Colonialism, forced labor and mortality in Java

Pim de Zwart, Daniel Gallardo Albarrán, Auke Rijpma. Show Author details

How did colonial extractive institutions affect the development trajectories of countries over the long term? Whereas most of the literature on the economic impact of such institutions, much less is known about local health conditions and mortality. Our new research addresses this knowledge gap by examining the mortality consequences of the Dutch Cultivation System in nineteenth-century Java with new detailed data for many regions within the island.

The Cultivation System was one of the most extractive forced labor regimes ever to exist. In this system, Javanese peasants were forced to devote a substantial share of their land and labor to the production of various cash crops, such as sugar, coffee, tea, tobacco and indigo. Despite the fact that these commodities were highly valued and therefore sold at very profitable rates in international markets, the local peasantry in Java received very little in return for their labor and land devoted to the cultivation of these crops. During most of the 19^th century, the system was highly effective at increasing the output of cash crops from the island. For instance, sugar production increased from 6,700 tons in 1831 to over 100,000 tons per annum in the late 1850s, while coffee production rose from 20,000 tons in 1829 to almost 55,000 in the late 1850s. Unsurprisingly, the system was also extraordinarily profitable for the Dutch government. At its height in the 1850s, net transfers from the East Indies – known as the batig slot – were almost 4 percent of Dutch Gross Domestic Product and accounted for over 50 percent of total government revenue. 

The system deeply intervened in local society as it involved over 70 percent of all Javanese households at its height. Both European and local elites were incentivized to push for higher production by the peasants via the so-called kultuurprocenten, or cultivation percentages, which represented a share of the total sale proceeds of these crops in the Netherlands. Among the higher echelons of the colonial administration – both Dutch and indigenous – the income from these proceeds was excessive and their wealth became legendary. For lower indigenous heads, such as village chiefs, these proceeds obviously were lower, but in terms of their relative importance in total income they may have even been more important as they received little income besides it. Via these kultuurprocenten, market prices exerted an influence on the demands placed on the local peasantry. Liability for work in the cultivation system was related to access to land as in principle labor duties were requested from household heads of landholding families. Landholding could mean both private ownership of plot, but also rights to cultivate a share of the communally owned village land. The landowning household head, could send a replacement, however. Our sources suggest that most work was done by young men that were slightly poorer than most in society. The system was gradually abolished after 1870, when the Dutch East Indies were opened up for private enterprise, and forced cultivation for the government slowly declined. 

Labor conditions and health

How was this system related to health conditions and mortality rates? Much of the labor in the Cultivation System had to be done on plantations, as shown in the image at the beginning of this article. There were workers who spent the night on those plantations, but also those who travelled each day between their villages and the fields. The plantations themselves, described as unhygienic sites by contemporaries, formed particular hotbeds of disease, while the movement of workers contributed to illness spread throughout a region. Colonial investigations from the early 20^th century confirms this link between plantations and disease diffusion, although this information was not well received by everyone. As a result, one of the chief colonial medical officers in the middle of the nineteenth century, Willem Bosch, risked his career by alluding to the role of work in the Cultivation System in the spread of a fever epidemic throughout central Java in the 1840s. Additional qualitative evidence suggests workers on government plantations were often malnourished, which impeded the proper functioning of their immune systems, making them more susceptible to disease. Overall, Java suffered from high mortality rates as we can see in the figure below (grey areas show known high-mortality events: famine and disease outbreaks). 

To see whether labor in the Cultivation System was indeed related to higher mortality rates, we compiled a dataset for 16 residencies (that is the largest administrative regional unit in the Dutch East Indies) for the years 1834 and 1879; a period that covers the rise and decline of the Cultivation System. The dataset includes figures on the number of workers called for Cultivation System labor each year in the different residencies, as well as the overall size of population, mortality and birth rates, and finally, other indicators of local economic and social development that allow us to isolate the effect of forced labor on mortality. Some of the information we gathered is shown in the map below, where we can see that, roughly, places with high forced labor demands were particularly deadly. Through our quantitative analysis, we indeed find that, ceteris paribus, a higher number of workers called up for labor in the Cultivation System is related to a higher mortality rate. Our estimates suggests that in 1840, for example, for every additional 1,000 workers engaged in the Cultivation System, total deaths increased by 30. Over the long-run, our calculations show that the abolishment of this system of exploitation by the late 19^th century resulted in higher health levels. We estimate that the effect of this development was sizeable: as the number of forced laborers called up each year was about halved between the 1830s and 1870s, this caused mortality rates to decline by between 10 and 30 percent. In our investigation, we find that variations in market prices for coffee and sugar in Amsterdam – that cannot have been influenced by mortality rates in the different residencies of Java – influenced the number of forced laborers. As such, we suggest the link between forced labor in the Cultivation System and mortality was, in fact, causal. 

Towards a better understanding of colonial legacies

In sum, earlier research into the effects of the Cultivation System have highlighted beneficial effects of this extractive institution on commercialization, the rise of the sugar industry and long-run consequences for industrial development and educational attainments. Our research, instead, shows that the forced labor demands of this system had important negative health effects, as measured by death rates. The contrast between our results with those of studies analyzing a longer period of time are not necessarily contradictory. In fact, this shows that the effect of colonial policies on development trajectories was manifold and that we need to expand our approaches to fully comprehend colonial legacies, a topic that has recently received substantial attention (read more about this here). Focusing on other indicators than economic performance is one way to do this, as we have shown, but future research could also consider how different sub-national regions may have been unequally affected by similar colonial policies, because their geographic endowments or local institutions and culture differ.

---