•  What the 2026 death toll actually shows, how elephants are dying, and why coexistence must replace crisis management

One-hundred-and-eighty elephant deaths reported by 12 August. 

The number has travelled quickly because it is almost impossible to forget: more than 800 elephants dead in Sri Lanka this year (2026). It sounds like the kind of statistic that should stop a country in its tracks.

It is also inaccurate.

Sri Lanka has not recorded 800 elephant deaths in 2026. The latest publicly reported Department of Wildlife Conservation (DWC) figure is 180 wild elephant deaths as of 12 August. The phrase “over 800” refers to two earlier years combined. Sri Lanka recorded 439 elephant deaths in 2022 and 488 in 2023 - a total of 927.

Correcting the number matters. Conservation cannot be built on an inflated claim. But the correction must not become an excuse for comfort. One-hundred-and-eighty deaths in 224 days is still roughly one elephant every 30 hours. During the same period, 56 people were reportedly killed in elephant encounters - one human life every four days.

This is not a story in which only elephants suffer, or in which rural families should be treated as villains. It is the story of two vulnerable populations being pushed into the same shrinking, poorly governed landscape and then left to absorb the consequences.

What the official numbers say

Of the 180 elephant deaths reported by last month (August), the DWC attributed 75 directly to human activity. The published breakdown is brutally specific: 30 elephants were shot, 23 were electrocuted by illegal electrical connections or fences, 19 were killed by explosive bait known as hakka patas, and three were poisoned.

That means that about 42 per cent of the reported elephant deaths had a clearly identified, direct human cause. It does not mean that all 180 elephants were killed by people. The August announcement did not provide a complete cause-by-cause account of the remaining 105 deaths. Those cases can include natural causes, illness, injuries, accidents and deaths still under investigation. Accuracy requires keeping that distinction visible.

Earlier DWC data also show the range of hazards. By 30 April, when the cumulative toll stood at 108, officials had recorded five deaths from train collisions in addition to shootings, electrocutions and explosive bait. Other reporting during the year has identified poisoning, drowning and falls into unprotected cultivation wells among the recurring causes. These are not all deliberate killings, but many are preventable.

This year’s data snapshot: Wild elephant deaths by 12 August – 180; Deaths attributed directly to human activity – 75; Shot – 30; Electrocuted – 23; Killed by hakka patas explosive bait – 19; Poisoned – three; and Human deaths in elephant encounters – 56.

The methods reveal more than a death toll. A gunshot usually signals retaliation, fear or illegal hunting. An electrified wire points to unsafe crop protection and weak enforcement. A hakka patas device - an explosive hidden in food that detonates inside an animal's mouth - reflects a level of desperation and cruelty that no society should normalise. Poisoning is deliberate. Train collisions, exposed wells and drainage structures expose failures of infrastructure design.

The crisis is therefore not one problem. It is a chain of land-use decisions, livelihood insecurity, weak deterrence, delayed policy, under-resourced enforcement and infrastructure that ignores elephant movement.

The map explains the conflict

Sri Lanka's elephants do not live neatly inside national parks. A peer-reviewed countrywide survey found that elephants across 59.9% of the island, while people were resident in 69.4% of the elephant range. It also estimated that the elephant range had contracted by 16.1% since 1960.

Those figures dismantle the comforting idea that conflict can be solved simply by pushing elephants “back into the forest.” Much of the land that elephants use is already a shared landscape of villages, farms, roads, railways, reservoirs, scrub forest and development sites. Elephants remember seasonal routes to water and food. A new settlement, plantation, highway or poorly placed fence does not erase that memory; it converts a movement route into a conflict zone.

The Government itself acknowledged the scale in June. A Cabinet of Ministers decision stated that the human-elephant conflict (HEC) now affects 19 of the 25 Districts and 134 of its 341 Divisional Secretariat Divisions. It linked the escalation to population growth, expanding development, deforestation, habitat destruction and the loss of traditional migration routes. The same decision cited a 2024 national estimate of about 7,450 wild elephants.

The distribution of the conflict is important because it exposes the limits of park-centred conservation. Protecting parks remains essential, but, parks alone cannot protect a species that spends much of its life outside their boundaries. Nor can long fences built along administrative boundaries succeed if elephants are stranded on the human side.

A policy cycle that keeps restarting

Sri Lanka is not short of plans, committees or fencing projects. It is short of sustained implementation, measurement and public accountability.

Yet, in June, Cabinet approved the preparation of another National Plan because conditions had changed. In August, a monitoring committee met again to formulate it using updated information on migration, forest cover and agricultural expansion.

The lesson is not that electric fences never work. It is that location, ownership and maintenance decide whether they work. A well-maintained seasonal fence around an active field can protect crops. A community fence around a village can protect homes. A long boundary fence without local ownership may fail, be breached, fall into disrepair or trap elephants in inhabited land.

This is not a war between farmers and elephants

It is easy for urban commentary to condemn a farmer who protects a harvest. It is harder to spend weeks guarding a field at night, knowing that a single raid can erase a season's income. Families in conflict areas face crop loss, damaged homes, injury, sleeplessness and death. Compensation can be slow, difficult to claim or too small to rebuild a life.

Hardship cannot justify shooting, poisoning, electrocution or explosive bait. The answer is to reduce the conditions that produce retaliation and to make lethal acts more difficult, more detectable and more consistently punished.

Language matters here. When every elephant death is described as a killing, the public receives a distorted picture. When every human death is described as an unavoidable “elephant attack”, the policy failures that placed a home or road across an active movement route disappear. We need precise records and an honest account of responsibility.

What SL should do now?

1. Publish a national mortality dashboard

The DWC should release a frequently updated, District-level record of every elephant and human death: date, Global Positioning System location, age and sex where known, confirmed or suspected cause, post-mortem status, enforcement action and case outcome. “Unknown” should remain unknown until evidence changes it. 

2. Protect villages and fields, not arbitrary boundaries

Community-managed village fences and seasonal agricultural fences should be expanded where local conditions support them. Responsibility and funding for inspection, vegetation clearing, batteries and repairs must be assigned before construction. New fences should be mapped against elephant tracking and field data so that they protect people without blocking essential movement or trapping animals among settlements.

3. Treat illegal electrocution as a preventable enforcement failure

DWC officers, the Police and Electricity authorities need joint rapid-response teams in high-risk Districts. Illegal mains-powered fences should be removed immediately, with inspections focused on known mortality clusters. 

4. Disrupt the supply and use of hakka patas

These devices should be addressed as an illegal explosive supply chain, not only as isolated wildlife incidents. Veterinary response teams also need the capacity to reach injured animals quickly, although many victims cannot be saved once the device explodes in the mouth.

5. Secure corridors before approving development

Strategic environmental assessment must come before land allocation, not after the conflict begins. Corridors need enforceable boundaries, ecological restoration and negotiated solutions for people already living or farming within them.

6. Make railways, wells and canals safer

The Railways Department began introducing thermal cameras on locomotives in 2026. Technology should be paired with enforceable speed restrictions at verified hotspots, vegetation management for clear sightlines, real-time warnings to drivers and a shared incident database between the Railways and the DWC. Unprotected cultivation wells should receive parapets, covers or escape ramps, while canals and reservoirs in the elephant range need safe exit points.

7. Make rural safety and compensation part of conservation

Rapid warnings, trained village response teams, safer grain stores, lighting that does not provoke confrontation, crop insurance and prompt compensation reduce the pressure to retaliate.

8. Fund outcomes, not kms and announcements

Success should be measured in fewer deaths, fewer crop raids, shorter compensation times and functioning.

Coexistence is sometimes dismissed as an idealistic word. In Sri Lanka, it is the practical conclusion of the map. Elephants and people already share the land. The national choice is whether that shared land will be governed with evidence, fairness and foresight - or counted in bodies after every preventable failure.

The writer is a climate specialist with a background in environmental management, and a sustainability and development professional

----------------------

The views and opinions expressed in this column are those of the author, and do not necessarily reflect those of this publication