by Desconhecido
The good agreement between the land-cover reconstructions based on pollen data and the well-documented landscape development for the last century increases the credibility of the reconstruction also for earlier periods. Furthermore, the fact that modelled land cover of heather vegetation shows a sharp decrease in the twentieth century, when grazed heathland disappeared but natural peat bogs in the uplands changed very little, indicates that most of the heather vegetation, according to the diagram, belonged to pastures.
We may now return to the decrease in landscape openness that took place in the fourteenth century. Apart from cereals, particularly heather decreased whereas birch and some other trees expanded. It reflects how pastures on poor soils were abandoned and gradually became overgrown particularly by birch in a natural succession. Also sedges and grasses show a slight decrease. The only open-land vegetation that increased at this time, according to the reconstruction, was mixed herbaceous plants. It may reflect an expansion of weeds and other plants in abandoned arable fields when crop growing stopped and they were turned into pastures or meadows.
An interesting observation is that the decrease of open-land vegetation from the early to the late fourteenth century was proportionally much smaller than the decrease in crop growing. Whereas the total land cover of open-land vegetation decreased from 38% to 33% (a decrease of 13%), the land cover of cereals was halved. Obviously landscape openness did not decrease as dramatically in the wake of the plague as the extent of arable fields. Even though some of the open-land vegetation grow in natural wetlands that were not affected by agricultural change, the much smaller relative decrease in landscape openness than in cereals indicates that grazing (and possibly mowing) did not decrease as much as crop growing during the crisis.
Hence, in addition to the change in landscape openness, there was also a change in the proportional relationship between different types of agricultural land. In the traditional agricultural landscape of southern Sweden, areas connected to animal husbandry (pastures and meadows) usually covered much larger areas than areas used for crop growing. The relationship between arable fields, meadows and pastures differed between regions and also changed through time, but everywhere, except for in the most densely populated plains, pastures and meadows dominated the agricultural landscape. In the uplands, where animal husbandry played a more important role than in the lowlands, arable fields represented only a minor part of the productive land. The hamlet Lönsboda in the southern part of the uplands is a typical example.53 According to a map from 1696, it consisted of two households with 2.5 hectares of arable each. The total arable made up less than 2% of the land belonging to the hamlet, whereas the rest consisted of meadows, pastures, woods with clearings, swidden land and bogs. The meadows were more than ten times larger than the arable.
To conclude, animal husbandry has always been important in the uplands, but it became even more so after the Black Death. Population drop and farm abandonment resulted in a distinct and sudden decrease in crop cultivation, whereas land associated with animal husbandry decreased much less. The pollen record thus indicates that animal husbandry gained in relative importance after the Black Death. The same conclusion has been put forward by several authors based on the historical record.54 In Sweden indications are found in cadastral registers from the fifteenth century onwards. According to them, many deserted farms were used as pastures and meadows.55 Another indication is the increased production of butter, reflected in declining prices of butter relative to grain prices.56 Even though it is difficult to quantify the increased importance of animal husbandry, the landscape reconstruction shows that this change of the agricultural system resulted in vegetation changes that were distinct enough to show up in the pollen record on a regional scale. Because animal farming is more land demanding but less labour intensive than crop growing, this relative increase in animal husbandry at the expense of arable appears to have been a strategic, labour-saving change in a time of land excess but shortage of manpower.57
It is not possible from the pollen-based landscape reconstruction to distinguish hay meadows for fodder production from pastures. Even though animal husbandry in general was less labour intensive than crop growing, the collection of winter fodder by mowing was rather labour intensive. At the time of hay harvest, mainly in late July to early August, the size of the workforce was a limiting factor. The development during the Late Middle Ages of longer scythes and better rakes made hay mowing more efficient and less time consuming, but fodder production still remained a bottleneck in the animal production.58 Cadastral registers mention deserted farms used for hay mowing, at least during the late fifteenth century and later, but probably the landscape was kept open first of all by grazing. Fenced-in meadows and arable were grazed after harvest but for most of the grazing season animals roamed freely in the outlands, watched over by herders. The shift from male herders to children and women, which started in the fifteenth century, probably reflects an adaptation to the shortage of manpower after the Black Death, and helped to keep a larger livestock in relation to the number of people.59
In the wake of population drop and abandonment, arable fields were turned to meadows and pastures, and some meadows were turned to pastures as well. At the same time, the decrease in land cover of heather vegetation after 1350 indicates that poor, less productive heathlands were completely abandoned and left to reforestation. The only open-land vegetation that increased was the one represented in Figure 12 by mixed herbaceous plants, which increased in the late fourteenth century from 3% to 4% of land cover. It probably reflects an expansion of weeds and other open-land plants when cereal growing stopped and arable fields were turned to meadows or pastures. Such an expansion of grazing indicators at some sites in the fourteenth century has been shown earlier in a few individual pollen diagrams. Examples of plants that expanded were ribwort plantain, cow-wheat, goosefoot, sorrel and grasses.60 The reconstruction presented here shows that the increase of this type of vegetation is detectable also at a regional scale.
The decrease in land cover of heather and the increase in mixed herbaceous plants indicate that the average quality of pastures and meadows increased. It was the combined effect of two processes – the reforestation of the poorest heathland and the establishment of rich herbaceous vegetation on abandoned arable fields. This appears to have been the result of strategic decisions. By turning arable to pastures and meadows and by leaving the poorest pastures to natural succession, peasants took the opportunity to increase the land productivity of their stock farming.
The changes discussed so far happened in the fourteenth century and were connected to the Black Death and the associated population drop. Eventually, after more than a century, the first signs of agricultural expansion and recolonisation are visible. Both crop growing and animal husbandry (as reflected in landscape openness) started to expand again, but there appears to have been a time lag between the two. According to the landscape reconstruction, cropland reached a nadir in the late fourteenth century (Figs 8b and 12). In the first half of the fifteenth century it remained almost the same with only a slight increase, but for the second half of that century it shows a relatively distinct increase. It marks the beginning of an expansion of arable land that continued more or less uninterrupted to the early nineteenth century. The total land cover of open-land vegetation (Fig. 12) shows a similar development but slightly delayed. After the decrease in the mid-fourteenth century landscape openness remained the same throughout the fifteenth century, and the first marked increase according to the reconstruction was not until the first half of the sixteenth century. From then on, open land expanded to reach maximum openness in the late nineteenth century.
A plausible interpretation of this time lag is that the first agricultural effort, when population numbers had recovered enough to enable expansion, was to increase crop production. Arable fields once abandoned were now put under the plough again. On the stony ground that characterises much of the uplands, it was an obvious advantage to r
eturn to old fields that had already been cleared from stones. In this way the great effort once invested in stone clearance was not wasted.61 The re-establishment of previously abandoned fields is indicated by the pollen-based reconstruction. The increase in cereal cultivation in the late fifteenth and early sixteenth centuries is matched by a decrease of mixed herbaceous plants. This group of open-land plants was the only one to expand in the late fourteenth century (interpreted above as the turn of arable land to pastures and meadows), and it was the first to decrease its land cover when arable started to expand again.
The pollen record from the uplands thus indicates that agricultural expansion after the crisis started by the re-establishment of arable land. By doing so crop production and the total land productivity increased. In other words, more people could be fed on the same land. The introduction of crop growing on the deserted farms was probably associated with the establishment of new buildings and, of course, the moving in of people. Slightly later, in a second step of the expansion, forest clearing begun. This two-step process reflects a difference in the labour input required. To gradually expand pastures in the woods by girdling and fire, for instance by grazing of swidden land, may not have been a very labour intensive task, but to turn woodland to permanent arable certainly was. The felling of trees and clearing of roots to prepare the soil for cultivation was hard work. In contrast, abandoned fields that in the meantime had been used for grazing and mowing would have been relatively easy to put under the plough.
An interesting question in connection to this is to what degree vegetation and land-use during the late-medieval crisis had any influence on the duration of the crisis. As evident from the historical record and further supported by the pollen record, many deserted farms were used for pastures and meadows by other farms. In this way surviving farms got the opportunity to expand their land and in particular their animal production in the wake of the population drop. It is interesting to note that deserted farms used for pasture were frequently registered even though no rent had to be paid.62 Probably they were not seen only as pastures, but also as land that could be put under the plough in better times. In other words they were a resource for future expansion. Hence, the grazing and mowing of deserted farms did not only provide valuable pastures and fodder – by holding the forest back the threshold for future expansion was lowered. Regardless if this was an intentional strategy or not, it was a land use that facilitated the expansion and recolonisation that followed. If all deserted farms had been left for natural succession and reforestation, the threshold for clearance would have been higher and the recolonisation of the uplands delayed.
Expanding woodlands
Reforestation on abandoned land is frequently associated with social crisis and depopulation in the same way as its opposite – forest clearing and deforestation – is associated with agricultural expansion and population growth. Many of the most famous and well-studied examples of social collapse were followed by reforestation, for instance in Central America after the Maya decline in the tenth century, in North America in the sixteenth century, and in Amazonia after the sixteenth century.63 Also the late-medieval crisis in Europe is frequently associated with reforestation.64 According to a much-debated hypothesis by William Ruddiman, reforestation in Europe after the Black Death and in the Americas in the sixteenth and seventeenth centuries was significant enough to result in a reversed greenhouse effect and climatic cooling.65 Although the hypothesis has been questioned, the possible effect of past land-use changes on the climate has received increased attention during the last decade. However, few empirical studies have focused on environmental changes after the Black Death and we still know very little about the degree of reforestation. Several pollen studies provide individual examples of forest expansion, but there have been no attempts to quantify late-medieval reforestation for larger areas.66 Here pollen-based reconstructions using LRA, like the one used in the present study, have great potential.67
Reforestation throughout most of history – in Sweden before the last two centuries – was not by planting but by secondary succession. It was a natural process of woodland re-growth on abandoned land. The forest was held back by agricultural land use and as soon as the impact of this land use stopped or decreased a secondary succession of shrubs and trees would begin. In the traditional agricultural landscape of Sweden, there were sometimes pollards among settlement and in the meadows (Fig. 2 in Chap. 2), but most trees and woodland were to be found in the outlands. Woodland was the natural vegetation and grazing and continuous clearing was needed to keep the vegetation open and to hold the woodland back. Natural seeding of trees occurred also in open pastures but saplings were bitten off by grazing animals, and in the meadows mowing effectively stopped any sprouting of unwanted tree vegetation. Still, tree saplings that germinated in pastures and meadows could survive for many years even if browsed or cut down, ready to sprout whenever they got the opportunity.68
As soon as the impact of people and their animals decreased or stopped, trees would invade the open vegetation by sprouting and seed dispersal. In the uplands the first trees to establish would be birch and rowan.69 (Rowan is insect-pollinated and therefore poorly represented in pollen records.) Also oak and hazel may be quick to expand, as well as goat willow. Of these early succession trees birch was the most important. It is a typical pioneer tree, which quickly occupies cleared surfaces.70 It is also fast growing and will soon produce a dense canopy that covers lower vegetation. The process would be quick on barren soil and slower on a well-developed grass sward, but generally it would take 20–50 years for a canopy to develop.71 Birch and other pioneer trees dominate the first phase of succession. They are, however, shade-intolerant and after a while they will be outcompeted by more shade-tolerant and usually more slow-growing trees. Typical such late-successional trees in the uplands are beech and spruce.
Turning to the result of the pollen-based landscape reconstruction, which was based on all the 21 upland sites, the diagrams in Figure 13 shows the regional land cover for a selection of tree taxa. Of the different trees to expand in the late fourteenth century, birch shows the most distinct increase. It started to expand in connection to farm abandonment in the late fourteenth century and reached a temporary peak in the early fifteenth century. The expansion of birch was a temporary interlude in a long-term declining trend of birch woodland. Oak displays a similar development. A long-term gradual decrease of oak woodland was interrupted by a temporary expansion in the late fourteenth century.
Another tree that shows an increase in land cover in the late fourteenth century is spruce. However, in contrast to birch and oak, the long-term trend of spruce during the Middle Ages was expansion. Spruce is the latest tree immigrant in southern Sweden and the only one to have expanded from north to south.72 Its slow but steady conquest of the uplands started in the northern parts in eight century and by the nineteenth century it had reached also the southern parts.73 The increase in the late fourteenth century appears to have been part of this long-term development, but locally the establishment of spruce forest may have been facilitated by agricultural abandonment.
An expansion of birch in connection to late-medieval farm abandonment, which is evident from the landscape reconstruction presented here, is a distinct feature also of several individual pollen diagrams.74 In at least one of them, the peak in birch-pollen percentages is followed by an increase of pollen from more shade-tolerant trees, reflecting a gradual transition from early to late-successional woodland.75 Such a complete succession from open land to woodland of shade-tolerant trees certainly occurred locally, but is not evident from the regional landscape reconstruction presented here. Probably the time between abandonment and re-expansion was not always long enough for a complete succession. However, it is interesting to note that oak, which similar to birch starts to expand already in the late fourteenth century, peaks somewhat later. The land cover of birch woodland peaks in the early fifteenth century, 50–100 years after the Black Death,
whereas oak peaks in the next 50 years period, in the late fifteenth century. It may reflect the difference in life span and growth rate between the two species. Oak as well as birch quickly colonises abandoned land, but because oak is more slowly growing, birch dominates in the first phase and soon establishes a canopy. However, birch is a short-lived tree that does not usually exceed 100 years.76 Its short life span together with its inability to regenerate under a closed canopy explains why the birch phase of a natural succession is only temporary. Oak trees grow much older and are more shade-tolerant. Therefore the reforestation of oak woodland is a longer process that continues after the first generation of birch trees has died off.
The strength of the landscape reconstruction presented in Figure 12 is that it is based on a large number of sites from a relatively homogenous region – the South-Swedish Uplands above 100 m above sea level. From the perspective of woodland development, however, this region is not so homogeneous. An important difference within the uplands during the last 1000 years was the different spatial distribution of spruce and beech, respectively. Both species are competitive and shade tolerant, but with different climatic preferences. As mentioned above, spruce is a late immigrant from north to south and did not reach the southern and western parts of the uplands until a century ago.77 During much of the last 1000 years it was restricted to the northeastern parts of the uplands. Beech is also a late immigrant in a Holocene perspective but expanded from south to north. It reached the southernmost parts of the uplands in the fifth to tenth centuries and slowly expanded northward. Even today beech woodland is much restricted to the southern and western parts of the uplands with only scattered stands further north.78