7. Reconciling Boserup with Malthus: Agrarian Change and Soil Degradation in Olive Orchards in Spain (1750–2000)

In the mid-eighteenth century, Montefrío had 5,108 inhabitants, a scant population in comparison with other territories in northern Spain (23 inhabitants/km²), with high land availability, in excess of 4 ha per inhabitant. Land use was primarily devoted to livestock, offering future possibilities for the agricultural “colonisation” of the town. It was, as Herman Daly said (2005), an “empty” territory where labour, rather than resources, was the main limiting factor.

Land use was organised according to different levels of labour intensity. Land occupation occurred was based on a primary population nucleus and on small population clusters known as cortijos scattered throughout the municipality. Cultivation decreased in intensity with distance from these clusters.² The land that surrounded the town (ruedos) was cultivated intensively because more labour could be applied there. Some areas had access to irrigation and thus were where fruit, vegetables, cereals and leguminous crops were grown. The available manure was dedicated if possible to these lands because of the transport economy and because the costs in labour could be easily borne by the town (for more details, see González de Molina et al. 2010; Infante-Amate 2011) (Table 7.1).

The majority of the cultivated land (28.8 %) where “al tercio” (wheat/fallow/fallow) rotation was practised was located adjacent to this area, with fairly modest sowing and harvesting. Finally, grape vines and olive trees, which received low levels of labour during this period, occupied a marginal area (0.7 %), and their products provided subsistence for the locals. The land area dedicated to permanent, natural pasture (51.1 %) reveals that a very significant portion of the agro-ecosystem remained uncultivated, as home to significant livestock numbers (21 LU/km²).

In accordance with the tradition inaugurated by Marshall Sahlins (1972), Montefrío appeared as a kind of “opulent society”, which, through high levels of labour productivity, was able to feed a scant population more effectively. Indeed, the available food per capita in 1750 had a higher energy content than in the nineteenth century (González de Molina et al. 2011). Recent anthropometric studies have shown that the biological standards of living for its inhabitants were among the highest in Spain at that time (Cámara 2007).

During the second half of the eighteenth century, population growth was practically zero; however, in barely a century, it nearly doubled to 10,101 inhabitants in 1877.³ Following the fall of the Ancien Régime, new liberal policies facilitated access to land to a great number of landless peasants. Over the course of the nineteenth century, there were as many as three land distributions in Montefrío, which attracted significant population numbers (Martínez 1995). Other liberal measures, such as land seizure and the redemption of tithe rent charges, also facilitated access to land ownership. The availability of labour for more intensive management of the agro-ecosystem was greater and consequently so were the feeding requirements, which in turn spurred further reclamation and crop intensification.⁴

Analysing the structure of ownership and agrarian exploitation in Montefrío demonstrates that the number of small landowners increased as the number of large landowners decreased. Farms between 0 and 10 hectares in 1752 accounted for less than 3 % of the land area registered on the cadastral register. One century later, this figure had increased to 7 %. In contrast, in the mid-eighteenth century, farms over 500 ha accounted for over 40 % of all lands, whereas in 1850 this figure had fallen to 18 % (see Table 7.2). A large group of peasant farmers were able to access land ownership (Martínez 1995, p. 163).

However, the population growth and the consequent increased pressure on resources were beginning to decline by the end of the nineteenth century. Indeed, the number of small landowners continued to increase through the liberal measures outlined above, and as a result of the opportunities for clearing and intensifying production, they were favoured by the egalitarian legacy characteristic of Spanish legislation. Relatively isolated from the main commercial points of the region, the town was by necessity self-sufficient. By the beginning of the twentieth century, Montefrío had over two thousand landowners. Approximately 90 % of them, 40 % of the registered plots of land, had fewer than 50 ha. The greatest demographic increase was recorded among landowners with insufficient land; the number of peasant farmers who owned fewer than 5 ha rose from 192 in 1752 to 533 in 1852 and had reached 1,490 by 1901, representing an approximately eight-fold increase. The average area of land owned by this segment also decreased, from 1.67 ha per owner in 1852 to 1.51 in 1901 (see Table 7.2). Certainly, the number of landowners increased at a much greater rate than did the population as a whole, but this occurred under increasingly precarious conditions because of the scarce provision of land and because the productive possibilities of this land were being exhausted. This meant increasingly difficult access to labour and to income livestock and fewer opportunities for work, which stagnated in availability. The composition of their diet also deteriorated (González de Molina et al. 2011), and anthropometric studies display a sharp decline in biological standards of living, also accompanied by waves of emigration (Cámara 2007) and increasing development of social conflict (Cruz 1994).

At the end of the nineteenth century, the amount of land dedicated to agriculture had increased to levels never before observed, and more intensive rotations occupied more space than ever before. The only way of fulfilling the dietary needs of the town was to replace livestock farming, which could only be sustainable under low population densities, with another form of agriculture based on cereal crops. This change seems to support the classic strategy described by Boserup (1965, 1981), whereby territories that begin to become densely populated intensify their rotations and increase production to adapt to the limitation. However, focusing solely on population dynamics ignores the manifest importance of the institutional dimension of change, which in this case encouraged population increases or immigration into the town and also governed the way in which the rural community was able to colonise the territory or intensify its management.

The increasingly precarious situation of the peasantry of Montefrío, given the imbalance between population and resources, explains here and elsewhere the increasingly intensive use of the land to guarantee subsistence, which led to processes of soil degradation (Blaikie and Brookfield 1987; FAO 1983).

Olive growing is the best metaphor for the changes that have occurred in the Spanish Mediterranean in the past two centuries. It encapsulates the intensive progression of agro-ecosystems in the nineteenth century, and its level of expansion in recent years has meant that the area of land dedicated to this activity in certain territories such as Montefrío occupies the total utilised agricultural area.

In the mid-eighteenth century, olive trees were scattered in Montefrío, integrated with other uses and subject to fairly non-intensive management. Olive production barely exceeded 100 kg per ha, whereas production was greater than 500 kg by 1900. Although the production as total biomass of the tree remained relatively stable (including the use of pruned material and ground cover plants), there were increased efforts to improve olive production, which is the most valuable part of the crop. Human labour increased from 11 days/ha in 1750 to 28 days in 1900, and animal labour increased from 4.39 obradas/ha to 10.65 obradas/ha. Although higher olive yields were obtained, total crop productivity, measured as total biomass, decreased. In spite of this, olive orchards were a good solution for peasants with insufficient land.⁵ In addition to olives, they provided wood and timber from pruning for domestic fuel, leaves to feed goats, ground cover plants to feed sheep, and pomace to feed livestock. In terms of energy, these products represent between 80 and 90 % of the total usage (Infante-Amate 2011, 2012), and these products were decisively important for these peasant economies.

Therefore, olive orchards charted a transition from an “empty” (Daly 2005) and “opulent” (Sahlins 1972) world towards more intensive cultivation, partially linked to the market, which required more intensive labour despite the costs in productivity (Boserup 1965). This is a similar story to what the changes in land use throughout Montefrío: increased agricultural intensity based on a loss of labour productivity.

However, the major transformation of agriculture in Montefrío, as in the rest of the country, came in the second half of the twentieth century and was characterised by the industrialisation of the sector and the full expansion of olive growing. Spain joined the European Common Market in the 1980s and its agricultural policy in the early 90s while also achieving high levels of industrialisation and consolidating its integration into the global economy. The importation of great quantities of energy and materials has allowed the cultivated land area to occupy essentially the entire utilised agricultural area (Guzmán and González de Molina 2009). This socio-ecological transition in Spanish agriculture eventually isolated the processes of food production and consumption from the territory, as has also occurred in other countries (Erb et al. 2009; Würtenberger et al. 2006).⁶ Because of the mass importation of grain, meat consumption levels have returned to those last observed in the eighteenth century (Infante-Amate and González de Molina 2013), and land has become available for other uses. Production specialisation was accompanied by the expansion of olive growing, primarily because of the ecological advantages of this tree for development in Mediterranean climates (Angles 1999) and also because of the public subsidies of the Common Agricultural Policy (De Graaf and Eppink 1999).

Today, there are approximately 10 Mha under olive cultivation in the world, over 80 % of which are in the Mediterranean basin (FAO 2013), primarily because towns such as Montefrío have focused their agrarian activity fully on the production of olive oil; approximately 90 % of cultivated land is allocated to this crop, which is largely managed industrially. This latest leap towards agricultural intensification has substantially improved labour productivity (see Table 7.3). As recently discussed by Fischer-Kowalski et al. (2010), the productivity of labour could once again increase in an unprecedented way because of the application of substantial amounts of energy, largely from fossil fuels.

The model of agrarian growth that accompanied the industrialisation of Spanish agriculture was based on the forced transfer through the markets of a portion of agrarian incomes (González de Molina and Guzmán 2006). This resulted from the on-going deterioration in the exchange relationship between the agrarian sector and the urban industrial and services sector. In comparison with other sectors, agrarian income suffered a significant decline in real terms. Between 1990 and 1999, income in the province of Granada—to which Montefrío belongs—had fallen by 15 % (Analistas Económicos de Andalucía 2000).

Furthermore, the process by which the subsidies of Common Agricultural Policy incentivised productivity through the intensification of cultivation and through savings in labour costs is well known (De Graaf and Eppink 1999). Much of the technological package integrated into the management of olive orchards responds to the new olive-growing reality of the region, including part-time agriculture, an ageing population and a loss of profitability⁷, factors that are separate from the dynamics of population pressure.

The expansion of the amount of land used to produce food for human consumption was achieved at the expense of space used to grow feed for livestock (Table 7.1). Livestock numbers declined between 1752 and 1854, a trend that continued until the end of the nineteenth century. Livestock density was reduced to a third of its previous size. The expansion of cultivated land and the net increase in demand for animal traction explain why the decrease in working animals was not as marked. However, the decline was most obvious in income livestock, which were fed on products of the agro-ecosystem that could not be exploited to grow food for humans or labour livestock. By 1901, this type of livestock had fallen to just over 12 % of the figure registered in 1752.

One consequence of this process was a 54 % net reduction in the fertilisation capacity of livestock. The imbalance between cultivated lands and those dedicated to animal feed, at a time when significant transportation of grain, straw or manure was limited (González de Molina 2010; Sieferle 2001), diminished fertilisation capacity. In truth, the possibilities of agrarian growth in Montefrío had reached their practical limit in the final decades of the nineteenth century. The possibilities for increasing the amount of cultivated land were very limited. In general terms, yields had begun to level off and, in some cases, even to decline. The decrease in production per inhabitant shown in Table 7.1 is explained by this fact. However, harvested biomass continued to grow during this period. How was the global increase in production between 1852 and 1901 possible, therefore? It was achieved at the expense of the nutrient reserves in the soil. The decline observed in the amount of manure applied per cultivated hectare confirms this: it was reduced to a fifth of the former levels.

In previous articles (García-Ruiz et al. 2012; González de Molina et al. 2010), we have examined the balance of nutrients on an aggregate scale and for the main types of crops. Table 7.4 provides a summary of our primary findings in this regard, showing that the extraction of nutrients exceeded inputs as crops became more intensive. In the mid-eighteenth century, the nutrient balance for crop rotations and on an aggregate scale was positive, with the exception of a few losses in potassium that were not significant because of its abundance in the soil. By the mid-nineteenth century, the balance of potassium had become even more negative, and the phosphorus balance became negative for the first time. Something similar occurred at the end of the century with perhaps the most important macronutrient in preindustrial agriculture: nitrogen (Allen 2008; Cunfer and Krausmann 2009). This was yet another symptom of the difficulties facing the agro-ecosystems of Montefrío in the final few decades of the nineteenth century: the progressive exhaustion of the possibilities for intensifying production, the mining of nutrients (González de Molina et al. 2010), the declining biological standards of living, and emigration (Cámara 2007). These processes may have spread to other areas in southern Spain (Infante-Amate 2011).

Earlier, we associated the increasingly precarious situation of the peasantry in the late nineteenth century with the greater intensity in farming. According to our data, this association led to a process of soil degradation, which revealed itself in higher rates of erosion in the most expansive crop grown in southern Spain, olive orchards, and also in widespread nutrient mining that limited productive capacity.⁸

The negative balance registered in olive orchards continued to increase until the late nineteenth century. Although the balance of nitrogen was always positive, the mining of phosphorus and, to a greater extent, potassium increased. Although reserves of phosphorus and potassium were very high in the soil, the continued losses led to a long-term decline. According to García-Ruiz et al. (2012), the intensive levels of extraction practised in the late nineteenth century would have affected the production capacity of the olive orchards in a few decades.

In the final decades of the twentieth century, the availability of inexpensive synthetic fertilisers has spurred their uncontrolled use, especially for nitrogenous fertilisers. Over 60 % of the nitrogen applied is not absorbed by the plant (Hermosín et al. 2009), and that nitrogen consequently causes significant water and soil contamination (Beaufoy 2001; Tombesi et al. 1996). In addition, this same management approach is responsible for the loss of ground cover plants and, therefore, the increased erosional processes in olive orchards (see following section).

During previous research (see Vanwalleghem et al. 2011), we reconstructed soil losses in the olive orchards of Montefrío from 1750 until the present. This quantification was performed using measurements taken in three plots of land located within the town, following the methodology defined by Vanwalleghem et al. (2010).⁹

Table 7.5 shows the primary results, reflecting the average values for the three plots of land studied.¹⁰ The average soil losses (just over 40 t/ha/year) significantly exceeded the natural rate of soil regeneration (Montgomery 2007), but different values were observed over the course of the period studied. A clear upwards trend is observed from the lowest losses in the mid-eighteenth century to the present highest losses, with the exception of the mid-twentieth century when the losses declined.

Because the level of rainfall, the gradient, and soil type remained relatively stable throughout the period studied, the soil losses must be directly linked to changes in soil management.¹¹

In the eighteenth century¹², the olive orchards of Montefrío were barely tilled once a year, using a few people and animals. From the mid-nineteenth century onwards, with the new liberal reforms¹³ and subsequent demographic expansion, olive cultivation began to increase in the town. The sources describe greater intensity in soil management: the soil was tilled twice a year, between April and May, the bases of the olive tree trunks were dug out and, during the summer, manual reaping was replaced by very superficial tilling using harrow passes. In the late nineteenth century, the soil continued to be tilled at least twice a year, using more human and animal labour; hence, the soil remained bare for longer, and the risk of erosion consequently increased. This period also saw the introduction of the mouldboard plough. This technology penetrated the land even deeper, considerably increasing the risk of erosion.

From 1936 to 1975, the management of olive orchards changed radically. The autarchic policies implemented during Franco’s regime discouraged the production of olive oil (Tió 1982) and incentivised cereal crops (Christiansen 2002). Throughout the country, farmers sowed cereal crops between their rows of olive trees to alleviate these years of shortage, known as the “years of hunger”. Therefore, paradoxically, soil losses under Francoism fell to historically low levels by increasing the amount of plant ground cover between olive trees.

The major change came in the early 1970s, when the use of tractors increased, replacing animal labour. The greater efficiency of mechanical traction enabled farmers to significantly increase tillage. At the beginning of the 1990s, another important change took place. Mouldboard ploughing, which was used during the spring to eliminate spontaneous ground cover, was replaced with the application of weed killers.

Comparing the soil loss in the olive orchards of Montefrío with the changing number of inhabitants of the town, we see that these two trends have not been parallel over the past two centuries. The common history of soil degradation and land use practices indicates that agriculture in Montefrío intensified as the population increased during the nineteenth century until significant soil losses occurred. In other words, in the context of pre-industrial production, there was a high correlation between population growth and erosion. This close coupling between population and soil degradation disappeared in the twentieth century. The institutional features of Francoism resulted in a drastic decline in soil erosion, while population size increased further; in the final decades of the twentieth century, soil degradation increased again to high levels because of the fast expansion and industrialisation of olive production at a time when demographic pressure reached historic lows.

Agronomic literature commonly points to soil degradation in olive orchards as one of the main environmental problems in this region (Beaufoy 2001; Gómez et al. 2008; Kirkby et al. 2004). However, in relation to the discussion here, it is worth examining the causes that precipitated this change. Was population pressure responsible? Figure 7.2 reveals that in recent decades, the number of inhabitants has been declining to all-time low figures, whereas soil degradation in olive orchards is achieving maximum levels. Boserupian analysis also proves to be insufficient here to account for the socio-ecological transition towards an industrial metabolic regime that occurred in Montefrío (Sieferle 2001).