Transformation of the confessional structure of the population of Northwest Russia and the Baltics (late 19th — first quarter of the 21st centuries)

Introduction

The study of the dynamics of the confessional structure of the population has gained particular relevance in the context of significant transformations in the religious sphere across many countries. These changes are largely driven by secularisation processes that began in the 20th century and intensified in the 21st century. As Gorokhov [1] notes, all countries located east of the Baltic Sea experienced a wave of political secularisation during the Soviet period. Toward the end of the 20th century, this was followed by a brief phase of desecularisation — most notably in Russia and Lithuania, and to a lesser extent in Latvia and Estonia. However, in the early 21st century, the Baltic States (the Baltics) underwent a renewed acceleration of social secularisation, consistent with broader pan-­European trends. This development has been particularly evident in Protestant-­majority countries, with Estonia now ranking among the least religious countries in the world [2—4].

To a lesser extent, pan-­European secularisation affected Lithuania, whose confessional structure of the population is homogeneous and consists predominantly of Latin Catholics. To a greater extent, this phenomenon is noticeable in the Lutheran countries of Estonia and Latvia, which have large Orthodox communities. In the latter of these countries, there are also large communities of representatives of the Roman Catholic Church [5].

In the past, several regions of Northwest Russia exhibited a rather complex confessional structure, with a notable presence of Protestants in the northern areas and Jews in the southern areas [6]. Today, however, the confessional landscape in this part of the country can be regarded as predominantly homogeneous due to the dominance of the potentially Orthodox population. Compared to the confessional geography reflected in the First General Population Census of 1897, significant changes in the religious structure of the population have occurred not only in Northwest Russia [7], but also across the territory of the Baltic States [8].

Thus, the transformation of the confessional structure of the population in the countries situated east of the Baltic Sea reveals significant differences linked to their traditional religious affiliations. The present study aims to identify and analyse shifts in the confessional composition of the population in Northwest Russia, Estonia, Latvia, and Lithuania — examined at the level of the historical-­confessional regions delineated in the course of the research — over the period from 1897 to the present.

Current research landscape and literature review

Research has predominantly focused on the confessional geography of Northwest Russia and the Baltic region in the late 19th century, owing to the availability of comprehensive data from the 1897 census—the sole census to document religious affiliation across these areas. Territorial differences in the confessional structure of the population, based on the results of this census, are examined in the work of Safronov [9] and several of our studies [6; 8; 10; 11]. Also worthy of attention is the work of Kabuzan [12], who analyses confessional statistics at the level of the provinces of the Russian Empire based on the results of population revisions throughout the 18th—19th centuries.

The examination of subsequent changes in the confessional structure of the population of the region analysed was based on works devoted to the religious policy of the Russian Empire in the Baltic provinces [13], the history of the Estonian Evangelical Lutheran Church [14] and the Orthodox Church in the Baltics in the 19th and early 20th centuries [15—17], changes in the confessional structure of the population of the Belarusian-­Baltic region at the beginning of the 20th century [18], and the modern confessional structure of the population of the Baltic countries [3; 5; 19, etc.].

Previously, we sought to assess the current religious structure of the population in Northwest Russia by employing both the method of calculating the potential religious composition [9] and the findings of the 2012 regional population survey conducted across various Russian regions [20; 21]. Comparable methodological challenges arise in studying the religious structure of the population at the regional level in Latvia, as contemporary national censuses in the country do not include questions regarding religious affiliation.

The study of the transformation of the confessional structure in the region under consideration — from the late 19th century to the present — encounters the challenge of discrepancies between the administrative-­territorial divisions of the Russian Empire and those of the contemporary period. This issue is addressed in the present study through the historical and confessional zoning of Northwest Russia and the Baltic States, which takes into account the specific features of the confessional composition of the population in both the late 19th century and the present day.

Materials and methods

The empirical basis of the study consists of population census data and sociological surveys. Specifically, it includes the First General Population Census of the Russian Empire (1897), as well as the national censuses conducted in 2021 in Russia, Estonia, Latvia, and Lithuania. The results of these censuses are available on the website Population Statistics of Eastern Europe & Former USSR.1 Along with the national composition of the population in Estonia and Lithuania, their religious affiliation was also recorded during these censuses.

In addition, the study draws on the results of sociological surveys conducted in Russia and Latvia that included questions on religious affiliation. In Russia, such a survey, which allowed a fairly objective assessment of the confessional composition of the population of almost all regions of the country, was carried out by the non-profit research service “Sreda” within the framework of the “Arena” project in 2012.2 Taking into account the studies we conducted earlier, where the results of this survey were analyzed [20; 21], respondents who gave the answers “I profess Orthodoxy…” and “I profess Christianity…” were classified as the Eastern Christian population (Orthodox and Old Believers), and respondents who answered “I do not believe in God” and “I believe in God, but I do not profess a specific religion” were classified as the non-religious population.

In Latvia, the results of a sociological survey of the country’ s population conducted by the Latvian agency SKDS in 2018 were taken as a basis (the survey was conducted among residents of Latvia aged 18 to 74, a total of about 1000 respondents; the question was asked: “What faith do you belong to?”).3

Due to the lack of modern confessional statistics at the level of lower administrative-­territorial units of Russia and Latvia, the study provides an assessment of the confessional structure based on the ethnic composition of the population. More precisely, the potential confessional structure of the population was calculated based on the traditional religious affiliation of the ethnic groups living in these administrative-­territorial units. This approach was used, for example, in the works of Safronov [9], Manakov and Dementyev [8], Terenina and Krotok [21].

Due to the current existence of a fairly large category of non-religious population, a preliminary assessment of the religiosity of different ethnic groups is required [22]. Thus, in Russia, the results of a population survey conducted by the non-profit research service “Sreda” within the framework of the “Arena” project actually allow us to determine the religiosity of the potentially Orthodox population in the regions of the North-­West due to their mono-ethnicity. An assessment of the size of other religious groups in the event that the survey yielded results that fall within the margin of error of the sample is given on the basis of the ethnic structure of the population.

The assessment of the current confessional structure of the population of the regions of Latvia is more complex. It is based, firstly, on the potential confessional structure of the population of the country’ s regions based on their ethnic composition in 2021, secondly, on the results of a sociological survey in the country, which made it possible to assess the degree of religiosity of each potential confessional group, and thirdly, on the results of previous population censuses in Latvia (from 1920 to 1935, when the religious affiliation of the population was recorded). The latter made it possible to adjust the confessional structure of the population of the regions, taking into account the difference between the potential and actual confessional structure, or more precisely, to determine the share of Catholic Latvians and Orthodox Latvians, which was characterized by fairly high stability regardless of the year of the census.

Research results and discussion

The main indicator used in the study to assess the degree of homogeneity/heterogeneity of the confessional structure of the population is the confessional mosaic index (CMI). This indicator is calculated using the formula CMI = πi(1 – πi), where πi is the proportion of representatives of the i-th confessional group in the country or region. CMI is an analogue of the ethnic mosaic index (EMI) [23], which is widely used in Russian ethnic geography and ethno-­demography. In foreign science, this indicator is called the ethnic fractionalization index [24] or, as in the original version, the ethnolinguistic fractionalization index [25]. Most often in foreign science, the index is used to identify the relationship between the ethnic heterogeneity of the population of countries and regions and their economic development [26—28]. The name of the indicator was borrowed from foreign colleagues by Russian economists engaged in similar research topics [29—31].

The value of the CMI varies from zero (if the entire population of a country or region professes one religion) to 1—1/n (which indicates the maximum heterogeneity of the confessional structure of the population of a certain territory), that is, there is an implicit dependence of the value of the index on the number of confessional groups in the country or region used for its calculation. To neutralise the influence of this factor, Gorokhov [1, p. 102] proposed a modified version of the index, the values of which fall within the range from zero to one. If the value of the CMI is known, the modified version of the mosaic index (CMIm) can be calculated as follows: CMIm = CMI / (1—1/n).

CMIm allows for the assessment of the degree of homogeneity / heterogeneity of the confessional structure of the population of territories. Thus, Gorokhov [1, p. 106] proposed the following scale of CMIm values, which makes it possible to identify territories with different degrees of confessional homogeneity / heterogeneity: 1) extremely homogeneous (CMIm from 0.000 to 0.280); 2) relatively homogeneous (CMIm from 0.281 to 0.556); 3) relatively heterogeneous (CMIm from 0.557 to 0.820); 4) extremely heterogeneous (CMIm from 0.821 to 1.000).

The CMIm can also be used to assess shifts in the confessional structure of the population of territories over a certain time interval. A decrease in the CMIm value over the period under study indicates homogenization of the confessional structure of the population of administrative-­territorial units, while an increase in the CMIm indicates its heterogenization. Thus, a study was conducted in Latvia using a similar methodology [32], where, based on the calculation of the dynamics of the ethnic fractionalization index between 1925 and 1935, as well as between 1989 and 2009—2011, taking into account the change in the share of the Latvian population, an assessment was made of the degree of ethnic homogenization or heterogenization of the lower administrative-­territorial units of the country during these periods.

Given the significant proportion of atheists and individuals who either do not identify with any religion or choose not to disclose their religious affiliation — collectively referred to as the non-religious population — a question arises as to whether this demographic should be included in the calculation of the confessional mosaic index. It is more correct to calculate the CMIm only for the religious part of the population, taking the proportion of people who indicated their religious affiliation as 100 %. For example, Gorokhov [1] suggests calculating the indices of general and religious mosaic separately. We propose to designate the mosaic index for the entire population, including its non-religious part, in order not to confuse it with the CMIm, as a modified general index of confessional mosaic (CMImg).

The study area covers the historical part of Northwest of Russia and the Baltics. However, the Kaliningrad region and the northern part of the Kare­lian Isthmus (in 1897 part of the Vyborg Governorate, and now part of the Le­ningrad region) were excluded from the study due to the complete replacement of the population in their territory during the period analysed. In this territory, 20 historical-­confessional regions and 4 cities (Saint Petersburg, Tallinn, Riga and Vilnius) were identified, characterised by a specific confessional structure of the population at the turn of the 19th and 20th centuries or at present.

Based on the results of the 1897 and 2021 censuses, the CMIm was calculated for all historical and religious districts of the region under study. For 2021, the CMImg — an adjusted version of the index that includes the non-religious population — was also calculated. The results are presented in the table below.

The value of the confessional mosaic indices for historical and confessional regions of Northwest Russia and the Baltics, calculated for 1897 and 2021

In 1897, based on the values of the Confessional Mosaic Index (CMI), the historical-­confessional regions were classified according to the degree of homogeneity or heterogeneity of their confessional structure as follows: extremely homogeneous regions — most of Northwest Russia and northern Estonia; relatively homogeneous regions — St. Petersburg, Priluzhie, and most territories of Estonia, Latvia, and Lithuania; relatively heterogeneous regions — Ingria, southwest Es­tonia, the city of Riga, Latgale, and Eastern Aukštaitija. The only region classified as extremely heterogeneous was the city of Vilnius.

Currently, nearly all historical-­confessional regions exhibit a high degree of heterogeneity in the overall confessional structure of the population, primarily due to the substantial share of the non-religious population. At the same time, the degree of heterogeneity within the religious segment of the population has slightly decreased. The greatest increase in the homogeneity of this segment is observed in the regions of Northwest Russia and Lithuania. In contrast, the religious structure has become somewhat more heterogeneous in most historical-­confessional regions of Estonia (with the exception of Ida-­Virumaa County and South-­West Estonia) and Latvia (except for Latgale). The observed dynamics of the Confessional Mosaic Index can be explained by examining the changing proportions of various confessional groups over the period in question.

The main confessional groups selected for analysis in the region under study are Protestants (comprising Lutherans alongside representatives of other Pro­testant denominations), Catholics, Eastern Christians (including both Orthodox believers and Old Believers), and Jews. Firstly, these groups correspond either to individual religions (Judaism) or to the three principal branches of Christianity. Secondly, these confessional groups represented the largest shares within the religious composition of the population in the study region at the end of the 19th century. Thirdly, they allow for quantitative comparison with the potential confessional structure derived from the traditional religious affiliations associa­ted with the ethnic groups in the area.

According to the data utilized in this study, the share of the non-religious population — which includes atheists as well as individuals who did not specify their religious affiliation during censuses and population surveys — is currently 71 % in Estonia, 32 % in Latvia, 20 % in Lithuania, and 43 % in the regions of Northwest Russia. Within Northwest Russia, the proportions vary by area: over 48 % in the Novgorod region, 44 % in St. Petersburg, over 40 % in the Pskov region, and 39 % in the Leningrad region.

The decrease in the proportion of Protestants in the overall confessional structure of the population of the Baltic States is directly related to the processes of social secularisation that unfolded in the 20th and early 21st centuries, since secularisation primarily affected Protestant countries. In the period between the 1897 and 2021 censuses, the most significant reduction in the proportion of Protestants (over 70 %) occurred in Estonia (Fig. 1), which is currently one of the least religious countries in the world.

Fig. 1. Change in the share of Protestants in the total confessional structure
of the population between 1897 and 2021

A 50—70 % decrease in the proportion of Protestants is characteristic of the historically Lutheran regions of Latvia (Vidzeme and Zemgale), the city of Riga, as well as southwestern Estonia and Tallinn. In Russia, the most significant decrease in the proportion of Protestants occurred in the territory of Ingria (due to the deportation of Ingrian Finns in the 1930s—1940s), in Priluzhie and Eastern Prichudye (abandoned in 1943 by Estonians and Latvians who arrived in these territories in the second half of the 19th century) [6]. The minimal decrease in the proportion of Protestants characterises the territories of Russia and Lithuania, where the proportion of Lutherans was low to begin with. The only historical region where the proportion of Protestants increased during the study period was Dzūkija in Lithuania.

The share of Catholics in the overall confessional structure of the population grew mainly in the Baltic States and decreased in Northwest Russia (Fig. 2). The largest increase in the share of Catholics during the study period was observed in Vilnius and Riga, and less significant in most of Estonia, Latvia, and Lithuania. The exception was several regions where the share of Catholics decreased due to the growth of the share of Orthodox (in Latgale, Eastern Aukštaitija, and Samogitia) or the non-religious population.

Fig. 2. Change in the share of Catholics in the total confessional structure
of the population between 1897 and 2021

The most significant increase in the proportion of Orthodox and Old Believers (in total) in the overall confessional structure of the population was observed in Eastern Virumaa (Estonia) and the city of Riga (over 20 %), as well as in all regions of Latvia (from 10 to 20 %), a small increase — in Northern Estonia and the city of Tallinn, in Eastern Aukštaitija and Samogitia. A decrease in the proportion of Eastern Christians occurred in all regions of North-­West Russia (due to an increase in the proportion of the non-religious population), in the southern part of Estonia and in most regions of Lithuania (Fig. 3).

Fig. 3. Change in the share of Orthodox and Old Believers in the overall confessional structure of the population between 1897 and 2021

The decline in the proportion of Orthodox Christians and Old Believers in Lithuania is primarily attributable to the reduction of the potentially Orthodox population, largely resulting from partial migration outflows at the turn of the twentieth and twenty-­first centuries. In contrast, the decrease in the share of Eastern Christians in southern Estonia, particularly in its southwestern region, stems from a different cause. During the mid to late 19th century, Orthodoxy — often referred to as “the Tsar’ s faith” — was actively adopted by the local Estonian population in the hope of acquiring land following the abolition of serfdom in the Baltic provinces [15—17]. The island of Saaremaa (then Oesel) was particularly notable in this regard, with the proportion of Orthodox Christians exceeding 38 % of the local population in 1897. Similar processes occurred in other counties of the Livonian province — now corresponding to southern Estonia and northern Latvia — although the adoption of Orthodoxy by Estonians and Latvians in these areas was less widespread.

The schematic map illustrating the reduction in the share of Jews within the overall confessional structure of the population (Fig. 4) largely reflects the historical geography of Jewish settlement at the end of the 19th century, specifically within the so-called Pale of Settlement. By the present day, the Jewish population in the study region has become exceedingly small. This decline is primarily attributable to two factors: first, the genocide of Jews in the occupied territories during the Second World War, and second, the post-war migration outflow of the already greatly diminished Jewish community. At the end of the 19th century, the Pale of Settlement encompassed the Vitebsk Governorate (including Latgale and the modern southern part of the Pskov Region), as well as the Kovno, Vilnius, and Suwalki Governorates — territories that later fully or partially became part of Lithuania. Additionally, special residence regulations for Jews were imposed in the Courland Governorate (corresponding to Kurzeme and Zemgale). Outside these areas, except for major cities such as St. Petersburg, Revel, and Riga, the Jewish population—and thus the presence of Judeans — was insignificant.

Based on the difference in the CMIm and CMImg values between 1897 and 2021, as well as on the changes in this period in the general confessional structure of the population of the proportion of Protestants, Catholics, Eastern Christians, Jews and non-religious population, a typology of historical and confessional regions of Northwest Russia and the Baltic States was developed based on confessional dynamics in this period (Fig. 5).

In total, three types and eight subtypes of historical-­confessional districts were identified. First of all, the districts were divided into two main groups by the nature of the change in the general confessional structure of the population (including the non-religious population). Most often, an increase in the share of the non-religious population led to the heterogenization of the general confessional structure (increase in CMImg), but it was accompanied by either homogenization of the structure of the religious population (decrease in CMIm), or its heterogenization (growth of CMIm). These are the first two types of historical-­confessional districts, each of which has three subtypes depending on the dynamics of the proportion of individual confessional groups.

Fig. 4. Change in the share of Jewish population in the total confessional structure
of the population between 1897 and 2021

The first subtype of the first type includes historical-­confessional areas that experienced a homogenization of the religious population structure — despite a growing share of the non-religious population — due to a decline in the representation of other confessional groups. These areas encompass the entire territory of Northwest Russia, as well as Eastern Virumaa in Estonia and Latgale in Latvia. In these regions, Orthodoxy has retained or established dominance within the religious population over the study period. This outcome is attributable, in part, to the reduction in the proportion of Jews (notably in Latgale and the southern part of the present-day Pskov region, which historically belonged to the Vitebsk Governorate within the Pale of Settlement), as well as to the decline of Protestant communities. The latter trend is evident in Eastern Virumaa, Priluzhie, Eastern Prichudye, Ingria, and St. Petersburg — areas where Estonians and Finns previously formed a significant part of the population.

Fig. 5. Typology of historical and confessional regions of Northwest Russia
and the Baltic States based on the dynamics of the confessional structure
of the population between 1897 and 2021

The second subtype of the first type is characterised by an increase in the share of the Orthodox population; however, Orthodox believers did not come to dominate the religious landscape, as Catholicism retained its leading position. In these regions, the homogenization of the religious population structure was largely driven by the decline of the once substantial Jewish population. This subtype is represented by two historical-­confessional regions in Lithuania: Samogitia and Eastern Aukštaitija.

The third subtype of the first type is also characterised by the homogenization of the religious population structure in favour of Catholics, again due to a signi­ficant reduction in the share of Jews, but without an increase in the share of the Orthodox population. This subtype is represented in the eastern part of Lithuania (Vilnius and the surrounding region of Dzūkija).

The second type of historical-­confessional districts is generally characterised by the heterogenization of both the general confessional structure of the population and its religious component. However, in this type of districts, there is a small increase in the share of two confessional groups — Catholics (the first subtype, represented in the south-east of Estonia), Orthodox (the second subtype, including the city of Tallinn) and both Catholics and Orthodox (the third subtype, covering Northern Estonia and all historical districts of Latvia, except Latgale).

The third type of historical-­confessional regions is characterised by the simultaneous homogenization of the general confessional and religious structures of the population. The first subtype is represented by two historical regions of Lithuania (Aukštaitija and Suvalkija), where the structure of the religious population was homogenized in favour of Catholics, who strengthened their leadership not only in the religious structure of the population, but also in the general confessional structure (despite the growth of the share of the non-religious population) due to the loss of Jews from it.

The second subtype of the third type includes the southwestern part of Estonia. This is the only historical-­confessional region where there was a simultaneous homogenization of the religious population structure in favour of Protestants (due to the decline in the share of Eastern Christians) and the general confessional structure in favour of the non-religious population.

Conclusions

The methodology developed in this study—based on the analysis of changes in the proportions of major confessional groups and the calculation of the mosaic index across historically and confessionally defined regions — made it possible to identify territorial variations in the transformation of the confessional structure of the population in North-­West Russia and the Baltic countries from the late 19th century to the present. In total, 24 regions were identified, each characterised by distinctive features of the confessional structure of the population either at the end of the 19th century or at the beginning of the 21st century.

A typology of historical and confessional districts was carried out based on the dynamics of the confessional structure of the population for the specified period. Three types and eight subtypes of districts were identified, taking into account the change in the complexity of the religious and general confessional (including persons who did not indicate their religious affiliation) structure of the population, as well as the dynamics of the proportion of the four main confessional groups in the study region (Protestants, Catholics, Eastern Christians and Jews).

The first typе is characterised by the homogenization of the religious structure (in favour of either Orthodoxy or Catholicism), occurring alongside the heterogenisation of the broader confessional landscape due to the increasing share of the non-religious population. This trend is observed across all historical-­confessional regions of North-­West Russia, Ida-­Virumaa County in Estonia, the Latgale region in Latvia (where homogenization favoured Orthodoxy), and much of Lithuania (where Catholicism became predominant). The growing dominance of Orthodox and Catholic affiliations within the religious segment of the population can largely be attributed to the demographic decline of Protestant communities in the northern areas and the near disappearance of the Jewish population in the southern regions.

The second type is defined by a concurrent heterogenization of both the religious and overall confessional structure. This pattern is typical of most regions in Estonia and Latvia, where large-­scale secularisation among Protestants was accompanied by modest increases in the shares of Orthodox and Catholic adhe­rents.

The third type is distinguished by the simultaneous homogenization of both religious and overall confessional structures. In Lithuania, this trend is particularly evident in regions such as Aukštaitija and Suvalkija, where the dominance of Catholicism has intensified. The most illustrative case of this type is south-­western Estonia, where a substantial rise in the non-religious population coinci­ded with a marked decrease in the proportion of Orthodox Christians (primarily descendants of 19th-century Estonian converts), leading to a consolidation of Lutheranism as the prevailing religious affiliation.

The study was carried out with the financial support of the Russian Science Foundation within the framework of project № 23-17-00005 “Ethnic contact zones in the post-­Soviet space: genesis, typology, conflict potential”.