Friday, June 29, 2007

Deforestation in India

In the Introduction to the book Battles Over Nature, the editors Vasant Saberwal and Mahesh Rangarajan remark:

“…..That there has been a steady decline in the extent of undisturbed forest cover in the country over the past two hundred years is widely accepted. The causes for this decline are less widely accepted, although there is general agreement on some factors – the British demand for timber to meet demands imposed first by the British Navy and then by the building of the Indian railways; the demands from the two world wars; the post-Independence provisioning of forest resources to industry at highly subsidised rates. On these counts there is little debate…….” (emphasis added)

Note the term undisturbed forest cover. What exactly does it mean? The term is not defined in the book and to the best of my knowledge is not a standard forestry term capable of easy or unambiguous interpretation. Disturbance probably relates to human interference or use. This can cover a wide range of activities – from clear felling to a walk by a human being. Saying undisturbed forest cover has declined is so non-controversial and irrefutable that it is frankly quite useless. Who could disagree with such a claim?

Surely the use of the term is not a slip but deliberate. Why would the authors do so? For if the word undisturbed is removed and the same causes are ascribed the proposition becomes interesting and can be debated. Why are the authors playing it so safe?

It is very likely that few people would disagree that there has been extensive deforestation in the last 200 years and it was not merely a decline in undisturbed forests. If so how do we explain it?

The authors are content to state the reasons and the topic is not discussed in the book. A useful, well-regarded and oft-quoted source for such views is the book The Use and Abuse of Nature by Madhav Gadgil (MG) and Ramachandra Guha (RG). Additional references can be found in these two books.

It would be fair to claim MG and RG largely agree with the reasons listed in the opening paragraph (in fact they may have provided the basis for the claim cited above).

In their book MG and RG discuss the reasons at some length. Take the British era first. They trace “The Early Onslaught on Forests” (pg. 118) on the British demand for timber for shipbuilding. They say that colonial supplies including from India accounted for a large part of the merchant fleet built up by the British. However, this tells us precisely nothing about how this was affecting Indian forests. For that requires an estimate of timber extracted and area affected by such operations. There is nothing on that. And though a large proportion of wood used by shipbuilders may have come from colonies (as claimed by the authors) it implies nothing about how big or small an impact it made on the forests of the colonies.

They then discuss the felling of trees to make railway sleepers. They also present some data for the same. MG and RG also quote several accounts on the destruction that such extraction caused though those read as examples of bad forestry practices more than anything else (pgs.120-122).

Data on felling of trees for obtaining sleepers is presented but the numbers are too few and far between. There are a handful of figures from Madras and Yamuna Valley (pages 121-122). However, there are no figures for multiple years for these or other regions and none whatsoever for the country as a whole.

How then can one form an opinion about what was happening in India in the second half of the 19th century. One needs at a minimum time-series for a reasonably long period of time aggregated over the country. The British kept meticulous forest records and it is surprise that that such data is not analysed or presented though admittedly it would be a tedious exercise to collate and put it together for the whole of India.

In addition to perusing forest records the approximate requirements for railway sleepers can be worked out by an alternative method. Start from the length of railway tracks and using the number of sleepers required per km for new tracks and for replacement estimate the total number of sleepers required per annum. Next work out the number of trees required based on average yield of sleepers per tree. And finally using average density of sleeper species per hectare work out the likely area harvested.

The authors do mention some of these factors but don’t take the steps to build such estimates.

Let us do a small exercise using some data used in the book. Footnote no. 4 on pg. 122 mentions an annual requirement of 1 million sleepers (in the 1870s) and on page 121 one can derive an estimate of 7 sleepers per tree (250,000 sleepers from 35,000 trees). So a million sleepers would require 1,50,000 trees. Assume 5 trees were obtained from each hectare and that the rest of the area got completely denuded. We can thus estimate 30,000 hectares of deforestation per year. This simple and limited exercise indicates the meagre loss due to sleepers compared to the total forest area in the country which would have been at least 80-100 million hectares.

It is also odd that demand for wooden sleepers as a reason for deforestation drops off the list in early 20th century and also in independent India. Surely the Railways needed millions of them. Non-wood sleepers are a recent innovation.

For independent India they mention paper mills as a case in point. All that the authors provide is an account of paper mils going farther and farther away from their plant sites to obtain raw material. That, however, doesn’t help us understand the extent of India-wide deforestation. There is virtually no data. Interestingly a table on page 196 (reproduced below) gives a completely different impression.

The Loss of Forest Area for Various Purposes between 1951 and 1976

Purpose                                        Area(thousand ha)
River valley projects                 479.1
Agricultural purposes                2506.9
Construction of Roads                57.1
Establishment of industries       127.2
Miscellaneous purposes            965.4

However, there is something which is even more galling. This is the lack of discussion of regeneration. Even if trees were cut it doesn’t automatically imply permanent loss of forest-cover as trees grow back and the forest cover is restored. Even barren land is rapidly colonized. Hence accounts of deforestation necessarily need to account for what prevented regeneration.

It is hence impossible to accept the authors’ contention and the generally accepted notion about the reasons for deforestation in India. The evidence is anecdotal at best and the data is scanty and selective.

In fact a proper analysis of demand for wood including timber ought to start with pre-British times and then compare rates of deforestation over time. Was there a break with the past and did the trend change or accelerate and when. Finally other possible reasons ought to be considered. For example construction, furniture and fuelwood to name only three sources also contributed to demand for wood. Surely these factors need to be considered in any analysis.

It is also important to distinguish two tendencies. One, that land got diverted to other uses – agriculture, roads, towns, dams, factories, etc. - so there was a permanent loss of forest land. Two, that forests were degraded - loss of forest cover.

Clearly there is insufficient evidence to accept the commonly accepted hypothesis regarding deforestation. The industrial demand for wood (both under the British and independent India) may or may not be responsible or causing or accelerating deforestation. There could have been other forces at play.

That there is little debate shouldn’t imply that the matter is settled. It may well be that not enough work has been done and not all factors examined closely. A comprehensive account of deforestation and its causes still remains to be provided.