Urban Segregation in a Nordic Small Town in the Late-Seventeenth Century: Residential Patterns in Sortavala at the Eastern Borderland of the Swedish Realm

The general view of urban segregation in pre-modern towns has been that the wealthy lived near the administrative and economic center(s), while the poor were pushed to the limits of the town. This approach has been questioned by studies proving that urban spaces were socially mixed. This dilemma has been studied here by examining in detail the urban segregation in one small town, Sortavala, at the eastern borderland of the Swedish realm. The analysis shows that the town space was bipolarly segregated. The “gentry,” officeholders and the like, lived near the market square and town hall; the wealthy burghers along the main street. However, even the poorest taxpayers lived among the wealthy and those of high social rank. The segregation was relative: the proportion of the wealthy grew in the grid plan in the town center; the settlements growing “freely” outside the original grid plan were for the poor only.

In his influential work analyzing and generalizing patterns in preindustrial cities (1960) from all over the world, Gideon Sjöberg postulates that the dwellings of the rich and powerful were concentrated in town centers, while the poorest and powerless were pushed to the limits near the town walls. In status-oriented pre-modern cities, the administrative center, often surrounding the market square, was, Sjöberg maintains, the place for the dwellings of those of high social rank. In capitalist modern cities, the needs of trade guided the location of the wealthy burghers. He also remarks that along with social status and wealth, ethnicity and religion (often combined) formed remarkable patterns for the socio-topographical formation of urban spaces. 3 This has been the dogma repeated in many works on segregation, but it has also been disputed by many scholars. More recent research on urban segregation has offered evidence that the social space of early modern towns was mixed rather than strictly socially segregated. However, although all socioeconomic groupings may have lived in the same part of town, there has been a tendency to suggest that the wealthier groups had more weight in the center, while the poorer were overrepresented near the limits. 4 This phenomenon can be called "relative" segregation. 5 In short, this is the view given by research on urban segregation in the early modern towns of Western Europe and Great Britain. How was it in the early modern North, where the towns were small and often very young? Were urban spaces socially segregated; were town centers socially mixed; and can we find any residential patterns to make generalizations about urban segregation?
The results of studies undertaken in the Nordic countries vary. According to a Finnish group of researchers, 6 the academic dissertation of E. Brunnius from 1731 states that in seventeenthcentury Tornio, 7 the most able traders lived on the first street (i.e., the street nearest the strand and harbor), the "common people" on the second, and "the poorest" on the third. They therefore suggest that the town space of Tornio was relatively socially segregated at the beginning of the eighteenth century. However, they are unable to verify Brunnius's claim about spatial segregation with any primary sources.
In studying the social construction of the sixteenth-and early seventeenth-century town of Nya Lodöse in Western Sweden, the Swedes Rosén and Larsson combine the results of archeological excavations and archival sources. According to them, wealthier people lived in some parts of the town more than in others, while many artisans' dwellings were found in some areas. Yet the overall picture was that the settlement was socially and economically mixed. No parts of the town were specially designated to any special groups. 8 A GIS analysis of early eighteenth-century Copenhagen by a Dane, Mads Linnet Perner, suggests that the urban space was horizontally segregated to the streets of the wealthy and to the alleys of the poor, although they could live very close to each other. 9 However, Rosén and Larsen posit that segregation in small urban centers in Sweden may have changed during the first half of the seventeenth century. New towns were founded, and old ones were moved to new places. The renaissance grid plan inspired by antiquity became the ideal townscape in Sweden at the beginning of the seventeenth century. The change from freely growing medieval urban spaces to regulated town plans took place in many Swedish towns especially in the 1640s and 1650s. 10 These processes of founding a new town, moving old towns to new places to regulate medieval town centers to grid plans, may have broken the old social spatial structures of the urban spaces developed over decades and centuries. Were the old residential patterns transferred to the new grid plans? Did the gentry and burghers settle in new towns in the conventional order: the powerful in the center, the powerless at the limits?
Yet these changes in urban spaces may have paved the way for an intentional town planning, in which the social and economic needs of social groupings were taken into account. In the 1640s, when Kalmar was moved to a new place, social inequalities were inscribed in the town plan from the outset. When Jönköping was moved in the 1620s, separate areas by the seaside were allocated to German workers. On the contrary, in the 1620s, when Halmstad was rebuilt after its destruction, the tax records from the following decades show a mixed socioeconomic structure, with merchants and artisans of differing wealth living side by side. A general observation in these Swedish studies to which Rosén and Larsson refer in their article is that the wealthier merchants had plots closer to the market square and church, but no separate areas were reserved for special socioeconomic groups in early modern towns. 11 A problem with studying the socioeconomic spatial organization and verifying the patterns of spatial urban segregation of the settlement in early modern towns in the North has been the lack of cadastral sources and town plans in which the data about the socioeconomic, ethnic, or religious status of the inhabitants can be compared with and located in blocks and plots. Because some data on the social status and wealth of townspeople comparable with a contemporary geographical town plan and a list of plot-owners are available from the small town of Sortavala at the eastern borderland of the Swedish realm, we attempt to respond to this need and identify how townspeople spatially located themselves on a grid plan at the end of the seventeenth century.

Sortavala-A Borderland Town and Its Town Space
In the treaty of 1617 between Sweden and Russia, the provinces of Ingria and Kexholm were annexed to the Swedish realm. During the first half of the seventeenth century, new towns were founded in the new eastern provinces to promote trade in these remote areas. One of them was Sortavala, founded in 1643. Although the location of Sortavala by the northern bank of Lake Ladoga was favorable for shipping and commercial activities, the town never grew to be a remarkable trading center. However, it had local importance in collecting peasant products, mostly tar, for shipping via Lake Ladoga to the town of Nyen. 12 Nyen and Stockholm were the most important trading partners for the burghers of Sortavala.
When the Russo-Swedish War (1656-1658), a sideshow of the Northern War (1655-1560), raged in Karelia, Sortavala was badly damaged. Almost all the burghers, most of them Orthodox Karelians, fled to Russia. After the war in the 1660s and 1670s, the town was newly inhabited by Finns from the Savo region and some Swedes who had ended up in Finland. Only a few Karelian families continued as burghers or returned to the town. We can therefore study the spatial formation of the residential patterns on an almost uninhabited grid plan (Map 1).
In 1710, the town was badly damaged again in the Great Northern War (1700-1721). The treaty of 1721 between Sweden and Russia left the town on the Russian side of the new borderline. Sortavala shrank to a village-like trading center until it regained town privileges in 1783. In 1809, Finland was annexed to the Russian Empire as a Grand Duchy. According to an order of Emperor Alexander I, the territories annexed to Russia in the treaties of 1721 and 1743 (Sortavala included) were ceded to Finland. In the nineteenth century and at the beginning of the twentieth, Sortavala was an important school and merchant town for eastern Finland. However, as a result of the Second World War, Finland lost Sortavala with the part of Karelia annexed to the Soviet Union in 1944. Today, it is a small border town with fewer than 20,000 inhabitants in Russia's western borderland.
The appearance of the seventeenth-century town plan of Sortavala is known from a geographical map drawn by Erik Beling in 1697 (Map 2). Beling worked as a land surveyor in the Baltic provinces between 1688 and 1700. 13 Beling's map gives reason to believe that over the years, although the grid plan remained as it was planned in 1643, the settlement was enlarged quite freely in some corners of the urban space. 14 The town consisted of a rather small area: the northsouth length of the space inside the town's customs fence is about 280 meters, and the width about 300 meters. The size of the town space inside the customs fence was between approximately five and six hectares. 15 The population of the early modern town of Sortavala is usually estimated to have been about 600. This estimate is based on the number of plots (102) and the guess that the average number of people living in each house (plot) was six. 16 However, according to the estimates of Sven Lilja, based on a large research project on Swedish early modern towns, the average family size varied greatly, but the appropriate figure is between 3.5 and 4.5. 17 The tax roll of Sortavala from 1685 lists 103 burghers, but also seventeen names of landless (bobuler) living in the town and earning their living through different kinds of work. The gentry, that is, civil servants, schoolmasters, and the clergy, are missing from the tax rolls. The list of plot-owners (Notorium explicatio 18 ) in the map of 1697 names nine. In addition to these, the town council minutes mention several artisans like a tailor, a shoemaker, and a founder, who lived in the same houses as the burghers but are not visible in the cameral sources. With these prerequisites, we can assume that the number of households in the town of Sortavala was at least 140. Using 4.5 as an average household size gives almost the same result, 630 inhabitants, as previous researchers proposed with bigger estimated average family size but a smaller number of families. We can therefore accept 600 as the approximate population at the end of the seventeenth century. Sven Lilja has classified Swedish early modern towns according to their size. In his classification, Sortavala falls into the category of "small towns" (500-1,000 inhabitants) between the categories of "micro-towns" (less than 500 inhabitants) and "small medium towns" (1,000-2,000 inhabitants). According to Lilja, "micro-towns" were the most common category in Sweden throughout the seventeenth century. Almost all towns fell into these three categories, and only 11 out of 101 towns had more than 2,000 inhabitants. 19 According to Lilja, the median size of a seventeenth-century town in Sweden was less than 500 inhabitants. 20 We can therefore say that although Sortavala was a tiny town in the periphery, for its size, it was a very typical early modern town in the seventeenth-century Swedish realm.
In Beling's map of 1697, only one street is named. The 230-meter-long 21 "Great Church Street" bisects the town horizontally. Four streets crossing Church Street in an approximate north-south direction are unnamed. In a small community, the naming of all the streets was perhaps unnecessary. The streets divide the town into the upper part (five blocks), inner part (five blocks), and lower part (three blocks) on the shore. 22 Inside the blocks, the plots are separated by borderlines and numbered from 1 to 102. The plots in the grid's regulated blocks were roughly square-shaped, each side about 22 to 23 meters, 23 and the size of the plots in the inner part of the town was about 540 square meters, and in the upper part, about 470 square meters. The plots in Sortavala were very similar in size to other Finnish towns in the seventeenth century. 24 The town did not have walls or other fortifications, but it was separated from the countryside by the customs fence. The blocks and plots near the fence and shore were very asymmetrical, and the size of the plots varied greatly. The buildings on the plots were simple one-story single houses made of timber, with the shelters required for livestock in the yards. Burgher's granaries and storehouses were situated on the shore and are numbered in the map's legend from 103 to 134. Number 135 is the place for the gardens. 25 The town was limited by Lake Ladoga in the south, and fields in the north and west. In the east, the urban space met gardens, a stream, and the churchyard. In the northwest, the outermost plots were beside a rocky hill, Kisamäki (Sw. "Leekberget," literally "Playhill").
Sortavala was also a town that did not have any medieval structures, like an old town center, affecting the segregation process. After the mid-1650s war, the urban space of Sortavala where the burghers and civil servants settled was practically an empty grid plan. As mentioned, in rebuilding seventeenth-century Halmstad after the fire, some segregation took place according to the town planning. It is not known if any kind of plan existed, according to which the burghers and civil servants could claim possession of certain plots to build and settle in Sortavala. This makes Sortavala as an interesting "laboratory," as we can expect that if some segregation can be traced to the late-seventeenth-century city, then it was a result of natural segregation process produced in the social interactions of the settlers.

Tracing the Residential Patterns of the Sortavala Urban Space
Carl H. Nightingale has conducted a comparative study on urban segregation, beginning with Mesopotamian Ziggurats and reaching to twentieth-century cities. Nightingale postulates that the history of urban settlement is the history of segregation: segregation of some kind is an inevitable part of urban life. When not taking into account the regulations of town plans, the driving force of segregation has been, according to Nightingale, the free will of the wealthy to choose their neighbors. The perception that having the poor as one's close neighbors would bring down the value of one's property did lie behind the choices made by the wealthy. Urban segregation has been a result of choosing one's place of living in relation to wealth, ethnicity, race, religion, or other identity. 26 During the last two decades, the concept of space has been under re-evaluation. Modern human geography and sociology are looking at space as a socially produced and relative concept. Classical urban segregation research has been criticized from treating urban space as a container without interaction with its surroundings. 27 However, the early modern towns were trying to be containers. From the 1620s, the towns in the Swedish realm were surrounded by a customs fence. The traffic into and out from the town was controlled at the gates. The town had control over those who wanted to settle in the town. The town council decided if an artisan was needed in the town and, to become a burgher, one needed to apply for the right from the town council, showing a good reputation and obtaining guarantees from two burghers of one's solvency.
Urban segregation is often studied using segregation indexes, the most famous being the dissimilarity index. 28 These segregation indexes have been extended to measure spatial relationships to more accurately capture the spatial aspect of urban segregation. 29 Unfortunately, the data for Sortavala are too incomplete for such analyses. Spatial social relations must therefore be described by using a partly more descriptive approach. Several studies have used fiscal records to study early modern urban space in this way. Good examples are Tim Bisschops's visualizations of the social topography of late medieval Antwerp that map a variety of sources, 30 and the analysis of segregation in three cities in Holland by Clé Lesger and Marco Van Leeuwen. 31 Lesger and Van Leeuwen study segregation on three levels, between districts (macro), between plots (meso), and within buildings (micro). Studying on these three analytical levels, they find that in smaller cities (Delft and Alkmaar), segregation was visible mainly on the meso-level. The social and economic elites were in small clusters in good locations, but always near a cluster of lower classes, often even within the same city blocks. Only in Amsterdam do they find clear macro-level segregation, which they attribute to the large size of the population and the presence of a sizable Jewish minority.
Micro-level analysis is used in studying vertical segregation-whether the different social classes lived in the basement or on the upper floors. Although the sources from seventeenthcentury Sortavala sometimes describe households with more than one family living in a house, micro-level vertical analysis is irrelevant in our case study. Evidently, the houses were small one-floor wooden constructions made of timber, usually consisting of only one or two rooms. In studying the urban space of tiny Sortavala, we therefore analyze the horizontal segregation on meso-(blocks and plots) and macro-(town part) levels.
The socioeconomic status of townspeople is studied here in two dimensions: first, the social status given by the offices the person held; second, his wealth. We will study if the social segregation according to social status or wealth (or both) can be traced to macro-level town parts or to meso-level clusters of plots-or at all in the tiny early modern urban space that was Sortavala. Because of the smallness of the urban space-for example, the longest distance from the remotest corner of the town space to the market was about 500 meters-it was easy to go on foot everywhere. Perhaps in such a small area, it was all the same where the gentry or the wealthy burghers had their houses. The only noticeable minority in Sortavala was the Orthodox Karelians. The spatial distribution of their dwellings in the urban space is one aspect to be examined in our study.
Although our data do not allow the use of segregation indexes, we can trace the patterns of residence in Sortavala by measuring the distances of the plots owned by different social and wealth groups from the urban space's possible "hot spots." These possible densifications of the dwellings of people of a socially high rank or the wealthy are identified by locating the inhabitants on whom we do have data in the town's surviving plan. These identified or suspected patterns are then tested by graphically presenting the results of measuring the distances of the dwellings to the "hot spots." The main sources in our study are the town map drawn in 1697 by Erik Beling and the lists of tax contributions paid annually by the town's burghers. The map of Beling is furnished with a list of the names of owners of 102 plots. 32 One plot is unbuilt and empty, so the list consists of 101 names that mention the profession of the most local civil servants. 33 The amount of the paid tax contribution was based on the annual valuation of the burgher's property and trade. 34 Lists of the tax paid by each burgher in Sortavala have survived from 1681, 1682, 1683, and 1685 in the state's provincial accounts. The accounts from 1684 are missing. After 1685, the lists of the paid taxes were not included in the provincial accounts. Evidently, the tax was collected and used by the tax farmer, who rented the Sortavala fief from 1686 (see below for more details). 35 Several issues must be pondered when using the taxing lists as a source. First, the mayor, civil servants, schoolmasters, and clergy were omitted from the valuation. However, those gentry who owned a plot and house in the town can be recognized in Beling's map, because their profession is mentioned. Although we lack any information about their wealth, in tracing the socioeconomic organization of the town's settlement, we can refer to their high social status in local society. Not only wealth may have affected the location where a person settled in the town: social status may also have been an important factor. For example, the schoolmaster had a relatively high social status in local society, but compared with the wealthy burghers, he was certainly poor. Another difficulty with the taxing lists is that they were made about fifteen years before Beling drew his map and wrote the list of plot-owners. We can therefore find data about the wealth of thirty-eight burghers owning 41 plots out of the 101 plots mentioned in Beling's map. This makes our data quite scanty, but we assume that we can obtain some good clues about the spatial organization of the settlement of Sortavala in the 1680s and 1690s.
The main sources of this study are supplemented by the rolls of the town council (Sw. rådstugurätt). These minutes of the town council have survived as an almost continuous series between 1673 and 1706, and they cover the entire period studied by this article. The literature on Sortavala supplements the archival data. The history of the town of Sortavala in the seventeenth century has received little attention. There are descriptions of the outlines of the development, and some very detailed information in two town histories published in 1932 and 1970. 36

Mayors and the Gentry by the Market Square?
In 1651, Sortavala town and the parish around it were given to Gustaf Adam Banér. A manor was founded on the other side of the Vakkosalmi strait from the town. Banér never visited his remote shire. A bailiff, called a "hopman," took care of the manor and the fief, and acted as the town's mayor. The largest fiefs in Sweden and Finland, including Sortavala county, were reduced to the Crown by the decision of the Swedish Diet in 1680. In Kexholm, Ingria, and Estonia provinces, the reduced fiefs were rented to the tax farmers. In Sortavala, the former "hopman" of the fief and mayor Johan Mether rented the fief from 1685. 37 During the latter half of the seventeenth century in Sweden, the rule was adopted that mayors' posts were no longer to be filled by a town's burghers but by men who had at least some academic education or experience in jurisprudence. 38 One task of a mayor was to act as the judge at the town court. Some training in jurisprudence was therefore needed. 39 Bailiffs like Mether often met this requirement. The fief owned a plot (81 in Beling's map) larger than others at the "parade place" by the market square. The legend of Beling's map states that the plot had long been owned by the man responsible for the fief. 40 Moreover, the inspector of the fief and later the tax farmer had a residence in the manor outside the town limits, as well as in the town by the market square. The house on the plot was used as the meeting place for the town's council, so it also served in a way as a town hall.
We only know the locations of the residences of two mayors of Sortavala. Sven Hielmberg, who was nominated as the deputy mayor of Sortavala in 1687, evidently did not have a permanent dwelling in the town. When Hielmberg suddenly died in 1689, the burgher who had accommodated Hielmberg while he lay on his sickbed demanded compensation from his heirs for the food, beer, and spirits he had given him. 41 Mether was again forced to take care of his mayoral duties. 42 Carl Ottoson, the son-in-law of Johan Mether, replaced the deceased Hielmberg as deputy mayor of Sortavala. Finally, in 1693, he was nominated as the full mayor. 43 Ottoson owned a house on two adjacent plots (both numbered 32 in Beling's map) on the north side of the market square. 44 Both Ottoson and Mether died in 1697, and the new tax farmer of the manor, Salomon Enberg, was nominated as mayor. 45 Enberg had his residence in the manor outside the town. Enberg came into conflict with the town's burghers, and in 1700, Benjamin Krook was nominated for the mayor's post. 46 Later, he worked as a district judge in northern Savonia county. 47 It is clear that Krook lived in the town, but the location of his dwelling is unknown. Krook was the last mayor of Sortavala before the town was destroyed in the war of 1710.
Alongside the mayors, a small number of people owning a plot in Sortavala can be labeled as belonging to the "gentry." We use the concept "gentry" here for civil servants, foremen of the manor, schoolmasters, and members of families who otherwise held notable positions in the Kexholm province. The plots next to the market square seem to have been valued highly by the gentry. The manor house serving as a town hall and the plots of Carl Ottoson were by the market square, as described above. Most other plots by or near the market square were inhabited by the gentry, wealthy burghers, or artisans.
The young district clerk (Sw. häradskrivare) Lorentz Frese owned a house on a plot (24) only a few steps from the market square. 48 Lorentz Frese's father was a district judge in Ingria, and Lorentz followed in his father's footsteps. Having graduated from Uppsala University (in 1700 with the name "Lorentz Freese Kexholmia Carelius") and the Royal Academy of Turku (in 1702), he first served at the Court of Appeal in Turku. In the 1710s, he was nominated as a vice district judge in the Kexholm province, and in the 1720s as district judge of the court of Karelia. 49 Plot 93 by the market square is marked in Beling's list of plot-owners for Johan Amptman. 50 The bailiffs and tax farmers of the manors were often called "inspectors," "hopmans," or "amptmans." An "amptman" was a bailiff of lower rank, 51 an overseer or a foreman. However, an "amptman" was clearly better off from the common folk. It is clear that "Johan Amptman" in Beling's map is the amptman of Sortavala manor Johan Isaksson Looman (or Loom, as he is called in the council's minutes 52 ). 53 Plot 92 on the southeast corner of the market square was owned by the "widow of Daniel Bång." The background of Daniel Bång is unknown. However, the Bång family is well known as burghers and priests in Finnish towns and parishes. The most famous was Daniel's contemporary, the Bishop of Vyborg Petrus Bång (1633-1696). However, it is unknown if they were close relatives. The family had entered Finland from Sweden proper at the beginning of the seventeenth century.
Plot 94 next to mayor Ottoson's house by the market square was owned by Knut Skomakare, that is, "shoemaker." 54 Perhaps the plot beside the market square was suitable for an artisan's workshop. The vicar, curate, and to some degree even the sacristan can be counted as belonging to the literati of the tiny town's society. The church, vicarage, and curate's house were outside the town limits. However, the sacristan had a small plot (96) near the market square. The plot itself was behind plot 93, but Beling has drawn a narrow corridor between plots 92 and 93 from the sacristan's house to the market square. 55 The humble servant of the church lived in the backyard of the valued open space of the town. The vicar had died in 1692, 56 but his widow earned part of her living by keeping a tavern on the south side of the market square (plot 89). 57 Just beside her tavern was the tavern of tailor Staffan Sairanen (plot 90). In 1700, because of the fear of fire, the council ordered these two "huts" in poor condition to be torn down. 58 It is unknown if this was actually done.
We can conclude that around the market square, there was not only the town hall and houses of the gentry, but also a concentration of plots owned by artisans: in addition to the houses of the shoemaker and tailor, there was a plot (102) belonging to the carpenter Hinrich Koistinen. Two taverns complete the picture of the heart of the town.
However, not all the gentry lived near the market square. Another place which seems to have been acceptable for the lesser gentry to have their houses was block 4 in the upper part of the town plan. Carl Affleck, who owned a plot (74) in this block, was the oldest son of Simon Affleck. 59 His father Simon was in the service of Salomon Enberg, the tax farmer and mayor of Sortavala. Enberg had rented the taxes of both Sortavala and Pielisjärvi (the northernmost parish of Kexholm province) manors and fiefs, and Simon Affleck was the bailiff of the Pielisjärvi fief, famous for his heavy-handed deeds. The son seems to have had same kind of career in mind, because he is mentioned as an "amptman." 60 He very probably had the same master as his father. During the Great Northern War (1700-1721), Carl Affleck rose as an officer to the ranks of Lieutenant and Quartermaster. 61 Schoolmaster Petter Pomelius owned a house (plot 68) on the other side of the block where Carl Affleck lived. 62 Pomelius was a son of the late vicar of Puumala parish, Gabriel Carol, 63 and served as the schoolmaster in Sortavala for about twenty years in the 1680s and 1690s. 64 By the standards of Sortavala, amptman Affleck and schoolmaster Pomelius lived in wealthy company. This small cluster of gentry was complemented by three of the wealthiest burghers in the town, Anders Rautiain (plot 72), Tarasia Pukari (plot 73), and Thomas Immonen (plot 66), who owned houses in the nearest neighborhood in the same block.

Wealthy Burghers by the Main Street?
The mayor headed the town administration, the town court (Sw. råd), with the town councilors (Sw. rådmen), who were chosen from the well-established burghers. In Sortavala, there were usually five or six councilors at the same time. In practice, membership was a lifelong post. Only seldom was a councilor released from the duty because of his behavior, advanced age, or sickness. Membership of the town council was a mark of social status in the local town society.
To compare the wealth of the burghers of Sortavala, the households were divided in wealth quartiles of equal size by the amount of contribution tax paid in copper thalers in the 1680s. To make the households easier to compare, the limits of the quartile categories for all years are based on 1681 taxes when possible. 65 The taxes for 1681-1683 were recorded as copper thalers, while 1685 taxes were in silver thalers; a standard rate of 3-1 was used to compare silver with copper money. The wealth quartiles of equal size are given in Table 1: In Table 1, we have placed the known wealth from the 1680s of the town councilors mentioned in the legend of Beling's 1697 map and the wealth of the other burghers mentioned in the legend in the wealth quartiles. We must remember that it was possible to connect the data about taxpaying in the 1680s with thirty-eight plot-owners of 1697 of forty-one plots. The total number of plots with houses was 101. 66 The proportions given in the following tables are therefore not exact figures of the real situation. However, they indicate the trends in the studied phenomenon. Table 2 shows clearly that the town councilors were elected from the well-established burghers: the councilors were recruited from the wealthier half of the burghers, and the wealthiest quartile (4) is very much overrepresented. 67 A long career and stable position as a burgher in society was a prerequisite for becoming a councilor. 68 The councilors were not always chosen from the wealthiest burghers, but they can usually be described as at least belonging to the "well-to-do" category. However, the town's wealthiest burgher, Anders Taskinen, for example, was never elected to the town council. 69 According to Petri Karonen, the Crown favored a practice in which towns' strongest traders were not elected to town councils. It was more profitable that they put their efforts into trading and not into town council and magistrate cases. 70 However, the Crown did not get involved in the election of councilors in Sortavala. The councilors were elected by the burghers, gathered in the town hall, or the town council voted itself for its new member.
It seems that a burgher's education played some role in election as a councilor. In Sortavala, some councilors had an education of at least some degree. Councilor Martinus Canuti was teaching children before Peter Pomelius founded a school to the town at the beginning of the 1680s. 71 In 1681, Martinus Canuti is mentioned as a bridge bailiff (Sw. broofogde) in the minutes of the district court in Sortavala parish. 72 His Latinized name refers to academic studies, although we cannot find his name in the graduate registers of the Royal Academy of Turku. 73 We do not know on which plot Martinus Canuti had his house, but when he died in 1693, Brun Olofsson, who lived on plot 14-just a stone's throw from the market square-took his seat on the town council. Brun Olofsson was a son of bailiff Olof Brunsson. The minute mentions that the gathered burghers of the town wanted to elect Brun Olofsson to the town council because of his irreproachable life, and because he was a "scholar" ("Literatus"). 74 Education, wealth, and the councilor's position gave these burghers a status in the local society very close to that of the group we have labeled "gentry." One councilor who may have had some scholarly education is Jören Wallius. He was a son of the first vicar Jören Petri Wallius of Kitee parish near Sortavala town. He lived on plot 28, which was on the corner of the street heading from the market square to the main Great Church Street. Beside Jören's plot was the house of his brother Hindrich Wallius (plot 20). Jören was a wealthy man among the Sortavala burghers. In 1681, his tax contribution was thirty copper thalers, while his brother Hindrich, less successful in trade, paid only five. 75 These remarks justify an examination of whether any residential patterns were characteristic of the group of councilors among the burghers. Even a glimpse of Map 3 affords an impression that most of the councilors for whom we have data lived by the main street. Seven of the ten councilors whose plots we could locate lived along the main street. Two out of ten had their house only one plot away from the main street, and one had his plot in the "gentry" cluster in block 4 of the upper part of the town. We can therefore conclude that Great Church Street, which bisected the town horizontally from the common gardens in the west to the bridge leading to the church in the east, seems to have been a valuable place to own a plot. The above conclusions, that the gentry tended to live near the market square, and the wealthy burghers along or near the main street, are based on the qualitative description and the visual image obtained by mapping the data (Map 3). However, human brains tend to ascribe order and patterns in data where they in reality do not exist. To test the conclusions, we have measured the distance of the plots owned by the gentry and burghers by wealth quartiles (Q1-Q4) to the center point of the market square and to the main street. Distances were measured from the centroids 76 of the plots and calculated using a Python script, including the GeoPandas and Shapely software libraries. 77 The results are presented in Figure 1. The distributions of distances from plots to central locations are depicted using a box-and-whisker plot. A box depicts the five-number summary of each distribution, so that each horizontal line corresponds to one statistic. In ascending order, the summary includes the minimum (in this case, shortest distance), the first quartile, the median, the third quartile, and the maximum. The not available (NA) group represents the distance to plots for which no socioeconomic data are available. Figure 1 was created using the Python Seaborn visualization library. 78 Map 3. Spatial segregation of the gentry and burghers according to wealth quartiles (Q1-Q4).
The figure shows that the gentry tended to have their dwellings near the market square, while the wealthiest burghers belonging to quartiles 4 and 3 owned the plots near Great Church Street (median about twenty-three meters). 79 Meanwhile, the average distance of the plots of the poorer burghers (quartiles 2 and 1) was much greater (medians over 120 and almost 60) from the main street. The plots owned by the burghers do not correlate in any way with the distance to the market square, and vice versa, there is no correlation between the location of the plots owned by the gentry and the distance to the main street. Many of the poorest taxpayers seem to have lived in the corner furthest from the market square. This is the settlement of irregular plots in the western and north-western limits of the town. On the contrary, some of the poorest lived on these irregular plots on the eastern side near the market square. However, we do find members belonging to the poorest quartile (Q1) on the grid plan in the middle of the plots of wealthy burghers. The urban space of Sortavala seems to have been segregated to some degree and quite mixed at the same time.

Karelian Traders as an Ethnic Minority
Finally, we must study the possibility of the segregation of the only ethnic-religious minority in the town's urban space, that of the Karelian Orthodox families. The taxing lists of the early 1680s mention at least six burghers who can be reliably identified as Orthodox Karelians. 80 In addition, the list of plot-owners on Beling's map has four Karelian or Russian names, two of which differ from the names in the taxing list. We can therefore identify at least eight Orthodox families among the Sortavala burghers of the 1680s and 1690s. The proportion of Karelians among the inhabitants of Sortavala during the whole period was therefore about 6 percent. Following the guidance of Sjöberg's classic work, we must ask if this small group of Karelian trader families was a segregated sect in the spatial organization of Sortavala. Because of the data's shortcomings and smallness, this analysis must again be based on a qualitative description and visual mapping of the data (see Map 3).
In the 1680s and 1690s, the most remarkable of the Karelian traders in Sortavala was Staffan Klimpo. He had his house in the best place by the market square (plot 25) and according to another map made by Erik Beling, he owned most of the fields surrounding the town. Klimpo belonged to the wealthiest group of burghers in Sortavala. Although he was not a councilor, he was one of the oath-sworn men who were chosen to execute the evaluation of the contribution tax for each burgher. This shows that he was a respected member of the town's society. He was a shipowner, and from the minutes of the town council, we find that he sailed with his wares to Stockholm almost annually. Jöran Klimpo, one of the wealthiest burghers and a councilor in the town before the 1680s, was Staffan's father. With his own funds, Jöran bought church bells from Stockholm for the local Orthodox church. 81 This background may explain the excellent and valuable situation of Staffan's house.
We do not know the location of the plot of Staffan Klimpo's brother, Pamphilia. In the taxing list of 1681, Pamphilia is marked as even wealthier than his brother Staffan. He had to pay a tax contribution of ninety thalers in copper, while Staffan paid eighty thalers. Only one other burgher paid more tax than these two brothers. While Staffan remained prosperous, Pamphilia had another fate. The tax contribution lists show that his property and wealth were rapidly vanishing. In the tax list of 1685, he was among the poorest burghers. The minutes of the town council make it clear that Pamphilia's downfall was his destructive alcoholism.
Three other Karelians, Tarasia Pukari, Pafwila Federoff, and Ondruska Iwanoff Kuhnoinen, were also quite prosperous burghers. Pukari paid a tax contribution of twenty copper thalers in 1681, and in 1683, even thirty thalers. With this income, he belonged to the third wealth quartile of burghers (the fourth being the most prosperous). His house was in the northern part of the urban space (plot 73). The plot was not near the main street or the market square, but it was in the cluster of plots of members of the gentry and wealthy burghers. Pukari's nearest neighbors were amptman Carl Affleck (plot 74) and schoolmaster Petter Pomelius (plot 68). Pafwila Federoff was taxed five copper thalers, therefore, belonging to the poorest quartile of burghers, yet was the most well-to-do in this group. In 1681, Ondruska Iwanoff paid ten thalers and in the following two years, twenty thalers, but he is no longer mentioned in the tax list of 1685. We do not know the location of his house in the town. However, in the list of plot-owners of 1697, plot 98 on the easternmost edge of the town plan was owned by Gauril Pawilof. It is most likely that Gauril's patronym refers to Pafwila Federoff. We can conclude that although the Karelian traders in Sortavala formed a small religious minority in the town, there is no sign that where they lived had any spatial group cohesion in the urban space.

Discussion
At the beginning of this article, we saw that Finnish historians and archeologists suggested that the early modern town space of Tornio in Northern Finland was highly segregated: the wealthiest lived by the street near the shore, while the poorest were pushed to the remotest street near the town limits. However, the Swedish researchers Rosén and Larsson present the preliminary results of the excavations of the early modern Nya Lodöse, which suggest that the urban space was economically highly mixed. The poor lived side by side with the rich. They also guess that artisans' workshops were probably in some special parts of the town. Observations about such slight segregation have also been made concerning other Swedish towns. Studies of the spatial segregation in European towns show a tendency to find segregation on the macro-level (town parts) in large cities (Amsterdam and London), but social segregation in smaller towns is visible only on the meso-level (blocks).
The results of examining the Sortavala urban space of the 1680s and 1690s seem to support the results from early modern small towns elsewhere in Europe and Sweden. However, the segregation found in the data collected from late-seventeenth-century Sortavala seem to suggest that the segregation was at least bipolar. The market square was the "heart" of the town, especially for the gentry. The wealthiest burghers, on the contrary, lived along the main street or in the plots near it. Block 4, or the upper part of the town, had a cluster of gentry and wealthy burgher dwellings. Sjöberg suggests that living near the administrative centers was practical and socially valued by the early modern elites. 82 However, differences in the residential patterns of pre-modern elites and wealthy burghers have been explained by the different needs of administration and commerce because of the rise of capitalism before industrialization. 83 Because we have yet to collect any data on production in Sortavala from the studied period, we cannot even guess the extent of the differences detected in the segregation patterns between the gentry and the wealthy burghers. However, in all blocks inhabited by the wealthiest, we find plots of the poorest quartile of taxpayers. Settlements in the grid plan of seventeenth-century Sortavala were relatively segregated in their social and economic distribution.
The blocks furthest from the market square were irregular, and the plots near the customs fence were usually settled by the poorest taxpayers in the town. The block to the east of the market square was highly unorganized and perhaps settled by the many artisans living in the town. Although no town walls or ramparts limited the town space of Sortavala, the town was surrounded by a customs fence. Outside the fence, the town met fields owned by the vicar and the wealthy burgher, Staffan Klimpo. This was not the direction in which to enlarge the urban space. When the plots in the grid plan were built and reserved, anyone who wanted to settle in the town had to find the place for his house outside the grid plan. The only places this could be done were at the sides of the rocky Kisamäki hill and on the eastern banks of the cape where the town was built. This must be the explanation for the poor settlement and irregularly formed plots at the northern end of block 2 of the upper part of the town and on the eastern side of the market square (block 3 in the lower part of the town, behind the plots nearest the market square).
The irregularly formed and freely grown outskirts of the urban space of Sortavala indicate that although an early modern town tried to be a container, strictly limited to selected settlers and a planned spatially and socially closed space, it did not fully succeed in this task. People earning their living through craftsmanship, as farm hands or by other irregular works were able to settle at the limits of the closed urban space.
In conclusion, the settlement in the Sortavala urban space shows evidence of socioeconomic segregation at the meso-level. Those in socially respected positions and the wealthiest burghers occupied the plots near the market square and by the main street. At the same time, the spatial structure of this small town was socially and economically mixed. Among the wealthiest, we can find some of the poorest as neighbors. However, this rule is not valid the other way round. The wealthiest did not settle among the poorest in the rocky parts of the town. Much as in the smaller Dutch towns of the same era, the town space of Sortavala was more homogeneous on the macrolevel but segregated on the meso-level. This suggests that the urban space of small towns in early modern Europe was similar in both the core and the periphery of Protestant Europe, but further research is needed to confirm this.

Author Contributions
Antti Härkönen has prepared the cartographic presentations, figures, created the wealth classification, and written the paragraphs on explaining the visualizations and segregation studies methodology. Kimmo Katajala has written the rest of the text.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.