about two kilometers wide in higher ground between two tank cascade systems, each independently sharing water from parallel watersheds.

Islands in the sea of forest

Villages in the North Central, Northwestern, Eastern, and eastern half of the Southern Province remained as islands in this sea of forest. Collective personality embedded in the village was that residents could hunt, harvest timber, and make small chena plots in these forests.  The concatenation of many such forest buffers formed elephant highways that were major feeding grounds. Everyone lived happily until the government’s neglect in addressing the population explosion of elephants and our own created the present predicament like a Class 4 wound.

A village community is a swarm, usually numbering around 100 individuals. Increasing membership in a swarm trigger some to move out to new locations. In a colony of bees, for example, an alternate queen bee will lead a part of the overpopulated colony out to set up a new community. Similarly, in the village, where two or three couples marry each year, and if the space for housing sites is limited, as is the case in old villages, a couple might emigrate to another village or town. The one or two with what biologists call the ‘group mind’ stay in the village, becoming the seeds that begin to spawn more warms, amplifying the elephant-human problem.

The new couple is looking to build a house closer to their larger family. But as space for potential housing is gone, the next option is to move beyond the traditional village boundary, where the one- to two-kilometer elephant feeding grounds begin. On these grounds, this family finds not only a spot for a house but also timber that had been the property of elephants and other wildlife since before the village’s genesis.  In a nutshell, this is how elephants began to lose their land.

Land grab

With this land grab, though isolated, friction over space ensued, leading to physical confrontation with elephants. The government’s inaction in mediating this problem is telling.

As years go by, this progression has led to the appearance of dozens of new home gardens, each slowly taking up at least a hectare of virgin forest. In a few decades, hundreds of such hectares will have been devoured by these progenitors entering the village marriage fraternity.

Meanwhile, the explosion of the rural population seems to influence the mechanics of elephants’ behavioral evolution.  Back then, elephants were shy.  I remember a herd disappearing into the woods in seconds after seeing a moving firebrand tipped with glowing embers. Aiming a flashlight made the herd disappear into the woods like blowing smoke into a beehive.  In contrast, now a wild elephant caparisoned with a dry crust of mud bath walks casually on a road, duly giving right of way to motorists, and stops by a lonely roadside tea kiosk.  He waits patiently, not for tea, but until the kiosk owner offers him a bunch of ripe bananas!

Today, elephants are so common and share our space more often, villagers assign lovely names to identify tuskers. In our childhood, we rarely saw a tusker because he owned a large swath forest, so his contact with us was minimal. Hence, the name tag was the least he needed.

HOUSING IN CITY AND VILLAGE

Whether people live in a crowded city with sprawling multistory housing compounds or in a village with two dozen homes under an irrigation tank, their universal human need is housing. In the city, with limited horizontal expansion, the housing idea must become improvisational. Thus, it grows vertically because it’s the only direction the cramped city can build.  Having no such problem, after the old gammedda ran out of space, villagers moved horizontally to new tracts of forest beyond the village borders. Missing in the discussions on the loaded thesis of elephant-human conflict is this premise – the housing need of newly married villagers, the overarching subtext of this problem, not seen by anyone outside of the village. This married couple clear a track of forest, marking the beginning of the gradual encroachment of the village into an existing elephant range.

When it comes to housing, villagers in elephant-roaming areas are left to fend for themselves. Overcrowding in villages had not received the government’s attention because it never put a premium on housing in a village.

Both parties are victims here. Any steps to help them have become untenable due to poor management (of the problem) and the uncontrolled population upsurge of the parties. This drama is what the successive governments have missed seeing. Although the government and private sector have been generous with housing issues in the city, not extending the same kindness to villagers is why they are in this loveless embrace with elephants.

Meanwhile, beginning in the mid-20th century, the city has adapted to meet its residents’ housing needs.  The scale ranged from clusters of one-room homes like UC Quarters in Urban Council jurisdictions, to modest multi-unit housing compounds, ‘flats,’ like the eponymous ones at Narahenpita, Maligawatta in Colombo. Over the past couple of decades, towering megastructures catering to the new affluent residents have further diversified the city’s housing options.

The elephants are wanderers and have all the land to move around. But villagers are no longer the itinerant bands they once were in their distant past.  Due to their proclivity to acquire acreage from freely growing forests, they become fixed targets for elephants. But don’t accuse the villagers of being xenophobic towards elephants.  We see they never show schadenfreude – enjoying an elephant’s misfortune, while it struggles to climb out from an agri-well or a canal. Instead, standing on the edge, they speak kind, encouraging words to the traumatized animal. Some even throw banana stumps at him to eat.

FAILED HOUSING PROJECTS (WIYAPARA) IN THE 1980S

Often, the government itself is the culprit of expanding the village into elephant corridors by introducing new housing projects. Such housing schemes were called wiyapara gewal (project homes). It turned out to be a failed government idea.

   Near my village, in the 1980s, the government marked off housing plots along a cart road that ran through a 2-kilometer elephant corridor which began from the end of our tank bund.  This stretch separated us from a series of neighbouring villages in the upper reaches of the Mahakanadarawa reservoir, built in 1959. Until then, elephants freely moved between Padaviya, Nachchaduwa, and Kalawewa tanks, using forest corridors between tank cascade systems, including the above, rarely entering villages.

After Mahakanadarawa gobbled up an extensive virgin forest area, elephants circumventing it on the way to Kalawewa stumbled upon a surprise: a society of homes was sitting on the above forest corridor that had always belonged to them.  Villagers cut down the verdant forest and started home gardens in their place under the aegis of the aforesaid wiyapara project. It bridged the neighboring village into one extended community. One family even fenced off the kamatha-sised water hole that elephants enjoyed on the rocky outcrop called Wannamgala and enclosed it in the new garden lot with the sign “balla hapai” (dog will bite)!

The Member of Parliament for the area was behind this project, with a piecemeal aid package worth about Rs. 25000 to each land recipient.  With that kind of economic magic wand, a half dozen villagers yielded, and now this row of houses bisects what was an elephant highway, sending elephants’ equanimity to coexist with human settlers downhill. Historical blunders like this tell us to reconstruct untested housing concepts to fit the present.

FEW PROPOSALS

Discouraging villagers from spilling beyond village boundaries to build new homes must be a priority in any plan designed to address elephant management issues.  It is a way to stop the slow oxidation of elephant corridors where newlyweds continue to stake out claims for home sites.

It is unfair to deny villagers the opportunity to own a piece of their own home garden. On its part, the government can help by creating employment or home-garden opportunities by introducing them to garden crop methods and small-scale industries, which will provide them with a meaningful livelihood and a reason to stay within the village’s borders.

The government must also devise the same plan it uses to address overcrowding in city swarms, by building small irrigation colony-type houses within the village situated for newlyweds in villages on the borders of elephant habitats.  New families will appreciate this idea that their government is giving them a hand with a small house within the village limits.

My proposal may sound like a fictive reverie. But math speaks for itself. Consider this: conceivably, if we can prevent that one newlywed couple from carving out its space in a forest tract used by elephants, after a few decades, we would have saved dozens of hectares for elephants by preventing couples from moving there to build new dwellings.

I ask the government not to think of the pink elephant – the cost – in considering this project.   If it cannot build the house for free, recover a partial cost from new owners in easy installments based on their verified income.

There are many private and public tracts within a village that remain fallow or undeveloped. The government can offer to purchase these to build new homes.  What happened to Alfred House Gardens in Colombo – 3 over a century ago gives us ideas on how to apply the summary of that history to a village where, generally, real estate behaves similarly.

Towards the end of the 19th century, the owners of Alfred House Gardens partitioned the opulent estate, endowing it to the city for a greater purpose. My premise here is that a villager with his own Alfred House estate in the village, parts of which remain fallow, may wish to place a corner of it on the market for ready cash.  Suppose the government offers an enticing price. In that case, I have no doubt the owner might consider it. Haven’t we seen this in newly partitioned large coconut plantations elsewhere?

The government will then build a small house here and bestow it on the newly married couple.  This is one couple that will not pose a threat to elephants’ right of way.  If we can push this simple idea to fruition, in a decade or two, encroachment into elephants’ roaming lanes can be reduced considerably around this village.  A fitting paradoxical allegory for this is an African proverb that says: “The way to eat an elephant is to take one bite at a time.”

Furthermore, the government may amend the President’s Fund or create an Elephant Fund to provide small housing loans specifically to newlywed couples in such villages. It must suspend the irrational, sneaky and flagrant absurdity of tax-exempted vehicle imports, now allowed to certain privileged government officials.  Tax this exclusive club and use the money for this program.  Each new car landing on Sri Lankan soil will pollute the environment and be one more headache for elephants feeding at Minneriya tank.

To identify which villages are likely to encroach on elephant corridors, the wildlife department must survey and designate their boundaries. This step is every bit as essential as declaring stretches of forest as elephant corridors.   Also, an accelerated tree-planting program to rehabilitate deforested areas on the edges of elephant corridors must be a government priority.

The government must not reward large farming interests and the solar power industries by allowing them to take up elephant habitats. A papaya plantation has alternatives that an elephant family does not.

Finally, failure to resolve this problem will itch our nation’s conscience and shame us deeper. The few patches of forest we can keep uncleared are the ones that tell us just how many more hectares of them refuse to be cleared.

Lokubanda Tillakaratne chronicled life in a village in Gammadde Ninnadaya, and a defunct traditional judicial system practiced in Nuwarakalaviya villages in Rata Sabhawa (Sarasavi Books).

By Lokubanda Tillakaratne ✍️

Continue Reading

Trending