Contested urban commons: mapping the transition of a lake to a sports stadium in Bangalore

: Urban expansion is a global phenomenon during which many common spaces, often with complex histories of governance and stewardships become redefined within prevailing notions of urbanity. However, such commons often pose challenges that result in conflict with respect to their use, management, and ownership. In this paper, we use the example of a lake in the South Indian megapolis of Bengaluru (formerly Bangalore) to look at different changing notions of urban commons pictured against a backdrop of rapid urbanization, migration, and landscape change. We look at conflicts at each period of change and argue that many of these have shaped the landscape of today and perhaps may be responsible for current notions of ownership associated with the landscape. We combine landscape change analysis through geospatial means along with official archival records, oral narratives, and secondary information sources to describe gradual loss of an urban commons. We then pose that knowledge of historical contexts of access to ecosystem services, exclusion, conflict, and the mechanisms of conflict resolution around urban commons can help understand trends in contemporary management of commons. This knowledge would help shape more equitable and ecologically robust policy frameworks that govern these vulnerable resources. Acknowledgement: This paper from The authors would like to express their gratitude to the Karnataka State Archives and the Mythic for allowing us access to their records. We thank Smriti Srinivas for her valuable comments during the initial of this study; Seema Mundoli and Meera Iyer for helpful information that enriched our contextual understanding of the history of Sampangi lake. Many helpful comments received from previous presentations of this and and the Students went a long way in improving our work. We also for their helpful comments in earlier of this work. Funding for this study was


Introduction
We live in the age of the Anthropocene, an era in which landscapes are predominantly urban or urbanizing (Biermann et al. 2012). A sizeable fraction of that urban population is likely to be concentrated around cities in the developing world (UN World Urbanization Prospects, 2014 revision). Given that the urban landscape is likely to be even more pervasive in the near future, it is very important to study the impact of urbanization on the commons.
Urbanizing landscapes in many growing cities, including in India, are home to a wide variety of ecological commons such as lakes, sacred groves, and temple tanks (Agarwal and Narain 1997; Mundoli et al. 2015). Often home to rich biodiversity, these commons hold significant cultural and traditional significance, and have been managed as commons for long periods of time (Brown 2006;D'Souza and Nagendra 2011). Urban commons are especially vulnerable to threats including pollution and conversion into other land uses (Sudhira et al. 2007;Gidwani and Baviskar 2011). Many ecological commons have lost their identity by transforming into other forms of land use following urbanization (Nagendra and Ostrom 2014).
Despite their importance, the transformation of commons following urbanization is poorly studied (Nagendra and Ostrom 2014). This lack notwithstanding, there is an even lesser engagement with the influence of historical events on contemporary use and management of commons (Johnson 2004). Research has shown that between 1985 and 2005, scholarly literature on commons that provides a historical perspective has been very limited (van Laerhoven and Ostrom 2007;De Moor 2012). This "poverty of history in commons" can have important consequences in contemporary policies that influence the governance of common pool resources (Johnson 2004). For example, an ahistorical approach could result in the misreading of the actors associated with a commons and ignore or alienate its users, making the resource vulnerable to many threats (Fairhead and Leach 1995).
Recognizing these gaps in research surrounding urban commons, this paper adopts a historically contingent approach to study the case of a transformed commons (a freshwater lake) within the south Indian megacity of Bengaluru (formerly Bangalore). By 'transformed commons', we imply that the resource no longer possesses any character that distinguishes it as an ecological commons. We trace the history of transformation of this landscape between 1885 and 2015. In so doing, we add to the limited literature on historical studies examining changing perceptions around a common pool resource (Valk and Ewald 2013). In addition, by providing a historically contextualized understanding of transformations in common pool resources and common property relationships around a lake, we seek to address the "poverty of history" in the commons literature (Johnson 2004). We examine the importance of historical transformations in contextualizing and understanding contemporary political issues influencing the governance of surviving urban commons (lakes) in Bengaluru today.

Study area
The south Indian city of Bengaluru forms a useful location within which to examine the social and cultural contestations and transformations of commons. Established as a city in 1537 AD, Bengaluru is well known as the software capital of India (Sudhira et al. 2007). Lacking access to large rivers, the city was built around a networked system of storage reservoirs -tanks or lakeswhich provided water, to large parts of Bengaluru until the early 20th century (Annaswamy 2003;Sudhira et al. 2007;BBMP 2010). The importance of lakes for the city can be gauged by the fact that Bengaluru was also called 'kalyananagara' (city of lakes) (Nair 2005). The provision of alternate sources of water beginning in the late 19th century, coupled with pressures of a growing city and the demand for land led to conversion of many lakes into residential sites, malls, bus stands, and sport stadiums (Nair 2005). Many remaining lakes were reshaped from erstwhile commons (used for provisioning ecosystem services by traditional ecosystem users) into public spaces (accessed for nonconsumptive ecosystem services such as recreation) reflecting the priorities of wealthy urban residents (D'Souza and Nagendra 2011; Unnikrishnan and Nagendra 2014a).
In the sixteenth century, Bengaluru witnessed the large-scale migration of a community of horticulturalists -the Vannhikula Kshatriyas -from the adjacent state of Tamil Nadu (Srinivas 1999). Later known for their efforts in landscaping prominent parks in Bengaluru, this community was responsible for the initiation of the annual nine-day festival of worship -the Karaga in the city. Celebrated even today, this festival is centred on rituals conducted at a series of water bodies within the city: three of which have been replaced by buildings (Srinivas 1999). This study revolves around one of these lost lakes -the Sampangi lake, which carries with it a heritage of heterogeneity, conflicts and politically oriented resolutions between different groups of people. Each of these groups viewed the lake through lenses shaped by their social, cultural, and ecological dependencies on this former urban commons.
Located in the heart of the city, the Sampangi lakebed now consists of a roughly triangular patch of built up land containing the Sri Kanteerava stadium. This building, Bengaluru's main sports stadium, was erected in 1946 (The Cottonian 1946) and further modified in the late 1990s (Heitzman 1999). The lake is believed to date back to the creation of the city of Bengaluru by a local chieftain Kempe Gowda around the late sixteenth century (Nair 2005;Samana and Gopinath 2012).
In the colonial period, Sampangi lake represented a transitional area located between two distinct zones of Bengaluru -the colonial Cantonment and the Indian city or Pete (Srinivas 2001). These two regions in the heart of Bengaluru were managed under the dual jurisdiction of the British Crown and the native Wodeyar Dynasty of Mysore (Rice 1897;Gist 1957;Vyasulu and Reddy 1985). Sampangi lake, by virtue of its central geographic position, defied all conventional separating boundaries between the Cantonment and the Pete (native city). Populations from both sides of the divide had clear stakes in the appropriation, management, and governance of this resource ( Figure 1).
Adjacent to the lost lake, in a low-lying depression to the south-west, is an area called Sampangiramnagar comprising of a middle class residential layout adjacent to two slums. Three major groups of long-term resident communities can be found here, intermixed with recent migrants. The first and earliest group of settlers are the Vannhikula Kshatriyas (or the Tigalars) (Srinivas 2001), whose services as horticulturists and vegetable growers were highly regarded by the Mysore government (Government Press 1949). The second group of long-term residents in this area are the 'Devangas', a class of migrant weavers who have been in the city for centuries. They moved to the area near the lake in the early twentieth century. Around the early 1950s, the area called Sampangiramnagar was developed into a middle class housing project, containing a diversity of people from different parts of the country. At the same time a government led resettlement of slums from a nearby location was also undertaken. This community of Scheduled Castes and Scheduled Tribes form the third group of long-term residents (Bangalore Development Committee 1954). In addition, other communities including Vokkaligas (agriculturalists) and Vaishyas (merchant communities) also settled around this lake. The lake, due to this temporal and social heterogeneity, formed an interesting case to study the diverse perceptions of various communities, each of whom have had different perceptions, interactions and property rights associated with the lake commons.
Mapping the transition of a lake to a sports stadium in Bangalore 5

Methods
We utilised a mixed methods approach, combining information from historical and current maps of the city with archival records and oral histories. This provided an integrated geographic and socio-politico-cultural understanding of contestations and transformations of the Sampangi lake over time ( Figure 2). Historical information from various sources was integrated and examined both for convergent narratives (which would triangulate information) as well as divergent narratives (that would lend voice to undocumented stories). Both of these threads were examined to drive a story of transforming commons which bears influence on the contemporary time.
Maps of the city, from 1885 to 2014, provided an important source to map changes in the landscape. Maps are indicators of terrain and geographical features, providing a spatial representation of conversions of land within and around the lake. Yet maps often portray an image reflective of both the purposes for which the map was originally drawn as well as its dominant political orientations (Bassett and Porter 1991).
A map of Bangalore Cantonment and its Environs published in 1885, consulted from the records of the Mythic Society of India at Bengaluru, provided our earliest point of reference. A Survey of India toposheet from 1935, obtained from the Indian Institute of World Culture in Bengaluru, was the next map used. Survey of India topographic sheets of scale 1:25000 were obtained from the Survey of India's Bengaluru office, providing information for the time period around 1973. For current information from 2014, publicly accessible imagery from Google Earth was used. Maps were spatially overlaid and registered, and land use was digitized using Arc GIS. Information on land use change was related with information from archival and oral historical material (described below).
Interpretations of maps were supplemented with the analysis of archival records (Okihiro 1981). These gave us information on the perspective of the government on the utility of the water body and its surroundings, and in some cases described conflicts around the use of the commons. Archival records described the most politically compelling representations of the water body in each time period. However there are limitations arising from complete reliance on archival material. Written records, prepared in an environment where literacy was not the norm and coming largely from official sources, represent documentation influenced by positions that were clearly reflective of their time and environment (Okihiro 1981;Thompson 2000;Portelli 2010). Archival information was collected from the holdings of the Karnataka State Archives in Bengaluru. This information was supplemented with secondary records from books, websites, and other written material. A third source of information came from oral histories of local residents. These narratives conveyed a picture of perceived landscape change, nostalgia, and a sense of place and identity among the people. They also portrayed histories of displacements and migrations from and to this landscape. The different meanings with which diverse social groups imbibed the commons was obtained. Changes in dependency on lake resources, cultural dependencies along with social and cultural continuities in traditional uses of the commons were thus charted. Oral histories can be biased by cultural perceptions, or influenced by predominant memories of specific people or communities. Yet they offer valuable insight into the meaning that specific landscapes hold for individuals, and the communities they represent (Gold and Gujar 2002;Stedman 2003;Goodall 2008).
Given the shortcomings of relying completely on any one source of data, we employed a method which combined insights from all these sources in order to gain a more holistic picture of landscape change.
We conducted 45 open-ended oral history interviews with elderly members of long-term resident communities between November 2014 and January 2015. A snowball sampling approach was followed to conduct the interviews. We asked respondents a range of questions that covered their perceptions of landscape change and the impact of conversion on livelihood and cultural uses of the resource.
Based on a broader understanding of transformations around the lake (Table 1) we organized our data on historical change into the following time periods: a) Before 1900 b) 1900 -1935 c) 1935 -1973 d) 1973 -present Using information from oral histories and archives, we traced changes in the ecosystem services derived from the lake at each time. We identified connecting and divergent narratives from the oral histories collected, and used these to relate the past to contemporary trends in the governance of common pool resources.

Time period 1: before 1900
In 1885 the lake was much larger than the boundary occupied by the stadium today. It extended to cover parts of the current day boundaries of Cubbon Park ( Figure 3). The lake was surrounded by open land. Two sides of the lake were bordered by roads, and there were a few buildings in the vicinity. The area to the east of the lake was wooded: that to the west was cultivated. There were two outflows to the lake. Water from Sampangi lake reached the Cantonment by means of feeder channels that connected it with the neighbouring series of Millers tanks. These in turn provided water to the Cantonment. The lake supported numerous horticulturists and farmers, as well as Pete residents. Water was drawn directly from the lake, or from the many wells surrounding the lake (Srinivas 2001). The area upstream of the lake was used for the cultivation of a dry local millet (ragi, Eleusine coracana), while the area downstream was used to cultivate flooded paddy. Wetlands around the lake were acquired in 1898 for the playgrounds of a prominent Jesuit educational institution. (http://www.stjosephsindianinstitutions. com/?page_id=9). While wetlands around the lake were being used for agricultural activities, state led challenges to traditional lake-based activities such as brick making or the sinking of wells were issued in the form of repeated orders and revocations ( Table 2). Prioritization of aesthetic and recreational utilities offered by the lake had begun. Traditional activities such as collection of water were permitted only in conditions of drought and famine.
Elderly members of the Vannhikula Kshatriya community remember the area surrounding the lake, depicted as open in the 1885 map, to be fertile farmland primarily belonging to them. They grew flowers and green leafy vegetables on this land, crops that this community cultivates even today. Other accounts corroborate this indicating that in the 1870s, the area around the lake was known for its beautiful gardens cultivated by the Vannhikula Kshatriyas (Srinivas 2001). The priest of an old temple that predates the colonial period belonging to the community of Kurubas (originally shepherds) asserted that the village of Sampigehalli, associated with the lake, was in existence prior to the founding of the city in 1537 by Kempe Gowda.
Oral narratives suggest that the area was populated with temple tanks of which only one exists today. Others were eventually built over. Local residents described a wide well which provided groundwater to most of the village, which had twenty pulleys to draw out water.

Time period 2: between 1900 and 1935
By 1935, part of the lake seems to have become a playground. The lake disappeared, leaving only a small rectangular enclosed tank, all that remains of the lake today ( Figure 4). The boundary of the Cubbon Park was demarcated. The land occupied by the Cantonment had increased road networks, while the area marked as fallow land reduced. A number of public utilities in the form of schools, churches and hospitals were constructed by this time along with a golf course and the Bangalore Brewery (Nair 2005).
The period between 1900 and 1935 was one of intensive change, both in the social and ecological fabric of this landscape. It was also a period of intense   Mapping the transition of a lake to a sports stadium in Bangalore 13 conflict between different actors (Table 3). There were tussles of power between institutions and residents of the British Cantonment headed by the Civil and Military Station and of the native city or Pete, governed by the Mysore kings.
The lake began to transform from a commons to a recreational and aesthetic space. Demands were made for dredging the lakebed, to increase its water storage capacity and prevent the bungalows and the brewery adjacent to the lake from flooding. British regiments also asked to drain a portion of the lakebed so that they could play polo. Native horticulturalists resisted such demands, instead asking     He then instructs the Sanitary Engineer to ensure that the "channels are not interfered with beyond the clearance of silt".

The decision arrived at in these conferences was to a) To drain the portion of the Sampangi Tank to the north to RL 42. b) To retain water in the Southern Portion at its present level of RL 44.
Estimates were prepared and work sanctioned accordingly.
We did not procure evidence of any further correspondence concerning the tank till about 1935.  for increased water supply to irrigate their crops. A petition signed in Kannada (the language spoken within the state of Karnataka) by 50 horticulturists in 1903 asked the Dewan of Mysore for increased water supply from the lake to support agriculture. The eventual outcome of these conflicts was the draining of the lake, indicating that these voices were not given prominence in the decision making process. The archives also record debates between the colonial administration and the Mysore kingdom about the uses of the lake. An instance where this dynamic is evident is the response to Indian petitioners' demands for extra water that has been described earlier. Here, the Mysore rulers favoured the subaltern voices as opposed to the view taken by British administration. This was however done in order to sustain the revenue which horticulture brought into the kingdom.
Oral narratives from the Vannhikula Kshatriya community describe an abundance of water and fish in the lake in this time period, so much so that excess fish were used to provide manure for crops. Local residents described restrictions imposed on the entry of cattle for grazing, while people were permitted to use the lake for recreational activities such as swimming. Oral accounts also indicate that many farms were acquired during this period to build houses, corporation buildings, and roads.
Other communities began to migrate into the area. The earliest communities to settle here were Devangas (weavers), followed by Vokkaligas (agriculturalists) and Vaishyas (merchants). Restrictions were imposed on sinking of new wells in and around the lake. Yet there continued to be ten large wells and four temple tanks in the area. These supplied enough water to meet the needs of the people.
The Devangas describe the area surrounding the lake as a fertile agricultural landscape comprised of coconut groves and fruit orchards. The Kuruba priest at the temple described earlier recalled that the upstream area of the lake was drained in the colonial period and used as a playground (agreeing with the landscape as depicted in the map of 1935). Oral accounts also described a decrease in the level of water in the lake due to the construction of schools and hospitals in the vicinity. With the change in the agricultural landscape surrounding the lake the Vannhikula Kshatriyas began to pursue alternative professions and migrated away from the neighbourhood.
Oral accounts also associated the lake with important religious ceremonies. The most prominent of these was the Karaga (described earlier). In October, a festival called Ganga Pooja was conducted at the lake, where numerous lamps were lit and set afloat in the water. Oral accounts also describe a fishing community (Bastharu) who used to live on the banks of the lake, and have since migrated away.

Time period 3: between 1935 and 1973
Many important events occurred between the year 1935 and 1973, the most notable of which was the attainment of Indian independence in 1947. The Sampangi lake also transformed into the Kanteerava Stadium around this period in time. This change impacted the social fabric of the landscape (Table 4). Almost all the area surrounding the lake (apart from Cubbon Park) was built up by 1973 ( Figure 5). A small rectangular remnant tank within the stadium was the only remnant of its original identity as a water body and an ecological commons.
In 1937, a carnival ground was organized on the bed of the lake. A proposal was made to convert the lakebed into a stadium (Government Press 1949). This proposal followed similar transformations of many lakes in Mysore state into civic amenities (Government Press 1949). By 1946 the stadium was completed. The All India Olympics as well as the Karnataka Olympics both held in 1946, were conducted at this venue (The Cottonian 1946). In 1957, cattle fairs were held in the lakebed suggesting that at least a part of the lake survived as an open field (Venkatarayappa 1957). The social milieu around the lake experienced a trend towards increased diversity with many people from all over the country migrating into these parts and setting up their trade. The area formerly occupied by the farms now became a residential layout called Sampangiramnagar. A slum was established in the vicinity as part of a resettlement program initiated here by the City Improvement Trust Board (CITB) around the late 1950s (Bangalore Development Committee 1954).
The Vannhikula Kshatriyas recall the land as being cultivated till the late 1940s with paddy, arecanut (Areca catechu), coconut (Cocos nucifera) and millets using water from the now diminished lake for irrigation. Others remember grazing their cattle along the banks of the lake, while protest marches for Indian independence took place on the roads bordering the lake. Lake water was polluted, and most people refrained from swimming, drinking or using the water for domestic needs. Instead, water was obtained from the numerous wells that still dotted the landscape. Slum residents however remember using the lake for washing clothes and vessels until the late 1950s. The Vannhikula Kshatriyas described escalating conflicts with the CITB about land ownership and use.
The Devangas (weavers) had by this time distanced themselves from the lake. They remember the lake as being used only for two purposes -for open defecation and for irrigating a few crops. They recalled one instance in the late 1940s when during a period of intense rainfall the lake bund breached and some houses in the vicinity were flooded. The Devangas and the slum residents recall that by the late 1940s the lake had become a social hazard, a place where few people ventured. They recalled a few instances of drowning that occurred in the lake. Some accounts described parcels of land around the lake owned by rich landlords. When conditions changed they simply sold their workers (indicative of bonded labour) and migrated to other areas. The people who were most affected were those directly dependent upon the lake for their livelihoods. While some families of the Vannhikula Kshatriyas received higher education and took on professions such as that of lawyers and doctors, many others migrated away from the area to other lakes and continued as horticulturists. Figure 6 shows the present day landscape around the lake.

Time period 4: between 1973 and 2014
We now observe a further increase in urban cover and a greater density of road networks. The small rectangular portion conserved as a water body continues to be a location of importance for the annual Karaga festivities, now bordered by a small temple and lawn within the premises of the stadium. The stadium itself has been renovated and a state of the art indoor stadium added (Heitzman 1999). Outside of the Karaga festivities the Sampangi tank was further relegated into the unknown as far as the rest of the city was concerned.
The social scenario is very diverse today with pockets of Vannhikula Kshatriyas and Devangas existing alongside communities from all over the country. Many native residents sold their land and migrated elsewhere. A few others have remained and taken on alternative occupations such as running a cooperative society to help members of their community. Some have retained pockets of their original farmland, on which they have constructed their homes. Slum residents describe that urbanization which characterizes the landscape has somehow passed them by. They believe that urbanization is a result of the fancy stadium that has come up but that it has not improved prospects for them. One slum resident who rears cattle recalls how their occupation has progressively seen decay, both in terms of resource availability as well as social status. Obtaining fodder to feed the cattle has become a more difficult task over the years. They describe a time about fifteen years ago (around 1999) when the area now occupied by the temple and lawn next to the stadium was filled with fodder grass. Through a tender based process, pastoralists obtained rights to cut fodder. This process was stopped and the growth of fodder grass replaced by an ornamental lawn.
While one well and one temple tank still exist, others have been built over. The reduction in the importance of pastoralism has led to a decrease in social status with interviewees indicating that it has become more difficult for young men to find brides. Slum residents are also dealing with the potential challenge of relocation from this area.
Within the Vannhikula Kshatriya community who live here almost no one practices agriculture. However, they retain a deep sense of connection to the landscape as it is sacred to their cultural beliefs. The community campaigned for exclusive use of the tank and the compound it occupied. Some members of this community raise fish in the tank and harvest the fish for personal use, sharing a small portion of their catch with a flock of Brahminy kites (Haliastur indus) that circle overhead.

Discussion
Drawing on transformations in the use of Sampangi lake as a commons, Table 5 utilises the property rights framework proposed by Ostrom and Schlager (1992) to describe changes in the property rights bundles associated with the lake over time. Categories of ecosystem services accessed by these different actors are also elucidated.
It is clear from this table that there have been many actors in this landscape and that their user rights have been temporally dynamic and heterogeneous. Some actors such as British residents or the Municipal Council are no longer active stakeholders in the landscape. Other users such as the Vannhikula Kshatriya community have seen the balance of power undergo shifts over time. It is equally clear that users deriving provisioning and certain forms of cultural ecosystem services like spiritual services have always been on the lower rung of the power spectrum. Over time, their relations and requirements of the lake decreased and they became dissociated from the resource. Following its conversion into the stadium, the lake stopped being a provider of important ecosystem services to its dependent communities. This as we demonstrate above, has alienated these communities from their commons. The lake thus transformed gradually from being perceived as a sacred resource, important for local livelihoods into one  *A=Right to access, P=Right to appropriate, M=Right to manage, L=Right to alienate (Ostrom and Schlager 1992). **NA=Actor not present in the landscape at that point in time. ***None=Actor present but does not possess usufruct rights to the resource.
that was unpopular and hazardous. Physically too, it lost its identity as an urban ecological commons, transforming instead into a built space evoking the same powerful notions of aesthetics and recreation that dominated the colonial ethic. Except for a brief time every year when the small remnant water body forms part of the Karaga worship, the lake has shrunk in public memory. Today, this space only reminds Bengaluru of its former identity as a water body during torrential rainfall, when the entire stadium floods over. This historical picture remains representative of contemporary trends in the management of lakes within the city and more broadly, of commons across cities in the global South. Many lakes within Bengaluru and elsewhere are polluted and fast losing their connection to the user groups that once accessed these commons (Agarwal and Narain 1997). Some of them have already been converted to temples, malls or bus terminals. Efforts are being made to preserve lakes through modes that foster further inequality. The aesthetic and commercial priorities of privatized lakes have served to keep out provisioning and cultural ecosystem service users (Unnikrishnan and Nagendra 2014b) in some parts of the city. In other places, the increased gating and restrictions of access to lakes actively keeps out communities engaged in livelihood uses. This again prioritizes aesthetic and recreational uses over extractive uses of the resource. Measures such as these have further alienated traditional users preventing them from engagement with the commons that they once actively maintained. Similar trajectories of alienation, exclusion and acquisition have been described in other parts of the country for example in the cities of Delhi (Mann and Sehrawat 2008) and Mumbai (Parthasarathy 2011).
Engaging with the history of a resource such as Sampangi lake -engaging with its "historical political ecology" -has multiple benefits (Offen 2004). First, as has been demonstrated in this paper, a historical account helps to document the tangible and intangible uses of, and meanings attached to the commons. Second, it helps to draw out the political nature of interactions between society and the governance of commons. Third, it provides a means of documenting changing perceptions of the utility of a commons because of its changing property rights. In so doing, alleviating the "poverty of history" in the commons (Johnson 2004) would aid in identifying those actors whose user rights are in danger of exclusion. This can provide a more informed and inclusive frame for policy directives governing (De Moor 2012) commons in the Anthropocene. Agarwal, A. and S. Narain. 1997