A Feudal Way to Gentrify? The current understanding of gentrification and changes of social-topography in a medieval and early modern town

Gentrification is a current and often debated concept that concerns social changes in our cities. The concept relates to a development whereby areas earlier inhabited by less wealthy social groups are taken over by middle-and upper middle-class residents. In the discussions of these changes, two perspectives have dominated. Representatives for the consumer perspective argue that gentrification occurs as a result of consumption preferences in the middle class. Representatives of the producer perspective argue that inner city areas are gentrified as a result of the movement of capital. In the article it is discussed whether it is possible to use the concept of gentrification, and if the conflicting perspectives can be tested, in a non-capitalistic setting. The case-study focuses on the changes in the social topography of the town of Malmö, at the time the second largest town in Denmark. The main issues investigated are whether these changes were produced within the feudal structures, how consumer preferences and agency interacted, and the relations between agency and structural constraints.

Today we live in an urban world. Much of the societal resources are centralised to urban centres. The economical and cultural capital that is currently said to be globalised, is actually fixed within urban space. We can also see that we are in a demographic urbanisation; people are moving to cities, abandoning the countryside and small towns.
In almost all of these different types of cities, we can see struggles in space between social groups. A common phenomenon is when areas inhabited by poor people and/or the working class are refurbished by an influx of private capital and middle-class homebuyers and renters (Smith 1996). The proper concept with which to describe these processes is gentrification.
Gentrification is a common and debated concept with regard to more or less intellectual and literate discussions as well as everyday conversations during coffee breaks. (A search of the concept via Google revealed about 66,400 hits on the Internet.) The concept consequently describes a phenomenon within urban space that is recognised as controversial for a great number of people today.
However, similar changes within urban space have occurred earlier, and a reasonable question is whether these examples can be compared to the developments of today. Another interesting aspect of this issue is whether we can gain some understanding that can contribute to our current knowledge of changes in urban social-topography. Also, is it possible, in the proper context, to apply the concept of gentrification within archaeological and historical analysis of past societies? As such, it could be used as a means to identify social relations and conflicts within archaeological records. In the following, I will try to discuss the definition of gentrification and adapt it to a feudal context. The empirical analysis is based on a case-study of housing and social topography in the medieval town of Malmö, situated today in the southernmost part of Sweden. However, during the Middle Ages and early modern times it was the second largest town in Denmark.

THE CONSTRUCTION OF THE CONCEPT OF GENTRIFICATION
The term gentrification was coined in the 1960s by a British sociologist, when investigating social changes. Ruth Glass' definition implied that many of the working-class neighbourhoods in London were "invaded" by the upper and lower middle-class: Shabby, modest mews and cottages -two rooms up and two down -have been taken over, when their leases have expired, and have become elegant, expensive residences... . Once this process of gentrification starts in a district it goes on rapidly until all or most of the original working-class occupiers are displaced and the whole social character of the district is changed (Glass 1964in Slater 2002:3). Gentrification as a phenomenon was thereafter identified in numerous urban centres, foremost in capitalistic western cities (Slater 2002). The explanations of gentrification, according and prior to the urban geographer Neil Smith, can be divided into economical and cultural categories (Smith 1996:51f).
The cultural reasons for the middle and upper middle-class to move to workingclass neighbourhoods have, according to this tradition, to do with changed lifestyles. The desire to live in single-family homes in the suburbs decreased, and was replaced by a "...trend toward fewer children, postponed marriages and a fast-rising divorce rate, younger homebuyers and renters are trading in the tarnished dream of their parents for a new dream defined in urban rather than suburban terms" (Smith 1996:51). The economical explanations of gentrification focus on the combination of increasing costs for construction-projects and the likewise increasing distance from the city-centre to available plots in the suburbs. Rehabilitation of inner and central plots in the cities is in this perspective seen as a more viable economical investment.
Smith regards this kind of explanation as an emphasis on consumer preferences and the constraints in which these are implemented. He argues that this kind of consumer sovereignty is not a definitive explanation to gentrification; instead gentrification is produced by the land and housing markets, and in its longest reach, the capitalist mode of production (Smith 1996:55f, 70). The actual pieces of land and the housing are commodities. In our society, private property-rights give the owner almost a monopoly on land and improvements on housing. The market is the basic institution for steering the transfers on land and buildings, which are regarded as "capital". This kind of capital is well fixed in space, but its value is not. Investments in capital -i.e., improvements on the built structuresare subject to influences on their value, while a piece of land does not need similar investments to retain its potential for use. These investments have very long turnover periods in physical as well as in value terms, in order to gain profit. Therefore, investments in real estate, formally after the Second World War, were done further and further from the city core, creating a demand for urban expansion (suburbanisation). The land values were at a peak by the centre of the city and were lowest in the peripheries, creating a conical form, as shown in a diagram (x = land-value; y = distance from city-centre; see Fig. 1a). With further urban expansion, land at the centre increases in value while the base of the "cone" broadens. The investments in real estate and built structures in the later phases of urban growth were primarily made in the suburban areas. At the same time, the inner city-areas decreased in value, because the built structures constrained favourable investments with high returns (in relation to the suburbs where the prices increased and the returns of investments were high). This created a "valley" in the land-value curve, between the city-centre and the suburbs. Smith interprets this "valley" as a devalorisation of the capital (the built structures) on the landed property. This devalorization produces the objective economic conditions that make capital "revaluation" (gentrification) a rational market response. Of fundamental importance here is what I call the rent gap. The rent gap is the disparity between the potential ground rent level and the actual ground rent (land value, my comm.) capitalized under the present land use (Smith 1996:87;Fig. 1b). Under these circumstances, an investor can purchase the structure, invest in rebuilding, pay mortgage loans, and finally sell the end product for a price that gives a defensible return -the capital, not the suburbanites, moves back into the inner city (Smith 1996:88). In this way, Smith can show through a number of examples that gentrification is a structural product of the capitalistic mode of production.
Smith's work, and foremost the rent-gap theory, has been regarded with quite a lot of criticism (and of course also a lot of positive critique). The urban geographer Tom Slater has identified four main objections to the rent-gap theory (Slater 2002:6f). The first argument is that the theory does not identify the gentrifiers. The process involves persons, and it is vital to know something about them in order to understand gentrification. The second objection is that there is a problem with applying the rent-gap theory in empirical research. Some cities do not have an industrial past, and some do not have such deep disinvestments in the inner cities as in the US; still, there are examples of gentrification. The third objection, is that devalorisation of a neighbourhood can be intercepted by gentrification. Thereby, the creation of a rent-gap is not a condition for gentrification, and the inner city and the suburbs do not have to be treated as a systematic whole. The fourth and last objection is the lack of gender agency within the theory. Some researchers have been able to show the role of gender in gentrification processes.
The conflicting perspectives -consumption versus production as an explanation to gentrification -are, however, not inseparable entities. In one sense it is the "good old question" and discussion about agency and structure. However, since the late 1980s the literature on the topic has tried to overcome the differences (likewise within the discussion about "agency and structure"; the sociologists Bourdieu and Giddens, among others, have tried to create meta-theories to connect the debated entities), and applied both perspectives in the research (Slater 2002:12f). It is quite reasonable to assume that the conditions for gentrification are produced by movement of capital and the structure of the real-estate and housing-markets. But the actual decisions and the needs to gentrify have to do with disposition of lifestyles and taste. People identify with the use of space and the architecture of a particular area, with its connotations to persons, groups and history. These more "cultural" explanations are, of course, intersected and "fuelled" by speculations made by capital owners. An interesting point of the following case-study is to consider the relations between the perspectives of production and consumption in a non-capitalistic setting, emanating and developed during feudal circumstances.  Chicago (after Smith 1996, fig. 3.1). 1b. The devalorisation cycle and the evolution of the rent-gap (after Smith 1996, fig. 3.2).
Another important issue is the spatial use of a city. One factor to discuss is that city-centres are culturally determined. In part, this should be regarded as centralisations of societal (political administration, courts, etc.), commercial (markets, stores etc) and other communal (churches, associations, etc) resources to bonded space; and in part it is dependent on the personal preferences of the individual or corporation (and, in our system, their influences over the "market"). The point is that city-centres are not historically constant pieces of land; city-centres are structured and changeable domains of human action. A centre can also multiply, which means that a city can have more than one focus. In a longer historical perspective, it is problematic to assume urban growth, and difficult to analyse "value", from a fixed centre. Another important factor is changes in the infrastructure of a city. Paths, roads, public spaces, subways and so forth, force patterns of movements that create different opportunities for "profitable returns" and cultural preferences (see Smith 1996:81). Investments in infrastructure can, I must emphasise, change the hierarchical setting of a city, and thereby the conditions for the real-estate and housing markets. To use Smithian terms, investments in the infrastructure can (or will) change the potential use for a specific piece of land, and thereby also its value. The geographers Eric Clark and Anders Gullberg have, for example, shown that, through investment in the infrastructure during the latter part of the 20 th century, Stockholm city changed the social-topographical setting of the town (Clark & Gullberg 1997). In order to understand gentrification, it is vital to understand the structure of the city. THE CONCEPT AND FEUDAL SOCIETY Both the consuming and producing perspectives in the discussion of gentrification in some way presuppose capitalistic circumstances in order to be viable. Regarding the consuming agenda, it is dependent upon, on the one hand, a differentiation of the workplace and the home. On the other hand, there has to be a developed market to supply the gentrifying consumers with suitable and buyable objects (commodities), mediated through anonymous means to transact values (money). Another important presupposed factor is the private property, with the onedimensional and almost unlimited right of ownership. The Smithian focus on the production of rent-gaps and circumstances for gentrification has a more elaborated and outspoken relation to the capitalistic mode of production. The question to be posed, then, is whether it is possible to apply the concept of gentrification to a feudal society.
The presupposed capitalistic circumstances in the discussions are important parts in the construction of the term, but they relate mainly to the means by which gentrification occurs, and not its core-meaning. As I see it, gentrification can be described as changes in urban social topography whereby areas earlier inhabited by poorer segments of the population start to be settled by more wealthy persons. Interpreted in this way, the construction of the concept is done with a reference to a given structure (contrary to a presupposed or a pronounced capital-istic one) and how it interacts with consumer preferences (agency), whereby the means are integrated into the research. This could make it possible to test Smith's observations of the dependence and bonds between "consumer agency" and structure. Archaeologically, there are also some source-critical limits to how to detect refurbishment of a neighbourhood. Therefore, refurbishment has been used more loosely, to imply changes in land-use and/or architecture.

THE FEUDAL MODE OF PRODUCTION
It is, of course, an impossible task to present a thorough description of the feudal mode of production within the scope of this article. The focus here is on the passages important to the discussion of gentrification; that is, the organisation of the landscape, the relations within the processes of production and, obviously, the outer foundations of "markets" and "prices".

The landscape
The feudal landscape was organised through regulations and privileges. Towns were centralisations and regulations of trade and handicraft to specific spatial entities (see Saunders 1991:299f). Also, the spatial limitations (walls, ramparts, moats, etc.) defined the space for a specific jurisdiction. It was a distinct border separating the man-made, ordered and peaceful world within the urban space, from the more wild and "natural" countryside. In another perspective, the towns were a part of the feudal relations in the countryside. The medieval archaeologist Tom Saunders has stated that, "Towns were institutions central to the maintenance of the rural economy, not only centres for petty commodity production, supplying peasants and lords with everyday artefacts, but also urban markets functioning as nodes in the articulation of agrarian surplus" (Saunders 1991:300). Expressed more harshly, towns were produced by feudal structures in order to control, regulate and administrate trade and commodity production. The sharp borders between urban and rural also made urban growth more restricted than in western 20 th century cities.

Household, corporation and town
A major difference was that the "home" was not, as today, a place separated from the "work". The production of commodities was, for example, done in the households and by the masters; and often the apprentices lived together with the family of the master or in a nearby house. But the deliveries of raw material to the producers in the towns were sometimes made over great distances, and through a lot of different links of middlemen-merchants. Generally, also the merchants had their "home" in the same building as their "store", or the "store" was situated at the market place. When they were away from their original town for longer periods, merchants rented houses for both commercial and domestic purposes. Of course, in relation to current conditions, these circumstances led to different patterns of movement within the towns. Our daily routine of "go to work -go home" had obviously another meaning, and perhaps it was not mentally separated at all. Without suburbanisation of a thought "consuming middle class" in the towns, and without a spatial separation of "work and home", there would probably not be any favourable infrastructural conditions to move to the "inner city". The medieval and early modern Scandinavian towns were not large spatial entities. Even with the perspective of a pedestrian, there were no great distances within the urban borders. The roads and paths had a more domestic usage in that they were proportional and more used for the needs of the households. (The households were subsistence-producers regarding daily life, not usually consuming semimanufactured articles.) Often persons belonging to the same guild, profession (i.e., corporation) or ethnic group lived together in specific blocks or along certain roads (Lamberg 2001:139f). The ideological ideal of the time was, according to the doctrine of the four estates, to differentiate societal groups (Andrén 1998:162f). The spatial ordering of the orders -the same structural principles that underlay the sharp spatial limitations between urban and rural -created a broad normative framework for what was socially acceptable regarding movement (i.e., to change place of residence) in urban space (and indeed also urban growth). Taking the spatial ordering into consideration from another perspective, social relations produced urban space and urban space produced social relations. This dialectic of structure had a conservative effect on changes in the social topography; once started and established, the structure was almost self-generated. Movements (i.e., to change place of residence) in space with retained social standards, especially by single individuals disturbing the spatial order of the predominant ideology, must have been more complicated compared to today. Relocation of corporative entities in urban space seems perhaps less likely as a logistical phenomenon, but was likely more appropriate from an ideological point of view.

Markets and prices
The markets were constituted by privileges and were spatially fixed in the urban space. The access to the markets was spatially and socially constrained. As stated earlier, towns were defined by walls, ramparts and/or moats. As a rule, certain roads or other public spaces had specific rights to house commercial activities. Physical access to the market areas in the towns was acquired through towergates, where a person had to pay a fee. The markets were to a lesser or greater degree socially organised in craftsmen and trade-guilds. In principle, in order to conduct trade or craftsmanship, a person had to be a member of a guild and/or recognised as a burgher. These spatial and social rules were stated in normative legislation and custom, and administered by the town councils, representatives of the governing feudal lordship and the different guilds. For example, trade conducted by non-merchants or craftsmanship carried out outside the towns was regarded as a serious violation of the governing principles (Lindström 1991:215).
The prices of commodities were adjusted according to the principles of the "fair price" (iustum pretium) (Fenger et al. 1982:44f;Lindström 1991:134;Huberman 1986:64f). Much too high prices were regarded as usury by the church, and constituted a serious crime according to the canonical right (Fenger et al. 1982:44). In a study of the guilds in late medieval Scandinavian towns, the historian Dag Lindström found proof of regulation regarding the limits for the highest prices (Lindström 1991:134). Prices were settled according to socially acceptable negotiations as to the utility-value of a certain commodity. There was also an expectation of stabile prices over time. Inflation was not a recognised phenomenon, even if it was an experienced fact (especially and on a larger scale during the late Middle Ages and the early modern period, regarding the effects of the Spanish silver from the colonies in America. See Huberman 1986:83 -98). Acceptable arguments for raised prices were, for example, more expensive raw materials and higher expenses in trade.
The markets were regulated and consisted of actors with overlapping monopolies, which formed the guarantee for actors to gain any kind of profit (Saunders 1991:297f). There were also social and spatial limitations to recruiting potential actors within the different markets. This narrows down further if the market is defined as persons with an ability to invest accumulated capital in real-estate and housing-markets. The settling of prices -i.e., the axiom of the "fair price" -also made the accumulation of unbounded capital, at least, a long-term project. Compared to capitalistic circumstances, it was obviously more difficult to create a fortune solely on the trading of petty commodities on a local and small-scale basis. After all, the majority of the medieval and early modern capital was bonded in rural landed estates, and profits were predominantly made through rent on the agrarian surplus of the land. The nobility and the noble mentality served as a model for the accumulation of wealth and wealthy behaviour, that is, conspicuous consumption. Great fortunes were made by town merchants, but these were not created primarily around estate-and housing-projects. During the late Middle Ages and early modern period, profits were made on monopolies on segments of the markets, defined by privileges authorised by the king. The most prosperous regional commerce during the period was cattle-trade with German towns.
There was also a lack of developed financial institutions to fund, for example, great enterprises on a short-term basis. One type of "financial" institution during the Middle Ages was the different spiritual institutions. They were of course, like others, restricted by the church's ban on loans with interest (i.e., 5 %) (Johannesson 1947:128).

THE BROADER LIMITATIONS OF THE URBAN HOUSING-MARKET
The property-concept of the Middle Ages and early modern times was quite different from the almost unlimited private ownership as we know it today. According to normative written sources, the rights to real estate in Scandinavia could be acquired through inheritance, exchange, purchase, gift and deposit (Ågren 1995:117). In principle there were two conflicting parts in the property-concept (Olsson 2002:30f).
-Firstly, there was the right to collect rent on the production of certain pieces of land and to sell these rights (dominium directum). This part can be connected with what we today call ownership of land (Fenger et al. 1982:25f). In principle, in medieval Scandinavia all land either belonged to the king/crown (the regaliarights) or to the earthly and spiritual nobility (Olsson 2002:30).
-Secondly, there was the right to use the fruits of the land (dominium utile). During the investigated period, this kind of right (to possess) was much stronger than today. A proof that has been put forward to support this statement is that the person who possessed or used the "fruits" of an estate, in the different laws, was at the centre of juridical focus, not the holders of dominium directum. Also in the canonical right, as it was interpreted by the scholastic theology, no man could own the property of God, but he/she should enjoy the fruits of it.
Besides the written and normative legislation, property could be received as a result of non-written acts (through proven tradition, use of and work on the estate). All of these arguments focused on proof of possession and use rather than ownership (Ågren 1995:112f). Leasing-contracts to "apartments" and houses were among those rights included in the concept of dominium utile (Fenger et al. 1982:129f;Bøe & Sørgaard 1962). An important consequence of the legislative culture was the fragmentation of the possession-rights. There developed different kinds of rights to use the same resource, which created quite complicated and conflicting circumstances. An investigation of Danish towns has given examples of this complexity. Rights to resources that could be sold included, for example, town-farms (Sw. stadsgård), halved town-farms, buildings, booths/rentable rooms within houses, fruit and cabbage gardens and, less often, transit-permissions (Fenger et al. 1982:124).
Another consequence of the strong possession-rights was the division between land and building in the medieval towns. The owner of a piece of land did not necessarily have to own the house standing on it. In fact, findings in Danish towns suggest that those who exercised their rights as owner of the land (dominium directum) most often were spiritual institutions or the town council, while those who exercised their rights as possessors of buildings (dominium utile) to a greater extent were ordinary burghers (Fenger et al. 1982:124f); that is, one person could have owned the land, another the building standing on it (the right to use the land), and a third could have possessed the rights to use the building. This implies that the phenomenon of different owners/possessors with regard to land and buildings was at least not unusual, and that possession-rights to use land were secure enough to invest in built structures on rented land.
This differential system of different and conflicting rights to various resources within real estate naturally also created a complicated and multidimensional system of rents and charges. As I understand it from research on the topic, there were at least three different levels of payments for rights to property and houses: -Rents to the feudal lord of the town. This was a primary level in the exercising of dominium directum, and was paid to the founders and/or those who gave and guaranteed the town its privileges. In Denmark, it was mainly the king/crown who, on royal estates, founded towns (Andrén 1986), and thereby was the holder of the primary dominium directum. During the Middle Ages, several of these rights were, however, donated to ecclesiastical institutions. In the early Middle Ages, this rent was given by the individual holders of the rights. For example, the Midsommargäld (Eng: midsummer-geld) and the Arnegaeld (fireplace-geld) were taxes given to the king by the different holders of land in some towns (Rosqvist 1974). At the turn of the century, in 1400, the taxations had been altered. The new kind of tax, the town tax (byskatten) was collected by town officials (from the different property-holders), and was then given to the feudal lord (Rosqvist 1974;Johannesson 1947:134f).
-Rents to the individual holders of the dominium directum. This rent (jordskyld) was paid by the holder of the right to live on a piece of land, to the "owner" of the property (i.e., the one that had the right to sell the property). During the Middle Ages, it could be characterised as a rent or fee. After the Reformation the earlier ecclesiastical property (the most frequent receiver of "jordskyld" during the Middle Ages) came to be "owned" by the crown/king. The rules allowed the holder of the land to sell his/her right, but the buyer had to pay rent continuously to the crown/king (Andrén 1984:6;Johannesson 1947:131ff).
-Rents to the holders of dominium directum and/or dominium utile. These kinds of rents were paid by "tenants" to holders of the rights to use land and/or buildings or mortgagors (could be holders of directum as well as utile). Apart from the obligation to pay the rent, the tenant also had to make improvements and renovations to the rented houses. Also, easements were included in the contracts. Most common were rights to access, responsibilities of maintenance and cleanliness, or different personal easements such as the right to use a shop in the given building, transfer of the property to a convent after the death of the holder, or "eternal" possession of the property as tenant (Fenger et al. 1982:131). The actual rental levels were negotiated in the same manner as regarding prices on commodities (Fenger et al. 1982:122f), and were not so much based on speculations on the potential and/or calculated forthcoming increase in value. All three levels of rents were obliged to be on a constant level as long as the contracts continued (Johannesson 1947:137f), but the actual amount was conditioned by the size, placement and the state of the property (Fenger et al. 1982:122). Some of the preserved contracts were on a yearly basis, but generally researchers in the field of history of law mean that the possession-rights in praxis were inheritable (Andrén 1984;Fenger et al. 1982:122). The many gifts to the ecclesiastical institutions were constructed in the same manner. They were conditioned by easements and, according to the fulfilment of these, had an eternal basis (Andrén 1984:6). Alterations in the rental levels could be achieved between different tenants, and through investments in built structures.
The complicated ownership in feudal towns with, for example, "eternal" leases, and the likewise strong holding-rights of the tenants, constituted a limited and undeveloped real-estate market. If there was a market, it consisted of "leasingcontracts" rather than privately owned properties. The constitution of the market, together with the limitations of the rental-levels, must have made it less likely to make a profit on short-term speculation on raised values of certain urban properties.
THE ORGANISATION OF THE BUILDING TRADE From the high Middle Ages, the urban building-trade was organised in guilds, which had a monopoly on ordinary projects in the different towns. With regard to "ordinary" projects, an employer mainly used masters from the different guilds as contractors. According to late medieval and early modern examples, a contractor undertook a project at a fixed price (Allgulin 1932:63). The employer was, together with the contractor, in charge of the continued work on the building (Östnäs 1984:60). Regarding larger projects, the different guilds were mainly responsible for their part of the project. Sometimes they shared the role as contractor, sometimes one guild was in charge (usually the master of the masonry-guild), or the project was led by a well-known freelancing masonry-master (Östnäs 1984:34f; Svanberg 1994:77f). On great building-projects like cathedrals, the contractor and the leading masters of the involved guilds had a fixed yearly payment. They were, together with representatives of the employer, responsible for keeping everything within a decided budget (Svanberg 1994:42f). Payment to craftsmen and suppliers was the responsibility of the contractor in charge and the representatives of the employer. There are no notes in the literature on fluctuations in these kinds of expenses. Payment to the building-guilds must generally, however, have been adjusted according to the beliefs of the "fair price", and thereby been rather fixed. Prices of raw materials were, on the other hand, more exposed to fluctuations. With the organisation of the building trade, a contractor could consequently not choose between different and more or less independent actors on a market, and thereby get lower prices for a project.
The "market" contained mainly two types of projects. Firstly, buildings were mainly erected on the initiative of its forthcoming inhabitants (Östnäs 1984:60f). These houses were generally not built on the basis of an expected increase in value. In general the employers had personal connections to their houses. As stated earlier, buildings were constructed and formed under the direct participation of the employers, who often lived there for the rest of their lives. The second type of project was when houses were built on the initiative of an employer, by a guild and for anonymous residents. This type of project was to a higher degree an investment to gain a profit. However, as stated earlier, the rent was fixed and settled according to the principles of a "fair price". Higher rents could be achieved through investments in housing. But the tenant's rights were strong, and investments had to be made either with the participation of the tenants -but tenants were mainly obliged to improve the house and could therefore not be charged higher rents -or when the lease expired. Therefore, profits were probably not calculated on a short-term basis.
Unlike today, feudal society does not seem to have been changeable with respect to fluctuation in the prices of real estate, and therefore gentrification of certain areas for the purpose of profit is less possible as a factor behind changes in social space. Capitalising an estate through investments in housing was done in undeveloped markets and on a long-term basis. The possibilities for short-term and high profits were limited. Also, the structuring of social groups in the towns, according to the feudal mode of production, had a firm and tough hold in urban space, making consumption of houses a difficult task. Individual persons could move to other neighbourhoods and thereby change social group by accepting another space and taste (the ideal of the spatial order of the orders), but they are consequently archaeologically invisible and hard to detect. Because of the personal connections to housing, the employer's involvement in the construction and the tenant's obligations to improvements, archaeologically visible changes in the use of space and architecture (disposition of taste) must be understood as essential changes in social relations. What, then, was the "fuel" behind gentrification in a medieval town, and how were these processes expressed in architecture and the use of space?

THE SOCIAL TOPOGRAPHY OF MALMÖ DURING THE MIDDLE AGES -A CON-GESTED IMAGE
The town of Malmö was founded in the middle of the 13th century on a sand-ridge by the waterfront to Öresund (Bager 1971 replacing a great tower-gate; another great tower-gate was built in its eastern limits. The market place and court, together with the town hall, were placed on the central parts. There seems to have been a sparsely built, almost open space between the market area, the town-hall and the main church of Saint Peter. Most of the other ecclesiastical, only the hospital of Saint George was situated in the outskirts, institutions were scattered along the entire street (see Rosborn 1984: fig. 8). The street was called Adelgatan. In the medieval county-laws, the suffix adel is used to characterise the most important, the principal, the most valuable and the genuine (Hoff 1997:289). The name Adelgatan is quite common in Danish towns, and refers to the principal or main street, and the place of the oldest settlement (Ljungberg 1960:35). Adelgatan can accordingly be interpreted and translated as the Main Street or the Original Street.
Besides the known medieval institutions, the majority of the prominent architecture came to frame the street (Figs. 3 & 4). The architecture consisted of brickwalled houses with cellars, built in a profane burgher-interpretation of the Gothic style. As in other towns of the Baltic region, nearly all of the houses were built on narrow and deep plots and had their gables toward the street (at the very end of the Middle Ages, some stone houses were built with their long-side facing Adelgatan). Those who lived in the prominent houses were, of course, the prominent citizens of Malmö (Fig. 5).
Through research done by the historian Anders Ulrik Isberg back in the 19 th century, the majority of the homes of the known mayors and councillors as well as town officials have been traced (Isberg 1897). In medieval Scandinavia, it was only the merchants that could be elected to a seat in the council, and it was primarily the wealthiest that entered (Lamberg 2001:54f). In other words, they  Rosborn 1984, andBager 1971). can serve as representatives of the town elite. During the period until the year of 1499, all except two, with known addresses, lived along Adelgatan (Fig. 5). A couple of other mayors and councillors, not marked on the map, had their houses in the western part of the street, which, in connection with the enlargement of the king's castle Malmöhus, was demolished in the middle of the 16 th century (Isberg 1897). The street functioned as the town-heart (i.e., the city-core) and the most important public space.
One of the adjacent streets, Kalendergatan, housed some noblemen-residences (Bager 1971:267f; Fig. 3). Among other things, this was the place of the residences of a town bailiff and the archbishop. Nevertheless the noblemen generally had their residences in connection to Adelgatan. The architecture differed from that of the prominent burghers, as many of the noblemen had their main stone-houses set back from the street and situated deep into the plots (Thomasson 1997).
The town was enlarged at the end of the 15 th century and the beginning of the 16 th . In a now lost inscription, once in the post-medieval town hall, it is stated that King Christian I, in the year 1456, "let the town be enlarged" (Bager 1971:272f). Written records also suggest that the mayor Hans Michelsen, in the very beginning of the 16 th century, constructed "the first ramparts that has surrounded Malmö"   (after Reisnert 1994:42). . Within archaeological investigations in the southern and eastern parts, remains of the older town moat were found, which, together with some written notes, indicate that the limits extended 50 to 60 meters to the north of the known border in the 16 th century (Rosborn 1984:40). This is sustained by the demolishing and subsequent rebuilding of the southern tower-gate further south, as early as in the mid-15 th century (Rosborn 1984:40). In the westerns parts, a new grand Franciscan convent was established in 1489, on what was supposedly "new" land (Bager 1971:272f). However, excavations imply that the southern areas were not built upon before the beginning of the 16 th century (Andersson 1979;Andersson-Koverberg 1982). The same seems to be the case regarding the surroundings of the new Franciscan convent.
These new areas of the town were mainly built-up with lodge-houses. As mentioned earlier, it was the mayor Hans Michelsen who initiated the new ramparts around the town. It was a couple of other mayors who stood behind some housing projects in these new areas. Both Jep Nielsen and Jörgen Kock financed a lot of lodge-houses, primarily around the new Franciscan convent (Bager 1971:293f). The majority of the known examples were constructed in brick, while others were of poorer construction. Brick-built lodge-houses are also known in the western parts along Adelgatan (Jörgen Kock), by the entrance-gate to the harbour (Jep Nielsen) and near the town church of Saint Peter (Rosborn 1984: fig. 8). Jörgen Kock, as well as Jep Nielsen and another prominent family, Anna Pedersdotter and Mogens Jensen Rosenvinge, built residences along Adelgatan in the beginning of the 16 th century (Fig. 6). The remarkable thing about these residences is that they were equipped with additional apartments for rent. The houses of Jep Nielsen and Anna Pedersdotter/Mogens Jensen Rosenvinge could be described as semidetached, while the apartments in the house of Jörgen Kock lay in an adjacent row. These investments in housing made by prominent burghers, although almost non-profitable, were a kind of conspicuous consumption that differed from that of the landed nobility.
Lodge-houses, constructed in poorer materials, have also been found in other parts of the town. The remains of this kind of housing, the oldest dating to the 12 th and the others mainly to the early 16 th century, have been found in the southern and eastern peripheries (Andersson 1979;Andersson-Koverberg 1982;Billberg 1979). Sixteenth-century notes in the written records, supported by stratigraphical observations, suggest that the peripheral areas of Malmö also were used for other purposes than housing. Brickworks were situated in the western parts, in the neighbourhoods of some of the described lodge-houses (Rosborn 1984:40f). In the southern peripheries there were many fruit and cabbage gardens, and in the eastern limits was the weavers' bleach area (Bager 1971:271f). The social topography of medieval Malmö was based upon a widely known feudal principle that vital institutions like town halls, different churches and prominent profane buildings that housed prominent persons, were all built framing the main street and market area of the towns. The town structure with different streets was all created in relation to Adelgatan; the paths of movement in the town were created, supported and sustained by this street as the main public space. Other usage of land with different architecture characterised the more peripheral areas of the town. These areas housed craftsmen-families, market-visitors, wagelabourers, poor people and so forth. Older street-and path-names indicate that certain sections were occupied by particular groups of craftsmen. Adelgatan, with the market and the court, the town hall, the churches and, perhaps, with costumers and employers living along the street, composed a vital public space even for craftsmen, labourers, etc. The patterns of movement could be described as almost self-generated, through the placements of vital meeting-points and forced movements.
The development of Adelgatan could be described as a slow and congesting process, whereby profane stone-houses and vital institutions during the Middle Ages came to frame almost the full stretch of the street. As a whole, in the beginning it was the place for the most expensive architecture and the homes of the most prominent persons in town; the street formed a socio-topographical zone with its own architecture and aesthetics. In the case of Adelgatan, it is difficult to differentiate what should be interpreted as natural consequences of economical and demographic growth, and what actually should be interpreted as social upgrading or gentrification.
The enlargement of the town was done with the permission of the king, but certainly on the initiative of the town council. At least three of the mayors were involved in the enlargements, and two of them made substantial investments in lodge-houses. According to written sources, Jörgen Kock collected 300 daler in rents from the tenants, which at the time was regarded as a considerable amount. But he was a minor property-owner compared with the ecclesiastical institutions that each year collected 3000 daler in rents (Johannesson 1947:133). These areas, and particularly the brick-walled buildings, created a different aesthetic in comparison to other parts of the town. The new buildings were not constrained by complicated ownerships and rights, plot-borders or older built structures. Seen in this perspective, it was cheaper to invest in "virgin soil", without expenses for the cancelling of contracts and for demolishing. The housing-projects should probably be regarded as investments to gain profit in a more long-term perspective. They can also be interpreted as a kind of suburbanisation of Malmö, but conducted within the feudal framework of privileges and tough spatial boundaries. Counting from the "city core" of Adelgatan, the "suburbs" were situated behind the brickwork facilities, which had the spatial potential to create a "rent-gap". But there are no preserved written records which could have illustrated a decline in "value", and neither are there traces of more accountable housing-investments (i.e., considerable changes in architecture or land-use), or an influx of more wealthy inhabitants, to support a further development that could be described as a gentrification of the area in question, at least not until the 19 th century.

THE REFORMATION AND THE NEW STRUCTURING OF THE TOWN
At the end of the Middle Ages, there was a strong Protestant movement in major Danish towns like Malmö. To make a rather long and complex story simple and short: the conflicting parts were on the one side reformative prominent burghers and on the other the old spiritual and worldly nobility. At the very beginning of the 16 th century the burghers "won".
In Malmö, it seems that Saint George was the first Catholic institution to be closed down. This was probably done sometime during the period 1506 to 1519, when building-material and gravel was sold and some of it was used in the erection of the new town wall (Thomasson, forthcoming). Next in the reformative chain of events, in October 1528, the monastery of the Holy Spirit and the Franciscan convent were closed down. The king gave the two institutions to the town council; the monastery was to be used as a town hall and the convent as a hospital for the "poor and sick" (Tomner 1977:20). In a letter dated the 5 th of June 1529, King Fredrik I announced that all houses, town-farms and properties in the possession of the different alters, vicarages and the guilds, should be handed over to the town council (Reisnert 1991:18;Johannesson 1947:139). Later that year, the possessions held by the town church of Saint Peter also came into the hands of the council (Johannesson 1947:142). After the Reformation, other ecclesiastical properties came into the possession of the king. During the period between 1534 and 1536 there was a civil war in Denmark (the feud of the counts; Swe: grevefejden). The strongest towns, Copenhagen and Malmö, declared themselves as "reichstadten", and tried to liberate themselves from the king (Lamberg 2001:37).
As the ecclesiastical institutions possessed rights to about 31 % of the properties in Malmö (Johannesson 1947:134), it can be understood that huge resources were handed over to the town council. These were then invested in the construction of a large market square and a town hall, during the period from the late 1530s to April 1545 (Tomner 1977:30;Fig. 7). In order to do this, the council used the area of the monastery of the Holy Spirit and other properties possessed by former Catholic institutions or guilds. It also acquired properties through barter transactions. But these resources were not enough to cover the expenses. The council also had to lend money acquired from Jörgen Kock, to the new town hall (Tomner 1977:30), and in this perspective Kock has to be seen as an investor in the new infrastructural setting.
The new market area was situated to the south of Adelgatan. The building of the new town hall, in the north-eastern limits of the new market, also meant that the earlier spatial connection between town hall and town church was abandoned. This could perhaps be interpreted in a reformative context, as a symbolic division between the church and the new worldly power -the town council. It could also be interpreted as a way to manifest independence from the central authority of the king, to become an independent "reichstadt" in the same manner as the Hanseatic towns.
With the construction of the Great Market, as well as the new town hall, the structure of the town became more multi-focused. The patterns of movement were no longer directed solely towards the old main street; the new market was a serious competitor to Adelgatan as the main public area. Streets that earlier had limited or no importance, became vital places of movement and transportation. The best example of this is the development of South Street.
Earlier, South Street together with Rundelsgatan composed a link between Adelgatan and the tower-gates in the southern part of the town. The hospital of Saint George, situated by the southern tower gate, housed persons who perhaps were avoided by other people. After the changes, South Street became the easiest and most suitable road (the gate at Rundelsgatan was closed down, see Figs. 3, 7) for out-of-town visitors to reach the new Great Market and town hall. From now on, craftsmen in the neighbouring areas had reason to direct their movement toward the street, whereby adjacent smaller streets and paths served as inflows not only to Adelgatan, but also to South Street. It became structured and competitive as one of the main public spaces of the town.  Rosborn 1984 andBager 1971).

THE PROMINENT PERSONS AND THE SOCIAL TOPOGRAPHY
The changes in the town structure, with more than one important public space, naturally also led to some unintended consequences for the social topography. The estates around the Great Market and along South Street became inhabited by important persons. From the middle of the 16 th century, mayors, members of the town council, town bailiffs and civil servants of the council started to settle in these areas (Fig. 9). In the first half of the 16 th century a majority of the people associated with the council lived along Adelgatan, but thereafter the medieval pattern began to dissolve. During the second half of the century, nearly as many had their homes in the new areas. This development became even more distinct during the 17 th century, when the majority of mayors, council-members, etc., lived by the Great Market and South Street.
The new settlers came to live in areas formerly populated by craftsmen and poorer social groups. The circumstances should be interpreted as a social upgrading or a gentrification of old craftsmen areas, to merchant neighbourhoods. This is especially notable in the blocks in the southernmost areas of South Street.
Prior to the changes, and apart from the institution of the hospital, the block of Saint George was occupied by tailors, brewers and smiths. Millers and lumbermen lived just outside the tower-gate (Bager 1949(Bager , 1961. The plot in the north-eastern part of the block was used in 1475 as a cabbage garden, and it was not until the very end of the 15 th century that a house was built (Reisnert 1991:14f). The people living in the block were among those who paid least in tax to the town council (Rosborn 1989:17). After the changes, the block was positioned along a crowded road. The buildings of the former poorer inhabitants along South Street were torn down. South Street became framed by two-storey buildings, with mayors and members of the town council among their inhabitants (Thomasson forthcoming;Thomasson 1997). After the demolishing of the hospital, the north-western plot of the block first became a noble residence, later the home of one of the mayors, and thereafter a member of the town council lived in the house. The plot in the north-eastern part of the block became inhabited by clergymen (Reisnert 1991:15). In the block, as well as the neighbouring areas, it was only the estates towards South Street that were built with prominent architecture. Towards the adjacent smaller streets, there were still rows of lodgehouses for petty craftsmen.  Compared to the old central space along Adelgatan, these new areas were not constrained by older firmly built structures. Therefore a new architecture emerged along the Great Market and South Street (Fig. 10). In the beginning it was characterised by solid brick-walled and timber-framed houses, built in two storeys. To a greater extent, they had their long-side towards the public spaces, as the plots were wider compared to Adelgatan. After a time, in the end of the 16 th and beginning of the 17 th century, new stone houses were built in renaissance-style. There are only a few known examples, but three out of four stood in connection to the new paths. Among them was the first stone house in the southern part of South Street, the house of Flensburg, built at the very end of the 16 th century (Bager 1971:347f). Also the new town hall, which dominated the Great Market, had an early renaissance design. In early drawings from the 18 th century, the western long-side, which was towards the market, seems to have been dominated by the entrance and three oriel-windows (see picture in Bager 1971:321). The partition between the two floors was marked by a moulding, probably cut in sandstone, and the brick-walled building had crowstep-gables. Seen in this perspective, the new vital public areas had a different use of space (with wider plots) and an architecture with a new aesthetic.

A FEUDAL WAY TO GENTRIFY?
The different definitions of gentrification all in some way presuppose capitalistic circumstances in order to be viable, and in that sense it is impossible to use this concept in another structural setting than the capitalistic. But analysing the core of gentrification -i.e., changes in urban social topography whereby areas earlier inhabited by poorer segments of the population become settled by more wealthy persons -and not the means for doing it, it should be possible to, contextualised, apply the concept in investigations of feudal towns. In this case-study, emphasis is put on the relations between agency and structure in changes of the social topography.
To begin with, urban space was produced within the structuring of feudal society. It was spatially defined by firm and tough borders and socially constituted by privileges and legislation. Together with the ordering of the orders, which meant an ideal based on the spatial separation of different societal (functional) groups, and the organisation of the production with little or no separation between home and work, there was little potential for socio-topographical change. An analysis of the town plan of Malmö during the Middle Ages revealed a town with a single focus for movement and institutional structure, which generated itself as the centre of the town and the main public space.
The actual value of buildings was not exclusively measured in prices. This was due to several factors, imbedded in what was considered socially acceptable. First, the organisation of the craftsmanship surrounding the construction of houses mainly produced houses in direct corporation with the employer as a consumer of the house. Second, prices were negotiated through the axiom of the fair price, and with a belief in a constant and permanent rent level. Third, the possessionrights, the dominium utile, expressed a strong holding in comparison to ownership (dominium directum) and presupposed what was called eternal leases. As a consequence of the conflicting and overlapping rights to different resources, ownership and possession-rights were complicated matters. Together with a lack of financial institutions, all of this created an, to say the least, undeveloped housing and estate market. The constitution of the market suggests that there was no reliable infrastructure, or social need, to translate house-and property-value into prices. Also, socially the houses were primary not regarded as commodities. However, even though buildings primarily were not produced as commodities, a kind of  Einar Bager 1971: 333, 350. value could be determined by, as Marx has stated, the quantity of socially necessary labour invested in their production; and as I interpret it regarding housing, this should be understood as investments in architecture.
In towns, investments in elaborate architecture must have been recognised as honourable and trustworthy, as an asset for the occupier, partly because of the employer's participation in the construction, and partly because of the holder's responsibility to maintain the status of the house. On a general level, of course, almost everyone could recognise prominent architecture and understand its value. But on a more detailed level, the aesthetics were directed to different fields of "readers" (in Bourdieu-terms: a group of persons that shares a common system of relations). In Malmö, this could be seen as differences in land-use and aesthetics between merchant and craftsmen neighbourhoods.
Even if there is proof of a kind of "suburbanisation" when the town was enlarged, there are no traces of rent-gaps in neighbouring areas, neither as new kinds of land-uses nor new architecture. The enlargements were done within the feudal structure, as the king (the feudal lord of Malmö) gave his permission. At the same time, three of the most prominent burghers of the town were among the investors, and must therefore be accountable as actors or producers of urban space. The lodge-houses that were built in these new peripheral parts housed some of the persons who paid least in tax to the council, most probably craftsmen, wage-labour workers and strangers. In that sense they did not challenge the spatial order of the different urban groups, as poorer groups traditionally lived in the peripheral areas of the town. This could be interpreted as such that the lodgehouses were bonded up to a segment of the population, and not to an anonymous and wide "market" of house-consumers. The firm social zoning of the town thereby prevented the creation of rent-gaps and valleys in a conceived land-value curve.
In Malmö, after the Reformation, the extracted resources from old Catholic institutions made it possible to challenge the built-up and structural constraints of the town. The investment in a new market place was of course partly done to increase trade, and seen as such, it was a favourable investment. In the different account books kept by the council, the incomes from trade in the Great Market increased considerably (Tomner 1977:71f). The new town hall was more important to the citizens than the older one. After the Reformation the town council took over many of the social responsibilities of the Catholic institutions (medical care, school, etc.). Further, their role in court-matters and communal affairs such as fire brigades, plumbing etc., increased. The construction of the Great Market and the new town hall created a symbiosis, which became a vital matter for the citizens as well as the visitors. The infrastructural investments changed the patterns movement, from a town with a single focus plan to a multi-focused plan.
The consequences of these changes were that the Great Market and South Street became vitally important public space. As such, it challenged the earlier hegemony of Adelgatan. The connections between prominent persons and important public space, and/or the understanding of vital public space as merchant areas, created a kind of cultural-historical core that "fuelled" a "social upgrading" or gentrification of the areas surrounding the Great Market and South Street. This development was not done through speculation of a potential profit on a capitalised piece of land. In these circumstances, it could rather be understood as a personal drive of the employer to achieve or maintain importance through investments in architecture, and, perhaps more importantly; as a corporative struggle in urban space that referred to collective social preferences. In contrast to modern examples, the blocks were divided among different social groups. The struggle was about the properties towards the main public spaces, which were gentrified. Adjacent estates towards less important streets were continuously occupied by rows of lodge-houses.
The relations between the conflicting perspectives of consumption and production as explanations to gentrification could, in the case-study of Malmö, be described as dialectic. Consumer choice in a medieval town like Malmö was not culturally defined by "independent" anonymous actors on a socially accepted "market" in a capitalistic framework; consumer choice and consumer preferences were structured by the relations between the different social groups. In a wider perspective this relates to the spatial organisation in the feudal society as a whole, and in an urban perspective it should be connected to the relations between the corporations. These relations produced the spatial settings, and the spatial setting produced the social settings. The tough socio-spatial constraints in feudal society had, however, a built-in unchangeability. When changes occurred in Malmö, they were dependent on extraordinary extracted resources that were invested in the infrastructure. These investments challenged the spatial order of the town, and the corporative relations produced the settings for gentrification around the Great Market and South Street. The corporative and individual agency followed the logic of the structure, and with regard to the Reformation as well as the changes in social topography in Malmö, agency indeed challenged and changed the structure.
English revised by Laura Wrang.