The diffusion of unsuccessful innovations: the myth of raised field agriculture

Pre-Columbian raised field agriculture is an extremely interesting topic that we have discussed in this blog before, here, here and here. We call raised fields “any prepared land involving the transfer and elevation of soil above the natural surface of the earth in order to improve cultivating conditions” (Denevan and Turner, 1974). Raised fields have received a lot of attention in studies related to pre-Columbian demography. They have been considered key in allowing dense populations of complex societies to inhabit seasonally flooded regions of South America in pre-Columbian times. Beyond their academic interest, raised field agriculture has also become popular among rural development workers and aid agencies working with small farmers in South America. For more than 30 years now, some archaeologists and NGOs have favoured the “re-introduction” of this pre-Columbian agricultural technique among modern day farmers. In their view, raised fields (which are considered analogous to the Chinampa system in Mexico) represent a promising pro-poor agricultural innovation which is more productive and sustainable than traditional agriculture (which, in the neo tropics is, basically, slash and burn).

A few days ago, Philippe C. Baveye published on-line a thorough debunk of such proposals  (Baveye in press) . It is a comment to a paper by Renard et al (Renard et al., 2012) where the authors advocate for the adoption of raised fields agriculture among modern farmers. The main points that Baveye puts forwards are: 1) There is no evidence that raised fields have supported dense populations in the past; 2) the pre-Columbian Chinampa system is unique. Therefore, it cannot be assumed that the productivity of the South-American raised fields, which are very different from the Mexican Chinampas, was comparable to that of the Chinampas; and 3) the purpose of raised fields was limited to water management:  soil drainage and/or irrigation. We reach very similar conclusions in (Lombardo et al., 2011). It is important to highlight that none of the projects designed to reintroduce raised field agriculture among small farmers in Bolivia and Peru have ever worked (actually, I am not aware of any successful rehabilitation anywhere in the Americas). Local farmers have no idea or knowledge about this ancient practice: this kind of indigenous knowledge was lost hundreds of years ago when the Spaniards got to America. Therefore, besides the technical problems described above, another reason behind this large scale failure is that raised field agriculture has been entirely “invented” by archaeologists and NGOs: the latest example I know of raised field rehabilitation project failures comes from an Oxfam project in the Beni - Bolivian Lowlands. The picture below shows the current state of the raised fields built by Oxfam in 2009, amid an important media coverage and support (for example, see this BBC article).

In 2011 raised fields were also built in the northern Beni.You see the state of the fields during the summer 2012 (photo below).

Hopefully, Baveye’s comment will contribute to make researchers more cautious before proposing the ‘reintroduction’ of raised field agriculture in rural communities.

References

Baveye, Philippe C. (2013). Comment on “Ecological engineers ahead of their time: The functioning of pre-Columbian raised-field agriculture and its potential contributions to sustainability today” by Dephine Renard et al Ecological Engineering

Denevan, W. M., and Turner, B. L., 1974, Forms, functions and associations of raised fields in the old world tropics: Journal of tropical geography, v. 39, p. 24-33.

Lombardo, U., Canal-Beeby, E., Fehr, S., & Veit, H. (2011). Raised fields in the Bolivian Amazonia: a prehistoric green revolution or a flood risk mitigation strategy? Journal of Archaeological Science, 38 (3), 502-512 DOI: 10.1016/j.jas.2010.09.022

Renard, D., Iriarte, J., Birk, J., Rostain, S., Glaser, B., & McKey, D. (2012). Ecological engineers ahead of their time: The functioning of pre-Columbian raised-field agriculture and its potential contributions to sustainability today Ecological Engineering, 45, 30-44 DOI: 10.1016/j.ecoleng.2011.03.007

Effects of the expanding agriculture frontier in the Bolivian Amazon

Last week, a paper published in Nature spurred a lot of debates on the internet about the future of agriculture and our ability to feed the 9 billion people that the world will have in 2050. One important aspect related to this debate is the availability of agricultural land. However, people do not always have a clear idea of what expanding the agriculture land means. Here an example of what is happening in one of the most biodiversily rich places of the world: the Bolivan Amazon. Let’s just have a look at a few Google earth’s images of the Lake Peroto in the Llanos de Moxos (dep. of Beni in the Bolivian Amazon).

Image taken the 17 of May 2003. Land use is extensive cattle ranching. Typical floating vegetation (locally called “yomomo”) is growing along the shores of Lake Peroto.

Image taken the 20th of Nov 2004. Land use is extensive cattle ranching. It is the end of the dry season, and the water level is extremely low. In the greenish areas inside the lakes water is almost completely evaporated. This is normal in the Llanos de Moxos, where seasonality can be extreme

Image taken the 2nd of July 2009. Land use is extensive cattle ranching. We are at the beginning of the dry season, water level is quite high, the water is clearly visible.

Image taken the 8th of August 2011. Land use is extensive cattle ranching plus industrial agriculture. 70 hectares of land in the surroundings of the lake have been used to produce rice.

What happened to the lake in 2011? Now the lake is almost completely covered with vegetation... Is it dry? Or is it the first effect of land use change: eutrophication? Based on my personal experience of having been working in the Llanos de Moxos for more than 10 years now and based on the fact that the nearby Lake Suarez is full of water (photo below) on the 15 of August 2011, I would conclude that Lake Peroto is in process of eutrophication.

Lake Suarez. Image taken the 15th of August 2011. The lake is full.

Industrial agriculture in the Beni is already causing deforestation and threatening archaeological sites (see this). Now it is also affecting lake ecology. Is industrial agriculture what people in the Beni really need? Does this kind of agriculture make any sense in a place with a density population of 1 person per square kilometre? Here, a very important part of the population lives in indigenous communities whose subsistence depends on local natural resources. What will happen to them in ten years’ time if these resources are depleted?

Ref:

Seufert V, Ramankutty N, & Foley JA (2012). Comparing the yields of organic and conventional agriculture. Nature PMID: 22535250

A story of people and rivers in the Amazon of 5000 years ago

This time I will tell you about a story that began in the Mid-Holocene (5000 years ago) and is set in the Bolivian Amazon. More precisely in the south-eastern part of the Llanos de Moxos seasonally flooded savannah, in what we call the Monumental Mounds Region MMR (Fig. 1).  Here, between 400 and 1400 AD, pre-Columbians built hundreds of monumental earth mounds, known locally as “lomas”. These earth mounds are planned, complex buildings made by one or more pyramids built on top of elevated platforms (Fig. 3-3).  Monumental mounds can be up to 20 meters high and can cover up to 30 hectares. In the MMR there are more than 350 of these pre-Columbian buildings.

Figure 1 The Llanos de Moxos, the MMR (yellow
box) and the paleo-courses of Río Grande (pink lines). 

During the early Holocene (between 11.000 and 5.000 years ago) this portion of Amazonia was relatively dryer than today, inundations were less frequent and rivers transported few sediments. During these stable climatic conditions there was no deposition of fluvial sediments in the savannahs and soils were forming all over the Llanos de Moxos. 

Mid- Holocene: Río Grande built the landscape

But things changed a lot during the Mid-Holocene, between 5 and 4Ky BP. The Rio Grande (pink lines in Fig. 1) entered in a period of frequent avulsions and high sedimentation, probably triggered by a climate change towards wetter conditions. As a result, in the South-eastern LM, it formed a ­fluvial distributary system (FDS) (Fig. 2-1). Suddenly, the landscape was transformed into a large swamp, dominated by something similar to an interior delta. This FDS deposited several levees, crating relief at a local scale (Fig. 2-2), and a sedimentary lobe, creating relief at a regional scale (Fig. 2-3). The former soils were buried and the landscape became a mosaic of patches of savannahs closely interwoven and sometimes enclosed by forested paleo-levees.

Late Holocene: pre-Columbians transformed the landscape

In the MMR, the lobe deposition favoured the development of a complex pre-Columbian society by increasing the region’s agricultural potential. Firstly, it created a convex-up topography, which greatly reduced its susceptibility to ­flooding; secondly, the construction of the elevated ­fluvial levees significantly improved drainage conditions at the local scale. Furthermore, the Río Grande also provided relatively younger sediments derived from its Andean catchment that are rich in nutrients. Thus, the Río Grande removed the two biggest obstacles faced by tropical agriculture in the rest of Amazonia: severe waterlogging and poor soils. But the Río Grande’s job was not perfect: fl­uvial levees enclosed patches of ­floodplain, resulting in ponding and pronounced waterlogging. Thus pre-Columbian people had to transform the landscape through the construction of a drainage system in order to further improve agricultural conditions (Fig. 3-1).

The network of canals had a significant impact on the edaphology of the MMR: it pushed the forest-savannah boundary towards the savannah, eventually increasing the area of well-drained, usable land. The inhabitants of the MMR were very lucky because they also had several lakes placed on the top of the sedimentary lobe. Building canals that transported the water from the lakes to the agricultural fields (Fig. 4), they were able to perform agriculture even during the dry season.

The spatial overlap (Fig. 3-2) between the monumental mounds and the area of deposition of the sedimentary lobe created by the Río Grande during the mid- to late Holocene  suggests that good edaphic conditions favoured the emergence of the monumental mounds culture (Fig. 3-3).

The ­fluvial landscape created by Río Grande was probably an important factor behind the emergence of the monumental mounds culture in the South-eastern LM, as it provided favourable environmental preconditions in terms of soils, nutrients and drainage characteristics. Pre-Columbians additionally modified and improved their environment by building a network of drainage canals.

Figure 4 (a) Drainage canal (A–B) and irrigation canal (C–D) 

built along topographic slopes. (b)The C–D canal in detail;

 the irrigation canal linking Lake San José to the Cotoca mound 

joins pre-existing natural channels. Coring in the middle of 

the canal shows that the bed of the canal is only 50cm below 

the present bed (which is about 50 cm below the modern savannah). 

Therefore, this canal was at most 1 meter deep but went through a slope 

of 3 meters. The only possible function of such a canal was irrigation.

You can find more on this story in my latest paper published on-line in the journal “The Holocene”. It is entitled “Mid- to late-Holocene fluvial activity behind pre-Columbian social complexity in the southwestern Amazon basin” and can be found here. 

Lombardo, U., & Prümers, H. (2010). Pre-Columbian human occupation patterns in the eastern plains of the Llanos de Moxos, Bolivian Amazonia Journal of Archaeological Science, 37 (8), 1875-1885 DOI: 10.1016/j.jas.2010.02.011


Umberto Lombardo, Jan-Hendrik May, & Heinz Veit (2012). Mid- to late-Holocene fluvial activity behind pre-Columbian social complexity in the southwestern Amazon basin The Holocene

Review of Betty J. Meggers‘ review of the Handbook of South American Archaeology

Betty J. Meggers has been working in Amazonian archaeology for more than 50 years. She has recently published a review of the Handbook of South American Archaeology, which is of great relevance to all of us working on the paleoecology of Amazonia during the Holocene. You can download her paper here.

The paper discusses 3 fundamental aspects of the South American archaeology: i) how contemporary Amazonian archaeologists’ interpretation of the archaeological record has been biased by the abandonment of the classical archaeological methods of pottery analysis; ii) the origin of new world pottery and iii) the relationship between environmental conditions and cultural development in Amazonia.

Classical archaeologists spent months on end drawing and analysing thousands of pieces of pottery. In the first part of her paper, Meggers explains why this meticulous work is a fundamental step in scientific archaeology: “Pottery can be decorated using an essentially unlimited number of techniques and motifs without affecting the utility of the vessel, making independent duplication of identical decoration unlikely”. Therefore, the analysis of pottery’s details and decorations allows us to distinguish diffusion (cultural aspects which are transmitted form one cultural group to another) form independent invention, where a cultural trait arises spontaneously in a given population. According to Meggers, some of the authors of the Handbook omit this important task, undermining the strength of their interpretation of archaeological records

Pottery analysis is the criteria that Meggers uses to discuss, in the second part of her paper, the origin of new world pottery. Many archaeologists that have contributed to the Handbook believe that pottery was independently invented in Amazonia. In Meggers’ view, the similitudes between Japanese Jomon pottery and the Valdivia pottery (which is the oldest pottery in America, dating 6000 BP) are too many to be the result of some form of “cultural convergence”. Moreover, she highlights that the period of Valdivia pottery coincides with a catastrophic volcanic eruption in Japan, which could have pushed groups of Japanese fishermen towards the American coasts. Hence, these Japanese fishermen would have influenced the pottery of early Americans. .

In the Handbook, Amazonia is depicted as a “manufactured landscape” or “anthropogenic cornucopia”. In the third part of her paper, Meggers assesses whether or not such descriptions of Amazonia are supported by the archaeological evidence. She shows that the evidence is actually very weak: the idea of widespread human occupation of pre-Columbian Amazonia is not supported by pollen, phytolith and charcoal analysis, which indicate that vast areas of Amazonia have never faced human disturbance; no systematic archaeological excavation has ever been performed that supports the assumption that large permanent settlements were common in Amazonia; many of the earthworks often cited in support of the “manufactured landscape” idea are concentrated in the Llanos de Moxos, which are ecologically quite different from the tropical rainforest that covers most of the Amazon basin; there is no evidence suggesting that ADE was created for intensive agriculture or to exclude that slash and burn agriculture was a common practice among pre-Columbian people.

Despite the fact that the bibliography of Meggers’ paper is not as large as it could have been, it can still serve as a good introduction to the archaeology of Amazonia for anyone who wants to get into the heart of the current debates.

Betty J. Meggers (2011). Handbook of South American Archaeology reviewed by Betty J. Meggers Revista de Antropología Chilena, 43 (1), 147-157

Soybean industrial production is bulldozing pre-Columbian archaeological sites in the Bolivian Amazon and nobody gives a damn

The journal Applied Geography and the journal Land Use Policy have recently published two papers, “Spatiotemporal modeling of the expansion of mechanized agriculture in the Bolivian lowland forests” and “Deforestation dynamics and policy changes in Bolivia’s post-neoliberal era” respectively, that depict a desolating panorama. The rate of deforestation under Evo Morales’ government is even higher than it was during the previous governments. Muller et al. say that “While overall dynamics remained relatively stable over time, the expansion of mechanized agriculture between 2001 and 2005 became more tolerant to excessive rainfall and less dependent on fertile soils. This mirrors the increasing penetration of mechanized agriculture into humid and less fertile Amazonian rainforests in the northern portion of the study area [Santa Cruz]. The map of deforestation probability substantiates these patterns and shows the highest propensities for future deforestation in the north”, while Redo et al. point out that: “Although neoliberal policies triggered an unprecedented level of forest clearing in Bolivia, rates have generally continued to increase and can be indirectly linked to the administration’s new agrarian reform and pro-environmental regulations”. The “pro-environmental” regulations of Morales’ government have actually increased deforestation rates!

The Northern Bolivian lowlands are not only an extremely important reservoir of biodiversity and home of many indigenous communities; they also hold an impressive amount of archaeological sites, most of which have never been studied or surveyed. As deforestation for industrial soybean production is moving northward, it now starts to affect the Llanos de Moxos, where most of these archaeological sites are found. Soy producers cut down the trees using bulldozers, and, in this way, they also destroy all the archaeological sites they encounter on the way. In the eastern Llanos de Moxos, in a stripe of forest called Monte San Pablo, between the River Cocharca and the River San Pablo (Fig. 1), Bolivian Mennonites are bulldozing pre-Columbian human-made earth mounds, which hold valuable remains and information about past Amazonian cultures. The destruction of pre-Columbian archaeological sites in the Bolivian Amazon is taking place while national and local governments look another way, and regrettably, with the complicity of some of the indigenous leaders who live in the area. 

Figure 1 Shaded areas are forests. Continuous line indicates the area of pre-Columbian monumental mounds. Monte San Pablo is east of the area (dashed lines) shown in Fig.2. (Lombardo and Prümers, 2010)

Figure 2. Modis Image taken a few days ago. The yellow stripes highlighted by the arrow are the portions of Monte San Pablo that have already been cleared for soybean production. The deforested area is already larger than 10.000 hectares

I have surveyed more than one hundred pre-Columbian monumental mounds and hundreds of Km of pre-Columbian canals and causeways in the area east of Trinidad, in the Beni (see map in fig.3). The monumental mounds are huge earthworks (the average mound covers 5 hectares and is 9 meters high) entirely human made, built from 400 AD to 1500 AD. They are full of pottery and burials and only very few of them have been excavated by archaeologists. The Monte San Pablo, just east of the area I surveyed, is full of monumental mounds. Those are the mounds that are being destroyed by the Mennonite bulldozers to make room for soybeans.

Figure 3. Triangles are pre-Columbian monumental mounds, lines are canals and causeways, shaded areas are forests. (Lombardo and Prümers, 2010)

Daniel Redo, Andrew C. Millington, & Derrick Hindery (2011). Deforestation dynamics and policy changes in Bolivia’s post-neoliberal era Land Use Policy : 10.1016/j.landusepol.2010.06.004

Robert Müller, Daniel Müller, Florian Schierhorn, & Gerhard Gerold (2011). Spatiotemporal modeling of the expansion of mechanized agriculture
in the Bolivian lowland forests Applied Geography : 10.1016/j.apgeog.2010.11.018

Umberto Lombardo, & Heiko Prümers (2010). Pre-Columbian human occupation patterns in the eastern plains of the Llanos de Moxos, Bolivian Amazonia Journal of Archaeological Science DOI: 10.1016/j.jas.2010.02.011