University of Helsinki

The concept of political ecology has been employed in two senses in studies on the subject. (1) In the tradition of French political geography, attention was primarily paid to climatological areas in which a given ideological trend had long been in a dominant position.¹ In fact, the method of relying upon cartograms and upon qualitative material in general has directed interest towards the static parts of the political map. This approach has the advantage that it is thereby possible to come to closer grips with the local background of the region. A party may receive support from highly varying conditions, and the regional approach provides a picture of the explanatory factor in question. (2) Statistical methods, on the other hand, encompass within the sphere of observations the whole region that is being studied: the communes of strong and weak support and those of stable and unstable support. From given basic units, say from the communes, the aggregate-level information about the variables that explain it are gathered, and their inter-relationship is ascertained by correlation coefficients or multivariable analyses. This provides a fuller and a more reliable picture of the sources of the formation of political opinion and of the social structure of the support of the parties, but the interpretation of the results by region raises difficulties of its own. Especially in studies of large regions it is difficult to point out the areas in which each relationship is specifically valid. Briefly, the regional aspect will remain somewhat weak in a study of this kind.

Neither of these two basic applications of political ecology replaces the other. Rather, they supplement each other. It is possible to use them side by side in the same study: in establishing general relationships, in defining political regions of different types and, subsequently, in analyzing the background of the support within the framework of these regions. If accurately delimited and politically uniform areas then appear on the political map, it will be possible to take a further step by examining more closely the structure of the region, its internal relations and its external characteristics.

It is intended to include all these aspects and research methods in a
political cartography of Finland which is now being prepared. This will be

a series of studies in three parts: (a) the defining of political regions and an analysis of their internal structure, (b) explanation of regional differences of support, and (c) an examination of regional changes.² Some results of the first part of the study will be presented in this article. To facilitate an understanding of the results, it may be useful to present a brief review of the Finnish electoral and party system, which naturally has had both a direct and an indirect role in the disposition of the political map.

The Electoral and Party Systems

In 1906, when Finland was still a part of the Russian Empire, a reform was carried out with respect to the institutions of popular representation: the Diet of the Estates was replaced by a unicameral diet of 200 members. The franchise was made general and equal, which brought the whole people and the whole nation into the sphere of electoral process, representation prior to that time having been mainly centered upon the south of Finland and especially upon the cities there. With a few ecologically insignificant exceptions, everyone who had reached the age of 24 years was given the franchise. In 1944 this minimum age limit was lowered to 21 years. Elections were held every third year, from 1954 every fourth year, and were normally held in summer so that travelling to the polls would be easier throughout the country. The Tsar — and, after Finnish independence in 1917, the President of the republic — have had the right to dissolve the Parliament and to decide upon the holding of new elections. This has occurred five times by gracious decree of the Tsar and, by 1967, six times during the period of independence.

By law the country is to be divided into no fewer than 12 and no more than 18 constituencies. In the first and the last elections in the observation period (1907-58) of the study that is discussed in this article, the number of constituensies is 16. The seats in the Parliament are distributed among the constituencies in accordance with the size of the registered population and not according to the number enfranchised. This procedure gives an evident advantage to parties that are relatively strong in regions with a high birth-rate - in Finland particularly in the north and the east, those parts of the country where the means of subsistence are scantiest. Exceptions to the system of the proportional distribution of seats have been the two singleseat constituencies: Finnish Lapland in the elections 1907—36 and Aland since 1948. In both instances the character of the constituency has stabilized the prevailing political tendency there and has kept voting turnout low.

The actual balloting takes place in voting districts, of which there may be one or more in commune, depending upon the size of the population. The scattered network of voting districts in the outlying areas previously curbed turnout at elections, but with the improvement in traffic communications and

with the growing number of voting districts (2462 in 1907 and 4983 in 1958)
this difficulty has eased.

The party system is still young in Finland. A more regular party activity has been maintained for only six to seven decades. Ecological studies are, however, facilitated by the fact that many of the political tendencies have played a role in politics throughout this period. In the most important historial stages of transition, for instance when the country had gained its independence after the Civil War or after the Second World War, the party system underwent certain changes but the new parties continued in many respects along the road staked out by their predecessors. All the most important ideologies are represented in the Finnish party system: the Liberal (at present the Liberal People's Party — 1966, whose earliest predecessor, the Young Finnish Party, was founded in 1894), Conservative (at present the National Coalition — 1918, whose chief predecessor, the Old Finnish Party, was founded in 1906), the Social Democratic (1899), and the Communist (at present the Finnish People's Democratic League — 1944, whose predecessor was the Finnish Socialist Workers' party — 1920, outlawed in 1930). Among the political movements of long duration, mention should also be made of the Farmers' Party, the Agrarian Union (1906), which became the Centre Party in 1965, and the Swedish People's Party (1906), which represents the Swedish-speaking part of the population. In addition a few small parties have been active, but their life-spans have rarely encompassed more than two successive elections, their proportion of the total vote usually falling short of five percent. The small parties have characteristically been splinter-groups of the larger parties, have been associated with the prevalence of smallholding in the country, or have been a result of the current political situation. From an ecological viewpoint it may be of interest to note that these parties have had fairly distinct regional strongholds.

Defining of Political Regions

The object of the study was four political groups: the Agrarian Union (A. U.), the rest of the Bourgeoisie (R. B.), which includes the Bourgeoisie other than Agrarian, the Social Democrats (S. D.), and the Communists (Com). As the question is one of a nationwide basic survey, the groups were selected in such a manner that they actually cover the whole political field. It was also the intention to divide both the Bourgeoisie and the Left into two, a more Conservative part and a more radical one.

An essential requirement for a political area is not only a given degree of support but also a stability of such support. In order to eliminate fluctuating and changeable areas any examination must be carried out over the longest possible period. The preconditions for this are good in Finland because since 1907 the number of votes cast for the parties have been recorded by the

commune in the election statistics. The said year could not, however, be taken as a starting point because the Agrarian Union did not begin nationwide participation in the election» until towards 1920, while the Communists first participated in 1922 (see Table 1). As an attempt was made in the study to compare the Bourgeois groups with the Leftist groups, these were incorporated in the observation periods in the following manner: when determining the regional support for the Agrarian Union and the rest of the Bourgeoisie, the parliamentary elections during 1919—58 were employed. In the case of the Social Democrats and the Communists the periods 1922—29 and 1945—58 were used, i.e. the times when the Communists participated in the elections. Moving the starting point of the observation period brought also the advantage that the study consequently covers a period of relative stability of the state of the parties. An exception is the outlawing of the Communist Party and the remarkable increase in its support after World War 11. But also in the case of this group, the regional basis of the support remained large by the same. The mutual comparability of the groups does not greatly suffer from the employment of different periods because the basic features of the political map have, as far as the tendencies examined are concerned, remained fairly similar. The strength of the* political tendencies could actually have been established from a few sample elections, but data from all the elections were required in other parts of the study series, particularly when changes were to be measured.³

The data on election support were gathered by commune. The length of the observation period caused inconvenience because, owing to changes of commune boundaries, it was necessary to make up a system of basic units with permanent regional borders for the entire period in question. As the population in the outlying regions and the islands have not always had good opportunities to reach the polling stations, support for the groups was determined by the ratio of valid votes. The parameter employed to express the size of the group in the commune is the arithmetic mean of the support percentage at the elections during the period. The stability of the support was confirmed by including in the political region in question only those communes in which the strength of the group had kept above the requisite limit at all elections during the observation period — perhaps disregarding a short exception caused by a party of brief existence.

There may be several grounds for the determining of political regions, and the results will vary accordingly. In a multi-party system it may be expedient to find out both the general political regions and the regions internal to a party. These two types are separated in this study on Finland. When comparisons are made of the quantitative relations of all political groupings, a picture is obtained (a) of the general political regions⁴ such as the communes

of absolute majority and of instances where a party occupies the position of majority group. Regarding (b) the internal regions of each party, use is made in the study of the terms general extension areas of support, actual support area, core areas, and area of influence. The lower limit for a general extension area was fixed fairly low, somewhere near the lower quartile of the party's support throughout the country. The support area is characterized by a greater-than-average popularity of the ideology and by a permanence of this popularity. For practical reasons a round ten was selected as central parameter instead of the national median. An attempt was made to distinguish from the support areas the core areas, where the party's share of the vote is close to the upper quartile level. The zone between the outer borders of the core area and the support area is called the area of influence. If large continuous areas arise in mapping, a (c) structure analysis is finally made of these.

The General Political Regions

Absolute majority in the commune

In order to be able to state that a group has a dominant position in the commune, it must have an absolute majority there. Only then can a party put its mark upon local conditions and the inhabitants feel that they are living in a specific political climate. When a majority is as great as this, the dominant position of the group will be apparent also in communal politics and, consequently, in occupation of the public posts. On the other hand, those belonging to minority groups are subject to considerable pressure, being in a politically alien milieu. It s difficult for them to maintain confidence in their

own opinion. One after the other may stay away from elections and gradually
become absorbed in the majority.5 The more isolated a region is, the more
massively will the political climate envelop the dissentient.

The groups studied show great variation in the achieving of absolute majority: the Communists have done so in one communal unit, the Social Democrats in five units, the Agrarian Union in 81, and the rest of the bourgeois parties in 101 (Table 3). The figures are averages, and the situation in individual elections is by no means always the same. Absolute majority has actually only been obtained by the bourgeois groups, and this is one of the most salient features of tho political map of Finland. Moreover, this high degree of bourgeois tendency occurs in the country as clear-cut continuous regions that — in accordance with the terminology used above — could be called regions of political climate. The communes of absolute majority, marked out with the symbol (E) on the Map, make up two regions of Agrarian Union climate. One is located in southeast Finland — in the Karelian isthmus and along the northwest side of Lake Ladoga — in the area surrendered to the Soviet Union in the peace treaty of 1945. The other, a much larger and a

powerful Agrarian Union group of communes, extends from southern Ostrobothnia to central Lapland, the area in which the party was originally founded. Both these regions are by nature areas of medium or small farming. The corresponding communes of the rest of the bourgeoisie are concentrated in three groups along the Swedishspeaking coast. In these cases it is the Swedish people's party wich lends color to the political climate. It is only in very few communes, e.g in towns and in the islands of the Gulf of Finland, that the Finnish-speaking parties have brought the average of the rest of the bourgeois parties up to over 50 percent.

The largest group in the commune

A position of political dominance cannot without reservation be ascribed to the largest group in the commune. In many cases the inhabitants are hardly aware of which party has most voters in the commune. Three groups may receive about equal supporl, but because of a minute lead one of them will be entered on the map showing the largest groups. A general political area will thus, to some extent, derive a different character, depending upon what degree of majority actually exists.

The difference between the said types of majority is illustrated also by the fact that the Social Democrats appear as the third considerable group, along with the Agrarian Union and the rest of the Bourgeoisie, on the map showing the largest groups: the Communists with 17 communes, the Social Democrats with 134, the rest of the Bourgeoisie with 171, and the Agrarian Union with 181 (Table 3). Again the result is not a gaudy patchwork, but the communes in which the parties have simple majorities find themselves close to one another — the only exception being the communist group, which pops up here and there on the map. The communes of Agrarian Union simple majority are immediately adjacent to the above-listed groups of E communes, and as a result the northern region in particular — partly, it is true, because of the extensive rural communes — seems to comprise nearly half the country. The rest of the Bourgeoisie is, on this map too, more dispersed than is the Agrarian Union. But also this former group has received addition its E areas - mainly from the nearest Finnish-speaking communes. The Social Democrats differ from the bourgeois groups in that their corresponding communes are, practically speaking, in one large bloc: a rectangle stretching from the southwest to the northeast, about half of which is located in the industrial part of the country while the other half located in the traditionally agricultural part. This rectangle is unusual also because of its direction, as other political basic boundaries in Finland generally run from southeast to northwest.

The map proves to be most broken up in the southwestern parts, where there are simple majority communes in all directions. The structure of the population is one in which there is a traditional agricultural overproduction — but at the same time it is also an essential part of industrial Finland, which

means that the service industries too are well developed there. Moreover, southwestern Finland belongs to that part of the country in which political activities have been carried on longest and where it has been easier also for the weaker parties, because of more effective communications and transport, to organize their informative activities than it has been in the sparselypopulated outlying districts. The more highly developed political character is also illustrated by the fact that it was the communal elections of these areas that the first coalitions — and also so far the largest number of coalitions — bearing the names of political parties were formed, while there have been relatively fewer coalitions there with non-political group labels.*

The Internal Areas of the Parties

The map of the general political areas is supplemented by a picture of the internal areas of the parties, especially the parties that have not been able to create majority areas. One such group is the Communist Party which has nevertheless played, politically and ecologically, such an interesting role in Finnish society that it deserves a separate examination. It is necessary, particularly when the formation of political opinion is being studied, to obtain an impression of the regional distribution of the support of all the parties.

The general extent of support

To obtain a background for a more detailed examination of the internal areas of the parties, a review is made (a) of the general extent of support in the country and, in addition, an analysis is made of the position of the groups in special areas which are essentially associated with the general picture, such as (b) the rural areas, (c) the cities, (d) the communes surrounding the cities, (e) Swedish-speaking areas, (f) the island communes, and (g) the national border areas. The general extent of party support is followed to the lower quartile of the averages, so that it may become more apparent whether support for a party is nationwide or local.

An inspection reinforces the picture that could be derived from the map of majorities: not one of the examined groups was evenly spread throughout the country, but neither was the support split up here and there in islands in different parts of the country. Instead of such extremes, we find an intermediate form: the strength of the group is fairly concentrated upon ascribed areas. It does not, however, follow from the concentration that the political activities of the group are limited to the said area to such a degree that the group could be spoken of as a regional party — not even in the case of the Swedish People's Party although its organizations and coalitions occur only in the bilingual constituencies or in the case of the Agrarian Union although its roots lie primarily in rural soil.

The group holding first place in the rural communes is the Agrarian Union, the only group that in the rural areas exceeds its national level of support. But the rest of the Bourgeoisie and the Social Democrats do not fall far behind. Although the lead over the Communists has grown considerably, the rural political picture remains quite heterogeneous, especially considering that the social structure of the population is more uniform there than it is in the cities.

The differences in the cities¹ are more noticeable and the order of magnitude among the groups is different: the rest of the Bourgeoisie, the Social Democrats, the Communists and the Agrarian Union. The first three groups are relatively stronger in the cities than they are in the country as a whole. The close contact with the milieu appears from the fact that in a region of strength the support percentage of all groups has grown in the cities to a higher than ordinary level. This observation also applies to the Agrarian Union — which is more alien in urban conditions — whose percentages in the cities of the north Finland agrarian majority area reach to the 10 percent. It is only the rest of the Bourgeoisie, the liberal-conservative grouping, whose support in the cities has such a favorable background that its success has not been dependent upon the basic political color of the region, for this grouping has achieved a considerable share of the vote also in regions dominated by its opponents.

The communes surrounding the cities' have a specific character of their own with respect to the order of magnitude among the parties: the Social Democrats, the rest of the Bourgeoisie, the Agrarian Union and the Communists. Attention should perhaps particularly be paid to the relatively strong grip of the Social Democrats in the zones between the cities and the open countryside. Even in such cases where the rest of the Bourgeoisie are a considerable majority group in the city itself, it is the Social Democrats who dominate the surrounding communes. The red belt surrounding bourgeois cities is explicable by the fact that the area is in a process of industrialization and by the existence of a v/orking population that travels to work in the cities.

The Swedish-speakinng areas¹ of the country are, because of the Swedish People's Party, so clearly dominated by the rest of the Bourgeoisie that 9 voters out of 10 support the* latter. The Leftists parties, being bilingual groups, have here been able to obtain a foothold to a greater extent than has the Agrarian Union, which once represented a pronounced attitude regarding language. The opportunities of the Social Democrats and the Communists to break the monopoly of tho Swedish People's Party by appealing to social class interests have been greater on the mainland than in the Swedishspeaking islands because the Swedish and the Finnish populations have had closer interrelations on the mainland. Where the Leftist groups are concerned, that group succeeds best which has, generally speaking, been in a stronger position in tho region in question.

In the island communes of the coastal seas, the political behavior of the

population, which lives mainly upon fishing, has followed the prevailing tendencies on the nearby mainland — which tends to mean success in the islands for the rest of the bourgeoisie. Support for the grouping, however has been greater in the islands than on the mainland because the inhabitants, living in isolated circumstances, have tended to adopt uniform attitudes.

Where the political circumstances of the border regions are concerned, greatest interest is directed towards the eastern border, which is also the borderline between countries representing the capitalist and the communist systems. Here the Bourgeoisie generally appear as the dominant groups, which, along with some other explanatory factors, — for instance in the Karelian isthmus - is due to the area having been a thoroughfare and a battlefield in many wars and subject to Russia when peace was made.¹⁰ In certain border communes farther north, where the historia! factors referred to above have not prevailed to the same extent and where the milieu of the population also differs otherwise from that of the border regions in the south, the Social Democrat or the Communist parties are in a majority position.

Support areas

When the map of the extent of general party support was studied, it was already possible to observe visually that support is not randomwise distributed here and there but is concentrated in certain regions. This observation raises new questions: (a) how many regions of this type can be discerned, (b) where are they located, and (c) what is the mutual regional situation of different ideological tendencies in Finland like? In this way we approach the support areas of the groups — these support areas being characterized by the popularity of the parties exceeding the national average.

The Agrarian Union, being the party of a population that is well rooted in its settlements and traditions, has remained cohesive: in this case too the communes in question make up two clear-cut areas of support, outside of which we find fewer than 10 communes of like support. The larger area includes the northern parts of the country (Ostrobothnia, the provinces of Oulu and Lapland), and the other area is also closely associated with the Agrarian Union climates, its location being in the southeast, a sizeable section of the province of Viipuri. In eastern Finland these support areas are connected by a corridor which may have served as a route for the dissemination of agrarian ideology between the southern and the northern areas.

For reasons already referred to when the majority areas were being discussed, the higher-than-average support for the resf of the Bourgeoisie is distributed in several directions in altogether five areas of support. The areas are located on the southern and western coasts, beginning at the Karelian isthmus and ending in Ostrobothnia. They include two Finnish-speaking and three largely Swedish-speaking groups of communes. Outside the continuous

areas there are only a very few communes in which the 30.0 percent limit is exceeded, and in spite of the greater number of support areas there is a close connection in this case too between the communes of strong support. As the rest of the bourgeoisie includes several political parties, the cohesive regional background is a consequence of the fact that the parties receive their main support from the same occupation groups and social classes,¹¹ which is partly explained by the ideological proximity of the parties.¹² There are, however, internal differences within the three Swedish-speaking areas insofar as the language divides them into two clearly distinct parts.

The rest of the Bourgeoisie have had commune groups where the 30.0 percent level has been exceeded in other parts of the country too, for instance in the northernmost regions, but these areas do not meet the requirement for permanence of support.¹³ The weakening of the group in this periphery of its extent area is associated with the withdrawal of Liberalism and Conservatism from north and east Finland towards the south and the west that began in the 19205. It was in this period that the proportion of urban population among the voters of the bourgeois parties increased markedly.¹⁴

The Social Democratic ciroup is regionally the most cohesive group on the support area map. The party is strongest in the rectangle that has become familiar in the political history of the country. It is cohesive unit also in the sense that the proportion of the vote inside it varies relatively little, but the grouping of the communes makes it possible to discern three Social Democratic areas of support. The largest and most central area comprises the western end of the rectangle (southwest and central Finland), the industrial and densely-settled part of the country in which the Social Democratic organization first arose and where, also, the center of gravity of party support has been located from trie beginning.¹⁵ The second area of support, the middle of the rectangle, is peculiar in shape in that the area has two branches towards the south in the direction of the Gulf of Finland, one along the river Kymi and the other along the Saimaa Canal. At one time the socialist ideology was carried from the coast to the interior in the very direction of these passages. The third area of Social Democrat support is located in the eastern border, in north Karelia, which is dominated by small farming and has relatively meager means of subsistence. The close proximity of the support areas provides good opportunities for the creation in south Finland of a vigorous atmosphere of social democracy that reinforces the party fidelity of the voters. The influence of this atmosphere can be seen in the fact that in election defeats the losses of the party have in these traditional regions been relatively smaller than elsewhere in the country.

It is characteristic of Communism in Finland that its two support areas should represent different types of regions which are far apart from each other. The smaller area - one of weaker support which is. somewhat broken up — is located in the rrore industrial southwest Finland on the traditional ground of Socialism. The actual chief area of Communism is found in north

Finland, where it extends from the southern border of the province of Kuopio to the northern border of the country. In this respect there is a uniformity with communism in Sweden, which likewise has its most cohesive and most vigorous areas in the north.¹⁶ In northern Finland the group has been labelled "backwood and forest Communism" because the industrial life of these sparsely-settled regions is dominated by smallhold farming and by logging in the winter season. Moreover, it is the geographic isolation which has influenced the attitudes of the inhabitants. The traditions of Socialism are weakest in the north, but in spite of this it was in this support area that Communism advanced most strongly after World War 11.

There is an important observation and conclusion to be made from the relative locations of the support areas of the different groups. Roughly speaking, two basic areas can be discerned on the political map of Finland — the more conservative south and west Finland and the more radical north and east Finland.¹⁷ The most radical groups of the Bourgeoisie and of the left — the Agrarians and the Communists — are strongest in the northern and eastern parts, while the point of gravity of the support of the rest of the Bourgeoisie, and also partly of the Social Democrats, is located rather in southern and western Finland.¹⁸ Neither the Bourgeois nor the Leftist tendencies have thus occupied the same regions, nor does either grouping represent a regionally cohesive movement. The Bourgeoisie occupy a majority position in the support areas of both the Agrarians and the rest of the Bourgeoisie, but this majority position is primarily due to of the two being stronger than usual. The Leftist areas, again, are chiefly dominated by the Social Democratic Party. It seems that in the more radical soil df the north and the east the more radical alternative within the Bourgeoisie and the Left wing is chosen. In a more conservative surrounding the choice is often for the more Conservative wing of the grouping. Political extremes are not located in the same regions in Finland.

Core areas and areas of influence

A further analysis of the political field attempts to find areas which are even smaller and where the percentage of support is greater. The next thing to be examined is (a) whether a group obtains support equally in the different communes of the area of support or whether it is possible within the support area to discern particular center areas, which in this study are referred to as core areas. The criterion for communes of this level is that the proportion of vote should reach an even ten near the upper quartile of the national vote (Table 2). If a concentration of support is observed for an inventory of the parameters of the said areas and for the establishment of internal differences, it will first be necessary to find out (b) the number of core areas, (c) their extent and shape, (d) the degree of support, (e) the communes of correspending

support level that remain outside the core areas, i.e. separate
pockets, and (f) how the communes of weaker support, i.e. the influence areas
of the core areas, fit into the area of support.

The study revealed that in the support areas of all the groups the communes of greatest and of weakest support were located in different directions.¹⁹ Core areas were found although the picture was not clear-cut enough for a single core area to dominate each area of support. Several core areas may be found in one area of support, the number varying between one and three. All in all, 18 core areas were found, i.e. four for the Agrarian Union, five for the rest of the Bourgeoisie, five for the Social Democrats, and four for the Communists. The core areas comprise 263 communes out of 503, and the distribution in groups is shown in Table 4. The sum total of the number of communes in the various core areas was 269, but six communes

in northern Finland were included in both Agrarian and Communist areas,
which is possible seeing that it is the internal areas of the parties that are
the subject under examination.

The number of background forces influencing the formation of core areas is so great that common features can hardly be observed in the extent and shape of the areas. The Agrarian core areas, like rural core areas in general, are large and of a solid shape. However, where the other bourgeois parties are concerned, the core areas, like the support areas, are characteristically small in size and irregular in shape. The Social Democratic and the Communist core areas are of varying dimensions. It is, however, possible to generalize from the core area map that the areas of smaller size and with fewer communes are located in »outhern Finland, where a greater heterogeneity of social and political conditions have made it more difficult to take and keep an extensive area.

Although it is not the intention here to explore the birth processes of political areas, there may bo reason to point out the evident effect of local traffic communications upon the formation of core areas. Wherever river valleys, railways and roads facilitate the spreading of information and of economic intercourse and interaction in administration and culture, the conditions have been favorable to the advance of political ideologies. As ideologies in certain instances have travelled also across big lakas, even these waterways have acted rather as routes of contact than as physical obstacles. The traffic routes and the population centres at their junctions appear to have had their greatest influence upon location of the core areas of the rest of

the Bourgeoisie, upon those of the Communists, and upon those core areas of the Social Democrats that lie to the west, apparently because the commonest direction of travel of ideologies has in these instances taken place from the city out toward the countryside. Also, the fact that industries are located along the main communications of an area has led to the location there of a most vigorous support for the left.²⁰

The median values illustrating the level of support in a group's core-area communes show the order that is familiar from what was said above regarding the areas of political climate: the rest of the Bourgeoisie 68.0 percent, the Agrarian Union 58.2 percent, the Social Democrats 43.9 percent, and the communists 36.6 percent. The strength of a group varies slightly depending upon the location of the core area. In addition to a suitable atmosphere, the lack of related political ideology is also significant, this being partly a question of atmosphere too. Thus the rest of the Bourgeoisie stand in good defensive positions in their bases along the seacoast, where they are not completely surrounded by competitive alternative groups. In the interior — as has been revealed above — the fate of the bourgeoisie has been to surrender voters to other tendencies, primarily to the Agrarian Union.

There are in the country relatively few separate pockets — only 62 (Table 4) — i.e. communes with the same level of support as, but outside, the continuous core areas. Most of these communes are pockets of the rest of the Bourgeoisie, whose voters are regionally the most dispersed; the separate pockets of the Agrarian Union include 21 cities and 4 islands in the Gulf of Finland. The support obtained by every group in the separate pockets shows a fair amount of variation. The rest of the bourgeoisie and also the Communist group have managed well in the separate pockets in a politically alien milieu where the pressure from different opinions is stronger than it is in the actual core areas. The Agrarians and the Social Democrats have, in contrast, had to be satisfied with a smaller share of the vote in the separate pockets than in the core areas, and their defense mechanism and the overcoming of the sense of political isolation have in these circumstances not been so efficient as those of the rest of the Bourgeoisie and the Communists. The importance of regional contacts to the Agrarian Union and the Social Democrats can also be seen since their share of the vote in the separate pockets tends to rise the closer the commune in question lies to the core area.

A core area may in parts reach the edge of the support area, in which case the share of the vote decreases sharply and the core area does not radiate its influences any farther. Such a situation is, however, an exception. The core area is usually located in the interior of the support area, which means that the core area in such a situation is joined by a commune group of weaker support, an area of influence. The consequence of our criterion for core and influence areas can be that the core areas of two political groupings may be linked with influence areas common to them both. An even

more common situation is that between near-lying core areas of one and
the same political grouping.

Like the core areas, the areas of influence do not follow a uniform pattern. A few areas of influence are dependent upon the size of the core area and on the level of support, in such a manner that a strong and large core area gains a greater area of influence around itself. Many of the core areas in farming regions are also a part of a larger unit of the same political color. Thus, the northern core areas of the Agrarians and the eastern ones of the Social Democrats are surrounded by influence areas of such extent that the core areas are thereby linked one with another. The core areas of the rest of the Bourgeoisie and of the Communists, in contrast, do not have an equally extensive outer defence. These parties have to a great extent occupied locations around population centers, and the rest of the bourgeoisie have also in some places taken up occupancy in the proximity of Swedish-speaking regions. It is possible to discern among the areas of influence two basic types: zones and blocks (F gure), the area assuming its shape in this case according to the social structure of the group and to some extent according to local traffic communications. In the event that the communications radiate closely in many directions the area of influence will easily take the shape of a zone, while blocks are more characteristic of areas dominated by small number of directions of traffic

Structure of the core areas

In ecological studies attention has been paid mainly to the geographic distribution of political support and to an explanation of the differences observed in this respect. Because political areas appear to form continuous and solid units clearly distinguishable from their surroundings, they provide

other aspects that are interesting from the viewpoint of the essence and the influence of an area. In what follows we shall examine the structure of one type of core area from three aspects: (a) whether it is possible in the core area to discern communes that have attained a position as a leading center, (b) whether it makes sense to speak of the borders of a core area, and (c) to what extent the political structure of an area varies with its main political color and its geographic location.

Centers

In political literature references are occasionally made to places that have been of local or national significance in the history of a given ideology.²¹ In attempting to discover which communes in the core areas have in this respect gone ahead of the others, the criterion employed was their contribution to practical party work and not their significance in the history of the ideology. Three external and easily checked factors were used for the criterion: the location of the party office, the commune where the party publication is issued, and the domiciles of members of Parliament.²² Three groups of problems have been the primary objective for clarification: (a) whether it is possible in the core areas to discern one or more centers on these grounds, (b) what distinguishing marks are associated with the centers with respect, say, to their character, location and degree of support, and (c) whether the center has had any influence upon the political power relations and upon changes therein in different parts of the core area.

It appears that the center is essentially a part of the structure of the core area because communes occupying this type of special position were found in all core areas. The number of centers varied however. Most frequently (12) there is one center, in some (5) cases the point of gravity is located in two places. In one core area, which is a core area of the rest of the Bourgeoisie - a multiparty and regionally dispersed grouping — there are three parallel centers discernible.

Most of the political centers are located in communes that have a dominating position in the internal non-political activities of the area, such as in commerce, services, traffic and, in the case of the core areas of the left, in industry. Further, many of the communes are administrative centers. It is already evident from this that it is usually a question of population centers: nineteen of them are cities, four of them boroughs, and only two of them rural communes. Out of practical considerations the parties have concentrated the management of their election and information work in places where the population of the area carries on its other transactions too. The party mechanism of the center usually includes a district office and a newspaper, which often go hand in hand, although there may be no newspaper, e.g. in the core areas of the leftist groups. A higher than average number of Parliament members have been elected from the centers, but this has probably

not contributed toward a strengthening of these communes. This considerable
number of members is rather a consequence of the fact that the candidates
from the center generally receive a large number of votes in Finland.23

The significance and the duties of the center vary somewhat with the party. A city that functions a» the center of an agrarian area is only a technical organizing center, but for the rest of the bourgeoisie and for the communists, whose roots are closer to urban settlement, the centers also provide ideological stimuli to their surroundings.

The fact that the location of the center is dependent on the location of the population centers that dominate the area explains a variation that is noticeable in this respect: nine centers (out of 25) have been located in the middle of the area, seven near the edges, and nine barely outside the core area. These last mentioned chiefly consist of Agrarian Union centers which for their practical service purposes may just as well be farther aside.

From the nature and significance of the center of the core area it is possible to conclude that its distinguishing marks do not include a greater than average support within it. Only one center (Com) is the strongest commune in its core area, and in one other instance the group (R.8.) has in its center exceeded the median level of the core area.

Nor are the groups always at their strongest in the communes adjacent to the centers. Nevertheless, the majority, i.e. 13 centers, have in their proximity a zone of stronger support beyond which the group weakens with increasing distance from the center. The phenomenon is characteristic, above all, of the leftist parties. In the Communist areas there are actually fortresslike core sections, a situation that facilitates a forwarding of arguments to their supporters, and the protection of the latter from outside influence by the creation of a regional and, within the region, a social community which is as far as possible a closed one.²⁴

The propensity to temporal changes in support varies somewhat in different parts of the core area. According to the data on standard deviation, a party easily suffers losses farthest from the center and in the border regions of the core area where the votor will more easily be subject to cross pressures than in the center or in the centrally located parts of the core area. Most of the exceptions to the rule occur in the core areas of the leftist groups, for which there may be two interdependent explanations: the parties in question acquire their chief support from among a single social class, i.e. the working class, and the whole political field — its strong as well as its weak points — became the scene of special competition in those two stages, the elections of 1922 and 1945, when the communist movement was looking for its place in Finnish society.

Although the center with respect to degree of support is not a model for the other communes of the core area, it can however be observed to have some part in the distribution and variation of the power of the group in the various parts of the area.

The borders of the core areas

The comment may be made that one should not speak of political areas because one should then also speak of their borders, and it can by no means always be clearly shown where the outer edge of an area runs. In estimating the extent of the core areas below, the attempt is made to avoid an absolute commitment regarding the border linnes of an area whenever the group is also relatively strong in nearby communes. In contrast, the clear weakening of a party's strength immediately outside the core area provides a better opportunity (a) of specifying the border of an area. In these instances another essential problem is associated with the extent of the core area: (b) to what degree do the core areas coincide with certain administrative, economic and natural geographic regional units.

As could be expected, the core areas do not completely follow either
administrative or economic borders, but political and non-political areas to
some extent interrelated.

Eleven core areas (out of 18) are located in their entirety inside one constituency, two extend to the area of 3 constituencies, and one extends to 4. In the eleven instances, the connection is further accentuated by the fact that the borders of the core area in many places coincide with the borders of the constituency. An outright unusal political-ecological phenomenon is surely the indentification that has taken place at the Communist core area and in many parts of the Western Kuopio constituency.²⁵ A similar example can be found in Sweden where the county of Norrbotten coincides fairly precisely with a red area, and the county of Västbotten with a white one." The relationship between core areas and constituencies in Finland has been further strengthened as result of the constituency reforms carried out in 1960. For instance in central Finland and in Ostrobothnia the administrative borders have, to an even greater extent than previously, approached the political borders. This bringing closer together of constituency and core area has been promoted by the constituency's being the administrative unit in parliamentary elections, which has in turn led to the concentration of the activities of the party district organizations upon the same area.²⁷

The communes of a core area stand on common ground also in the respect that fifteen of them are wthin a local economic region. Apparently the core areas have partly come about with consideration to factors that have influenced the shaping of economic regions, such as traditional communal links, uniformity of circumstances, a good contact network and centres of activity that dominate the region.

It is understandable from what established above about the contribution of lakes, rivers, railways and roads toward the transmission of political information, that these natural geographic boundaries lie rather in the inner parts of the core areas than at the edges.

The Political Power Relation» of Core Areas

It seems apparent from the two basic climates and the general political structure of the country that the basic color of a political area would have a role to play in (a) what other parties receive response in the area, and (b) to what degree they do so and (c) how the power relations vary with the geographic location of the core areas. Besides the mutual comparison of political tendencies, there is cause here to ascertain any changes between the different areas of the same group.

The strength and the order of magnitude of the groups studied in core
areas of different types appear from the following tabulation, the figures of
which represent the medians:

Although the core areas bolong to the politically most homogeneous regions, one should beware of considering them to be uniform to such an extent that the group in question would rule alone in the area, the roles of the others remaining non-existent or negligible. Every group examined has at least some amount of support in the core areas of other groups and, further, a fact that does not appear from the figures presented, in every core area.

With the exception of the Communists, the groups studied occupy the position of the largest political group in their core areas. But in light of the median figures only the Agrarian Union and the rest of the Bourgeoisie have achieved absolute majority. Because of the superiority of the Agrarian Union and the rest of the Bourgeoisie, the core areas of the bourgeois groups are more homogeneous in polilical structure for it is these areas that were characterised above as clima.tes.

Ideological relationships do not determine the order of the groups in each other's core areas. True, the rest of the Bourgeoisie occupy second place in the areas of the bourgeois-spirited Agrarian Union, but the Agrarian Union takes last place in the strong commune group of the rest of the bourgeoisie. Because the leftist groups Eire weakest in each other's core areas, citizens belonging to the Bourgeoisi«) or to the Left give their chief support — depending upon the area — to one sub-group and do not distribute their votes between the two. A close regional connection can further be observed even in the light of this examination, on the one hand between the rest of the Bourgeoisie and the Social Democrats, and on the other hand between the Agrarian Union and the Communists.

The position of a group in a core area of competitive tendencies varies
also somewhat with how close its own strongholds are. If the distance is

short, the group will improve its proportion and its rank although the question may be one of the core area of an ideologically distant group. In addition, within a core area too a group often obtains its highest proportion of the vote in a direction where it is strong itself.

An analysis of the strength relations of core areas shows that regionalism on the political map does not only appear in the tendency of communes of strong and weak support in their own direction. The total structure of the political field also has an influence on what position a group obtains in areas dominated by its opponents. The political field is consequently not made up of separate and detached parts but is a dynamic whole, the parts of which are in intimate contact and interrelation with each other.

Conclusions

(a) A highly developed regionalism is characteristic of the political map of Finland. Both within a narrower and within a wider framework the regional differences of support appear in such a manner that communes of stronger support belong to a group of their own which is moreover fairly cohesive, while the communes of weaker support also make up a group. The country as a whole is traditionally divided into two regions of political climate, the more radical northern and eastern Finland where the more radical groups of Bourgeoisie and Leftists (Agrarian and Communists) have more popularity, and the more conservative groups of Bourgeoisie and Leftists (the rest of the Bourgeoisie and the Social Democrats) are in stronger positions. But within these general regions of political climate, there were further discerned smaller areas that were particularly favorable to one or an other group and that differend fairly clearly from the areas surrounding them: the Agrarian Union finds itself strongest in southeast and northern Finland, the rest of the Bourgeoisie on the southern and western coasts, the Social Democrats in a rectangle stretching from southwest to northeast, and the Communists most clearly in the north of Finland.

As these clear-cut differences exist on the Finnish map, and as one and the same party — such as the Social Democrats and the Communists — is able to receive support from quite differing conditions, these observations have provided starting points for other ecological studies. In order to demonstrate the valid explanatory factors in the various parts of the country, Erik Allardt in his analyses of Communism has examined first the situation in the country as a whole, thereafter in the zones of political climate referred to above — although only with constituency accuracy — and finally in five different types of area (cities and boroughs, Swedish-speaking rural communes, the rural communes of the south and the west, the rural communes of the east, the rural communes of the north).²⁸ The two latter divisions are based on administrative borders, which to some extent breaks the natural political entities.

In the political-ecological cartography of Finland, the object of the analysis, besides the country as a whole, is the political areas in particular, which are based exclusively on the differences observed in voting behavior. From the study, we have the impression that the position and significance of the political areas is to some extent woakening. The decreased evaluation of tradition in politics, the equalization of economic differences through the progress of Industrialization and the consequent weakening of regional contrasts, the more dominant position of national issues, and the parties' increasing rejection of regional ties in their changing party program, break up the foundations of political areas and equalize, in this respect also, the differences between the various parts of the country.²⁹ But this change is taking place slowly. So far the regions still dominate the political map, providing also the bases for further analyses.

(b) The ascertaining of the degree and stability of support which are necessary bases in the determination of political areas was done in a study regarding the bourgeois parties at the parliamentary elections of 1919—58, when the bourgeois framework of the party system remained nearly unchanged; and with regard to the left in the elections of 1922-29 and 1945-58, when the Communist group in Finland participated in elections. The mutual commensurability of the bourgeois and the leftist groups does not suffer to any degree from the differences of time periods since the map of their support has been relatively static. The support of a group was determined as the proportion of the valid votes cast in a commune, and the parameter illustrating the size of a group in a commune was the arithmetic mean of the support percentage of all the elections, the support being determined as static if the strength of a group in the elections of the observation period remained above a requisite limit.

Two criteria for the determination of the political area were used in the study, the general political areas, by which the conditions of support of all parties were observed, and the internal areas of the parties, in which the question was one of a single political group. The general political areas were based either on absolute majority or on the situation as largest group. Areas of absolute majority, all of which are bourgeois in character, are dominated by the group in question to the extent that the group has tended to grow at the expense of smaller groups and has ridden out, almost unscathed, periods of political adversity. On the map showing the largest groups, which gives a clear picture of the political top level throuhout the country, not only the Agrarians and the rest of the Bourgeoisie but also the Social Democrats were entered. In this instance it is no longer possible to speak wholly about dominant parties. Depending upon the objects of the ecological study, there is cause to give careful consideration to the limit and the grounds employed when general political areas are being drawn up.

When there is a more searching analysis of the political field and especially
when the object of examination is the political ecology of a multiparty

country, the internal areas of the parties provide essential information on the problem. The fourth group examined, the Communists, appeared only in this examination. The study distinguished between support areas of parties and core areas within these, and also the zone or blockshaped areas of influence that linked up with the core areas. The national median level seems to be suitable for the lower limit of a support area, and the upper quartile level for that of a core area.

Two factors may be mentioned with regard to regional conglomeration of support. The Agrarians and the Communists had only two support areas, the Social Democrats three — which are indeed located in close proximity to one another — and the rest of the Bourgeoisie (a grouping of several parties) had five. Further, outside the regional entities in question, there were left throughout the country only a few communes of corresponding support level, and most of these were close to the main areas.

As the political areas are quite uniform entities, it is of interest to analyze their internal structure too. In this study, in which observations were made of the core areas, three aspects were kept in view: the centers of activities of the area, the borders and the political strength relations. Within its area of strength, a party is much better equipped than it is in an area of equal competition or where support for it is weak. Party offices, salaried employees and newspapers are located in the core area, and many of the group's Parliament members also live there. Most of this force is found at the same place, which thus becomes a local management and organizational center of the party. The center - or centers - thus appears to be an essential part of a core area's mechanism and possibly more generally of the mechanism of a political area.

By nature the political core areas are closely associated with non-political regional entities such as constituencies and economic regions and in many cases even with their borders. This association is probably not due to chance: certain junctures can probably be found in the area background. Nevertheless the core areas did not have their borders at such geographical features as lakes, rivers, railways or roads, which in their capacity as former or present routes of advance of political ideologies were rather found inside the core area. The political strength relations of core areas revealed surprisingly great regional ties in the political field. Not only the strength of groups but also their placing varied in different core areas and within the same core area, according to the direction in which the areas of strength of any group lay and according to the proximity of the latter. The political map is consequently not only divided into clear regional entities, but these entities moreover radiate their influence into the surroundings. When limited areas are being analyzed ecologically, it does not consequently seem expedient to separate them from their surroundings and connections because a great deal of essential information may be obtained from outside the area to explain the area's political structure.

---

NOTES

1 André Siegfried, Tableau politique de la France de l'Ouest sous la Troisiéme République, Paris: 1913, p. V.

2 For further information on this research project, the reader is referred to the article by Pertti Pesonen and Onni Rantala, "Current Election Studies in Finland", in the Review Section of the present volume.

3 Rudolf Heberle, "Principles of Political Ecology". Soziologische Forschung in unserer Zeit. Leopold von Wiese zum 75. Geburtstag dargebracht. Köln: Westdeutsche Verlag, 1951, p. 190.

4 Risto Sänkiaho, Uudenmaan läänin vaalipiirin poliittiset alueet vuosina 1907— 1927. (Political Areas of the Constituency of Uusimaa in 1907—1927). Licentiate dissertation in political science, University of Helsinki, 1967 (mimeo), p. 45. Sänkiaho defined the general political areas by means of inverted factor analysis.

5 Seymour Martin Lipset, Political Man, London: Mercury books no. 43, 1964, pp. 274, 302-303.

6 Cf. Stein Rokkan — Henry Valen, "Regional Contrasts in Norwegian Politics: A Review of Data from Official Statistics and Systems". In E. Allardt — Y. Littunen (eds.), Cleavages, Ideologies and Party Systems. Turku: Transactions of the Westermarck Society Vol. X, 1964, pp. 197—198.

7 Communes having received city charters by 1958 — altogether 38 — are included. Boroughs have in the study been included among the rural communes.

8 The question is one of 55 communes which in parts border on cities. The zone in question is not identical with the city's area of influence.

9 The language situation was determined in the study by the situation in 1930.

10 A similar case is to be :x>und in France where the population living in the proximity of the unstable eastern border is politically oriented to the right. Francois Goguel, Géographie des éleciions frangaises de 1970 å 1951, Paris: Cahiers No. 27. p. 131.

11 Martti Noponen, Kansant'dustajien sosiaahnen tausta Suomessa. (The Social Background of the Finnish Members of Parliament). Helsinki WSOY, 1964, p. 110.

12 Olavi Borg, Suomen puolueideologiat. (The Ideologies of the Finnish Political Parties). Helsinki: WSOY, 1964, pp. 266-267.

13 Onni Rantala, Suomen poliittiset alueet I. Poliittisten aatteiden levinneisyys 1907—1958. (The Political Regions in Finland I. The Spread of Political Ideas, 1907--1958). (mimeo). Turun Yliopiston valtio-opin laitos. Tutkimuksia Sarja C N:o 3, 1965, p. 35.

14 Göran von Bonsdorff, Studier rörande den moderna liberalismen i de nordiska länderna. {Studies on Modern Liberalism in the Northern Countries). Ekenäs: Skrifter utgivna av Fahlbeckska Stiftelsen, 1954, pp. 133—134.

15 Hannu Soikkanen, Sosialismin tulo Suomeen. (The Arrival of Socialism in Finland). Porvoo: WSOY, 1961, p. 344.

16 Sven Rydenfelt, Kommunismen i Sverige, (Communism in Sweden). Kristianstad: Gleerupska Universitetsbokhar dem, 1954, p. 219.

17 Cf. Eino Jutikkala, "Political Parties in the Elections of deputies to the Estate of Burgesses and the Estate Farmers in the Finnish diet Estates". Helsinki: Sitzungsberichte der finnishen Akademie der Wissenschaften 1960, 1962, pp. 176—177.

18 Erik Allardt, "Social Sources of Finnish Communism: Traditional and Emerging Radicalism". Publication No. 22 of the Institute of Sociology University of Helsinki, 1962, p. 57.

19 Cf. Goguel, op. cit. p. 128.

20 Cf. Soikkanen, op. cit. pp. 368—369.

21 Stuart A. Rice, Quantitative Methods in Politics, New York: 1928, p. 136; Rydenfelt, op. cit. p. 83; Soikkanen, op. cit. pp. 368—369.

22 The basis was the list of domiciles of the members of Parliament in the elections of 1919—1939 because the members of the earlier period evidently have had a greater influence on fie determination of center than have those of the later period. Attention is paid to the frequency of election and not to the number of persons elected.

23 Noponen, op. cit. p. 231.

24 Cf. Erik Allardt, Social siruktur och politisk aktivitet, (Social Structure and and Political Activity), Helsingfors: Söderström, 1956, pp. 47—48.

25 Cf. Jaakko Nousiainen, Kommunismi Kuopion läänissä, (Communism in the Kuopio Province), Joensuu.: 1956, p. 58.

26 Rvdenfelt, OP. cit. pp. 208, 242.

27 Cf. —16—19.

28 Allardt, Social struktur.. ., pp. 56—57. Allardt, "Institutionalized Versus Diffuse Support of Radical Political Movements". Publication No. 27 of the Institute of Sociology, University of Helsinki. 1962, p. 371.

29 Cf. Goeuel. on. cit. n. 134. Linset. OD. cit. D. 303. Rokkan — Valen. OD. cit. D. 179.