4 Gathering routes in the study area (Base map data from LMÍ) (fi gure by Oscar Aldred) 

Context in source publication

Context 1

... at either end of this temporal chain was an earlier emphasis on discrete land ownership in which only a few farmers grazed their sheep on the best pastures, versus a later communal approach to grazing. The switch was perhaps a gradual response to changing conditions in the constitution of the community, but it also stemmed from environmental change that was occurring to the land as vegetation became increasingly stressed (e.g., Simpson, Guðmundsson, Thomson, & Cluett, 2004 ). Arguably, the commu- nity-wide system of sheep farming that was used from at least the eighteenth century onwards was probably an environmental response, but it may equally have had come about for social reasons, or through economic and geographical changes in wool production. The nineteenth century was also a time of gradual increase in sheep numbers alongside increased industrial practices associated with textile production in the urbanized townships of Iceland, such as Reykjavík. In any case, in the nineteenth century, farms shared the grazing pastures, and as a result the workfl ow in the practice needed to be managed collectively. Committees of community elders or respected farmers were set up, and minutes of committee meetings were recorded. One of the results of this organization was a pre-calculated sheep tax that was based on a strict sheep-to-farmer ratio during the gathering process. In Skútustaðahreppur the tax can be estimated to have been 1 sheep to 0.15 Ikr (Icelandic Krónur ) or approx- imately 50 sheep to 1 gatherer, derived from looking at a neighboring community called Helgastaðahreppur and calculating an estimated cost of one sheep as a tax value for gathering (Sigurjónsson, 1950 ). Therefore, the consequences of establishing a committee that put into effect an increased organization of sheep-farming practice was the production of a highly regulated system of grazing, gathering, and sorting. Therefore, the terms of the relation that the environment afforded and the sociality between farmers and sheep were radically different at opposite ends of the temporal chain. Furthermore, it is likely that movements were different. In later periods the sheep farming involved larger “collectives” of humans and nonhumans. It is likely, however, that knowledge of the landscape in relation to previous gathering operations was retained through the continuity of specifi c aspects of much earlier, traditional forms of practice. This knowledge would have been reproduced and passed between generations of farmers, so that judgments about where to graze sheep changed alongside the actual movements involved. One particular problem in Skútustaðahreppur was the increasing rate of land degradation and a loss of vegetation from both over grazing in the past and worsen- ing environmental conditions during the late eighteenth and nineteenth centuries. These landscape environments were subject not only to dramatic changes between seasons that resulted in the gradual limited grazing of sheep in the southern grazing area but also to much longer-term changes across generations. Alterations in practice were not simply a response to recognized changes but rather part of a more gradual and emergent series of responses that were small at fi rst but which had a long-term effect on the shape and direction of practices. For example, in nineteenth- century Skútustaðahreppur, there were two grazing areas divided into a further nine zones. And in each area there was a sorting fold. Historical practices such as gathering along paths and selecting the areas within which sheep grazed and were gathered were cumulative. There would have been some resilience because of their operational “value” that derived from previous operations, but they would have been creative and innovative solutions that were worked out in the fi eld as the gathering took place. For example, locating sheep in the grazing areas was dependent on where they had been released and where there was suffi cient vegetation for grazing. In this way, responses to specifi c problems can be seen as either quick or slow , in the fi eld or between generations; for instance, recognizing environmental markers in a landscape was a quick response. On the other hand, slow responses accumulated such that an earlier practice and another, separated by several generations, may have been largely unrecognizable. It may very well be the case today that earlier practices have only a few, resilient vestiges left in the modern practice; earmarks were one of these, as were generalized routes, and to some degree, the sorting fold. In this respect, earmarks, paths, and sorting folds were outcomes of the relational entanglement between the resilient forms of the practices used at different times and the creative mediation of new forms, such as those infl uenced by the landscape at given moments of the gathering operation. There were a total of fi ve gathering paths, divided into a number of routes, in Skútustaðahreppur (Fig. 4.4 ). The timing during a single operation was carefully choreographed so that the multiple parties involved met together at the sorting fold. Each path into the main grazing areas took variable amounts of time (between 2 and 5 days) to traverse, and different numbers of gatherers traveled along each path. Archaeologically, the paths can only be generally reconstructed because the movements of sheep within the grazing area during each gathering would be different, and consequently each path taken had to accommodate changes in grazing patterns. In some respect, there was a sense of going along and remembering the way, but at the same time there was real sense of negotiating the path on the hoof. Like the quick and slow responses, there was a potential path, predicated as a virtual path before the gathering, while the actualized path took into account the structure provided by the virtual one alongside the conditions and settings during the moment of moving. What was predetermined was the eventual destination, and this had an effect on the dispatch and division of gatherers. But how the gathering was opera- tionalized was dependent on the comingling and entanglement of multiple forces, such as the state of vegetation cover, where the sheep were grazing, how many gath- ers there were, and the location of the sorting fold. Characteristic of all gatherings is the series of staggered movements. While there would have been long-term differences, many of the same places within the grazing areas served as structural anchors with respect to movement. Such paths and places were also places of knowledge and wisdom (after Basso, 1996 ) within the group, where specifi c acts were performed during the gathering process. Mountains, rivers, vegetation areas, and other topographic features guided directionality as well as hinted at changes that needed to be made in the gathering practices the following year as a result of differential vegetation cover, glacial melting, and snow pockets. The environmental markers were a form of observational knowledge learned not just by being in the landscape but by moving through it, a process that helped shape collective memory (Connerton, 1998 ). In this way, it was feasible to predict where sheep could be grazing, and in this way a virtual line became a kind of mental map. But the environments, as I have already said, were susceptible to fl uctuation that was diffi cult to observe year by year. In the nineteenth century, grazing areas that were once rich were depleted, so sheep were grazed in other locations. Thus, what an environment offered was liable to change, and farmers needed to be fl exible in light of these constantly materializing contexts. Not only were paths liable to change their routes through the landscape but also the eventual destinations at the sorting folds. Alongside the complexity of these intersecting moving elements, the sorting fold also needs to be considered. Throughout the late nineteenth and early twentieth centuries, the sorting folds for each grazing area changed location several times. For example, four different sorting folds were used in the southern grazing area called Suðurafrétt. The earliest sorting folds (pre-nineteenth-century ones) were located in the grazing area itself. From the late nineteenth century onwards, the sorting fold moved either to the margins of the settlement zone or into it (e.g., close to Lake Mývatn). Thus, a gradual movement in the location of the sorting fold from grazing areas to farm zone seems to have been taking place in Skútustaðahreppur. Driving this change were several interrelated forces, from inside the community as well as outside of it—population numbers of humans and sheep, climatic and environmental effects on vegetation, relations between the sorting fold and the location and use of the grazing areas, and the historical ties that particular farms and communities had to grazing areas. The sorting folds were not only places where the gathered sheep were divided into their respective farms on the basis of their earmarks, but they were also events in themselves. Each farm had its mark involving a combination of different features. Furthermore, each farm, or a group of farms, had its own chambers within the sorting fold (Fig. 4.5 ). But like the earmark, the sorting fold had multiple meanings attached to it, with a number of different fl ows operating within it. In many ways, sorting folds ( réttir ) were constituted by their fl ows not by the solidity of their materiality (cf. Aldred, 2006 ; Edgeworth, 2011 ). In many ways, it was the fl ow of movements by sheep and farmers that constituted the sorting fold not just the spatial arrangement of its architecture. This temporalization of space—in giving a location its set of events—is reminiscent of Tuan’s “space is movement and place is pause” ( 1977 ). The sorting fold was also a locale that was full of action in events that can be understood from multiple ...