by Chandra Arulpragasam

In 1958 I went to a lecture on coconut cultivation, because I knew nothing of the subject. The lecturer, a well-known coconut planter, started his talk with the platitude: ‘The duty of a coconut planter is to plant coconut, on coconut land’. But this set me thinking. First, who gave him the duty to plant coconuts? From his own point of view, he should be planting the crop that would give him the greatest returns, while from the country’s point of view he should be planting the crops that would provide the greatest return in terms of income, foreign exchange, employment and sustainability. Secondly, who decided that these were ‘coconut lands’? Was this not a terminology (‘tea lands’, rubber lands’, ‘coconut lands’) inherited from the British, who grew these crops because they could be grown on a plantation-scale for export?

A couple of years later (in 1960) when I headed the agriculture sector in the Department of National Planning, and again when I came with the ILO World Employment Mission to Sri Lanka in 1974, I had an opportunity to revisit these questions. If coconut was to be a mono-crop, it was important that it should meet the above criteria of greater employment, greater income and greater foreign exchange earnings compared to other crops. Coconut brings much lower financial returns than tea or rubber. As for employment, figures of 1960 showed that while one acre of tea employed 1.1 persons per acre per year, and one acre of rubber employed 0.4 persons per acre, one acre of coconut employed only 0.1 persons per acre per year. That is, only one worker was employed for every 10 acres of coconut, which was four times less than that employed in rubber and 10 times less than that employed in tea. This meant that the so called ‘coconut lands’ were under-utilizing the land not only in terms of income but also in terms of employment.

In physical terms, assuming that the coconut trees are planted in the usual spacing of 8m x 8m apart (which is accepted by the CRI as standard) and that their root system spreads only two metres around each tree (CRI stsndard), this would still leave 78 per cent of the land untouched and unutilized – the best lands in the ‘coconut triangle’.

This brings us back to the previous question. Why should these lands be called ‘coconut lands’? Is coconut the best or only crop that can be grown on them? Undoubtedly these lands are well suited for coconut, while coconuts are much in demand by our people. Not for nothing has the coconut tree been called ‘the tree of life’. But is it wise to relegate so much of our fertile lands to a relatively low-paying mono-crop? The British probably originated the nomenclature of ‘coconut lands’ when they grew coconut as a monocrop on a plantation scale, thus making it a land-extensive and labour-extensive crop, as opposed to the land-intensive and labour-intensive crops dictated by our factor endowments. Hence, this article is not against the planting of coconut: it is only against the planting of coconut as a monocrop on lands capable of yielding much more by way of intercropping.

The system of management of ‘coconut lands’ in the period 1960-1980 speaks for itself. Whereas tea and rubber estates were managed by resident estate superintendents or managers, coconut estates were ‘looked after’ by a ‘conductor’ or by a ‘watcher’, armed only with a torch and gun. The latter showed that the focus was on preventing the theft of coconuts, rather than on increasing yields or output. This locked large extents of these ‘coconut lands’ in a cycle of low expectations, low investment, low-level management, low income and low employment.

The Coconut Research Institute (CRI) in 1974 insisted that the optimum stand of coconut was 64 trees per acre, with an adequate distance (8 metres) between the individual trees and the coconut rows. It argued, on the one hand, that the growth of the intercrop would be stunted by the shade of the coconut, while insisting on the other, that the intercrop would deprive the coconuts of needed soil nutrients. After long discussions, the CRI experts ultimately agreed to the following propositions made by me in 1974.

First, it would be technically possible to inter-plant other crops during the first five years of replanting/new planting coconut without any adverse effects, since the coconut palms would be too small to block out the sunshine from the intercrop. This in itself was a big breakthrough, since an average of 9,300 acres was replanted or newly planted to coconut each year in Sri Lanka. Since intercropping would be possible for the first five years, the total acreage available for intercropping in the newly planted/replanted acreage in any particular year would be 46,500 acres (9300 acres x 5 years).

From this total should be deducted the 22 per cent of land that is actually occupied by the newly planted coconut, which would leave a net acreage of 36,000 acres for planting other crops. For purposes of comparison, this annually available acreage is more than double the extent of land opened up under land development/colonization schemes in each year, prior to the Mahaweli Scheme.

Secondly, the CRI ultimately agreed that in older stands of coconut (more than 25 years old), the trees would have grown so tall that they would not block out the sun from an inter-planted crop. It further now agrees that intercropping is possible without detriment to the coconut or its yields for 35 years of the trees’ 55 years of productive life. It is a pity that it has taken about 30 years for technical thinking to reach this conclusion!

But, thirdly, it was necessary to push the thinking even further. I argued that wider spacing between the coconut rows would result in less shade between the rows, thus enabling intercropping. The CRI in 1974 initially objected to this on the grounds that it would reduce the total number of trees per acre. But they ultimately agreed to my suggestion that if we increased the space between the rows but planted closer along the rows, the number of 64 trees per acre could still be attained, without any decrease in total production. Such further-apart spacing of coconut rows is now (40 years later) actually practised in Kerala and the Philippines, combined with intercropping. However, in Sri Lanka, although this was technically accepted in 1974, there has been little action along these lines by the Coconut Development Authority.

There remained the question of what could be grown as an inter-crop. When I travelled for FAO in Asia in the 1970s, I found pineapple, bananas, sisal, maize and manioc already inter-planted with coconut in the Philippines, and even cocoa under coconut in Indonesia, while livestock was common in most countries. Thus Sri Lanka lagged behind other South East Asian countries in this regard not only in the 1970s, but even so today.

Despite the government’s neglect, private planters in Sri Lanka have recently been adopting intercropping at an increasing pace. According to a survey done by the Coconut Research Institute in 2006, cashew was the most popular intercrop in the Dry Zone, while pineapple, betel and pepper were most popular in the Intermediate Zone. Tea, cinnamon and ginger were most popular in the Wet Zone, while bananas and livestock were common in all regions. Agro-forestry using tree crops (such as glyricidia) has also been recently recommended as a means of providing fodder for livestock, wood for fuel, biomass for fertilizer, control of erosion and soil moisture retention.

Obviously the possibilities of intercropping would be more limited in drier parts of the country with poorer soils. The inter-planting of cashew trees (pruned low) between the rows of coconut has now been adopted in the drier areas. I had also suggested (in the Short Term Implementation Programme of 1961) that groundwater was likely to be available at fairly shallow levels in the coastal areas north of Puttalam, which could be pumped up for higher value crops. I had also suggested the possibility of using windmills for such irrigation, which could be powered by the steady winds that blow during the dry season in these areas.

In 1994, I was able to revisit this question of inter-cropping under coconut in the drier areas. A women’s micro-credit in the dry north of the Puttalam District had used its loan to purchase a pump to irrigate an inter-crop on land newly planted to coconut. The women found groundwater at a depth of only four feet, which they pumped to irrigate chillie plants cultivated between the newly planted coconut rows. Their net return was Rs. 30,000 per acre within a four month period in 1994, which was more than treble the return from the adjoining coconut land for the whole year. Meanwhile, the fertilizer and water that they used for the intercrop were found to benefit the newly planted coconut too, in a win-win synergy. In the long run, the possibility of drip irrigation for coconut needs also to be considered. Such irrigation is needed only at the height of the dry season (cheap systems are now available) in order to reduce stress and increase yields.

To sum up, the Coconut Research Institute has now agreed to the following propositions that I proposed in 1961 and reiterated in 1974 (ref. ILO World Employment Mission, 1974).

· Inter-cropping between newly planted or replanted coconut can be done without prejudice to the newly planted coconut palms for the first five to six years of their life.

· In new plantings, the coconut rows could be planted farther apart, but with more trees per row, such that the total number of trees per acre will not be reduced. This would enable an inter-cop between the rows.

· Inter-planting among older coconut stands of over 25 years can be undertaken without detriment to the coconut trees or to the intercrop.

· Such intercropping can be done even in the drier regions using intercrops suited to the drier conditions, while irrigation would provide an added bonus.

· The yields of coconut actually increase because of the fertilizer and water used in the intercrop.

· There are other advantages of intercropping, such as providing biomass for fertilizer, increasing soil moisture and reducing erosion.

· The inter-crop (depending on the crop) is capable of yielding more than double the value of all the coconuts that could be produced from the same land.

Despite intercropping being both feasible and profitable, it was reported as late as 2007 that ‘in Sri Lanka, most of the coconut holdings are maintained as monocultures’ (Gunathilake, 2007). The question is why intercropping has not been more widely adopted when its feasibility and desirability were highlighted as early as 1974. The answers, in the opinion of the writer, are mainly structural and institutional.

The advantages of intercropping arise from its more intensive use of land and labour, with resultant higher returns per acre. However, the pattern of absentee ownership and management of larger estates raises the problem of supervising the casual, non-resident labourers needed for intercropping. Faced with this question, one of my estate-owner friends exploded: ‘Are you mad? The fellows (the labourers) will steal my coconuts’! Thus, although intercropping is recognized as feasible and profitable, the prevailing agrarian structure (with large holdings and absentee landlords not prepared to accept outside labour) seems to be the major factor inhibiting the wider adoption of inter-cropping on larger estates. Such estates (over 20 acres) occupied 18 per cent of the total area under coconut in 2002 (Agricultural Census of 2002).

Coconut, however, is mainly a smallholder crop in Sri Lanka, with 80 per cent of all ‘coconut lands’, covering almost 800,000 acres being made up of small holdings; 54 per cent of these are less than three acres in extent. Inter-cropping is gaining ground in this area, using mainly family labour. Although figures of comparative coconut yields between large and small coconut farms are not available for Sri Lanka, it is very likely that the coconut yields are higher in these small holdings compared to larger holdings, as proved in other countries. More importantly, the total value of agricultural production per acre in such small holdings is likely to be much higher than that in the large, well-managed coconut estates.

This is because the coconut smallholder invests more labour per unit of land to intensify and diversify his production by intercropping, in order to maximize his income. Most small coconut holdings are likely to include a papaya, banana or lime tree, some betel or pepper vines, some home-grown vegetables and some livestock. In fact, the small holder actually attains this higher level of total productivity per acre only by treating his land as much more than a ‘coconut land’.

Fortunately in more recent times, individual coconut planters in Sri Lanka have started to inter-crop on their own initiative, with encouraging results. The Coconut Research Institute has also helped by useful research into types of crops and land practices for intercropping. There has also been more forward-looking research and development abroad, in terms of ‘coconut based farming systems’ (CBFS) – a concept which is gaining ground in South India (Kerala) and some other South East Asian countries.

The purpose should not be merely to increase coconut yields, but to maximize the total productivity of these lands on a sustainable basis. This can best be achieved by a more holistic approach which seeks to develop the farming system as a whole, with each component synergistically supporting the other. While coconut would provide the pillars of such a farming system, inter-cropping would enhance its total productivity and ecological sustainability. Since coconut would still be the foundation of such a system, perhaps we could even be forgiven for referring to these lands affectionately as ‘coconut lands’!

(The writer who was a member 0f the old Ceylon Civil Service thereafter had a long career with FAO)