Assessment of the reliability of statistical data on fires in the Russian Empire in the middle of the 19th century

Avdeeva Anastasiya Andreevna PhD in History Press Secretary; Federal State Budgetary Institution 'All-Russian Research Institute of Labor' of the Ministry of Labor of the Russian Federation Kazakova str., 29-1, sq. 40, Moscow, 105064, Russia 9153588369@mail.ru

The Russian fire Department has a long history. Starting with the Judicial Code of Ivan III (1497), the regulatory framework for fire legislation was formed step by step. By the middle of the 19th century . The Russian Empire already had a fire fighting system, which was distinguished by the following features. Firstly, the regulatory and legal provision of fire safety covered all stages of the service's work - prevention, liquidation and restoration of affected areas. Secondly, professional training of firefighters has begun in the capital's fire depots. Thirdly, theoretical and practical firefighting skills have received scientific justification. Fourthly, all fire protection facilities of the Russian Empire were brought to uniformity by law and uniform fire protection rules were developed for all types of settlements. Fifth, the first fire insurance company was established and the foundations of national statistics were laid in terms of providing information about fires.

The history of Russian statistics dates back to 1802, when the manifesto "On the Formation of Ministries" was issued, which emphasized that each minister was obliged to submit to the emperor a written report on the powers entrusted to him ^{[1, p. 243]}. In order to carry out the quality work of the ministers in providing various types of reporting to the governors, a special circular was sent (dated September 19, 1802), signed by V.P. Kochubey, Minister of the Interior, on the procedure for providing statistical information ^{[2, p. 8]}. This circular can be considered the first nationwide attempt to conduct a statistical analysis of issues related to the Ministry of Internal Affairs and fire statistics, among others.

Specific recommendations on the structure and forms of presentation of the statistical governor's report were defined in 1804. Received in 1804 The reports published in the statistical collection "Timesheets to the report of the Minister of the Interior for 1804" have already revealed the shortcomings and general shortcomings of the submitted forms, which, unfortunately, were observed in the future, the main of which was the unreliability of the submitted data ^{[3, p. 5]}.

Nevertheless, statistical studies of various aspects of the life of the population, economy, industry and emergency situations in the Russian Empire continued. Gradually, they reached the scientific level. In 1806-1808 . The Statistical Journal was published in 1818-1819. The first scientific work based on the analysis of statistical data from the governors' reports was The Outline of Statistics of the Russian State ^[4]. K.F. German, head of the prototype of the modern Russian State Statistics service, the Statistical Department (Accounting) at the Executive Police Department, noted that the reason for the imperfection of statistical data The reason for the research was the formal approach to providing and publishing the data obtained, the absence of additional checks, various comparisons and their qualitative analysis ^{[5, p. 445]}.

In 1819, the procedure for submitting these reports changed. On November 28, 1819, it was legislatively determined that police officers, city officials and zemstvo police officers should no longer prepare reports to the Minister of Internal Affairs, but should report only to local authorities ^{[6, p. 399]}. The previous maintenance of statistics on all types of accidents (including fires) was complicated by duplication of information: it was provided to the minister by both the county authorities and the governors.

Since 1835, the formation of provincial statistical committees began, the main work of which consisted in verifying the information received, bringing them to uniformity, and compiling report cards according to special forms. The provincial administration began to play the role of the main link in the chain of institutions responsible for the reliability of the information provided. In 1842, almost all forms of reporting provided by provincial governors to the emperor were updated, and it was emphasized that the issue of accuracy and complete statistical certainty was a priority ^{[2, p. 21]}.

Thus, by the middle of the 19th century. In the Russian Empire, a legislative and executive framework was established not only for the collection, but also for the analysis of statistical information related to economic, demographic and social spheres of life. Fire statistics occupied an extremely important place in these reports. Firstly, the actual fire situation indicators were determined for the first time. Secondly, an attempt was made to assess the level of material damage caused by fires on a national scale. Thirdly, 5 main causes of fires were formulated: from lightning, from faulty pipes and furnaces, from carelessness, from arson and from unknown causes.

Assessing the reliability of statistical data on the basis of which conclusions are drawn is an extremely important aspect in historical retrospect. With regard to fire statistics, the main sources of information were the Most Reliable reports of provincial chiefs on incidents, submitted twice a month, and Incident Reports in Appendices to the Most Reliable Annual Reports. Thus, the responsibility for the accuracy of these data was assigned exclusively to the heads of the provinces.

The institution of the governorship of the Russian Empire, which began in 1708, acquired special features by the middle of the 19th century. The province became the main administrative-territorial unit of the empire, in which there was a unified system of provincial institutions, and the governor was the first person. However, the general imperial system of legislation forbade the heads of provinces to issue normative legal acts binding on the territory entrusted to them. The duration of the governorship was not legally defined either. For example, during the period 1825-1853, Penza Province had 2 governors, and Kostroma Province had 13 ^{[7, pp. 287-313]}. The post of governor could remain vacant for a long time (Tobolsk province - 1832-1835, Grodno province - 1842-1844, Kaluga Province - 1843-1845, etc. ^{[7, pp. 290, 295, 309]}). The names and the number of administrative divisions were changed. For example, in 1842. The Bialystok region was renamed the Kovno province; the Tiflis province was formed in 1846 from the Georgian-Imereti province and the Caspian region, etc. ^{[7, pp. 288, 309]}. New provinces, regions, and townships were established.: Samara province (1850); Kamchatka (1849), Zabaikalsky (1851) regions, Kyakhtensky municipality, etc. ^{[7, pp. 292-293, 296, 305]}. It is particularly necessary to note such administrative units as the Kingdom of Poland (became part of the Russian Empire in 1815 by decision of the Congress of Vienna, controlled through the Governor) and the Grand Duchy of Finland (as part of the Russian Empire since 1809, governed through the Governor-General – Chairman of the Imperial Finnish Senate), which have wide internal and external autonomy..

The administrative and territorial instability of the local government system naturally complicated the process of providing statistical reporting. For example, an assessment of the quality of the reports of the heads of provinces for 1837-1851 by the Department of the Economic Department made it possible to divide all provinces into 4 groups. Full statistical reports were received only from 4 provinces – Astrakhan, Vladimir, Nizhny Novgorod and Estonia. A fairly satisfactory reporting situation, when only a small number of reports were missing, was demonstrated by 17 provinces – Arkhangelsk, Vyatka, Yekaterinoslav, Yenisei, Kostroma, Vologda, Penza, Ryazan, Simbirsk, Kursk, Minsk, Olonets, Perm, Smolensk, Tambov, Chernigov and Yaroslavl. All other provinces were assigned to 2 groups: of which no reports were received in significant numbers (21 provinces) and provinces that were completely faulty in terms of reporting (10 provinces) ^{[8, pp. 152-154]}.

This quality check of the governor's reports assessed not so much the content as the very fact of the presence or absence of reports, which allows us to conclude that in addition to the inaccuracy and unreliability of data on fires, their very absence took place.

However, it is impossible to place all responsibility for the reliability of the data provided on fires solely on the provincial heads, since even the initial data collection was fraught with numerous difficulties. Firstly, only St. Petersburg, Moscow, and 51 cities of the Russian Empire had an approved staff of fire brigades ^{[9, p. 44]} capable of providing the most accurate information about a particular fire. However, statistics on visits to fires in fire brigades were not kept until the end of the 19th century, which was thoroughly proved in the study of gr. A.D. Sheremetev ^[10].

Secondly, the primary collection of fire data, which requires thoroughness and attention to detail, was most often assigned to police officials with a wide range of official duties. According to D.P. Zhuravsky, the facts of delegating the collection of primary information to subordinates were the norm: "In many other official cases, he often entrusts the collection of the required information to his subordinates, the so-called clerks, or messengers, for whom the accuracy of this information is not at all important" ^{[11, p. 157]}.

Thirdly, there are significant differences between the data provided by the governors on emergencies 2 times a month throughout the year and the final set of statistics for the reporting year. The annual reports seem to be the most accurate, as they included all the clarifications from previous weeks.

For the first time, the analysis of statistical data on fires in the Russian Empire was presented in the Journal of the Ministry of Internal Affairs in 1843. The analytical article contained information for the previous year. Thus, all-Russian information about fires can be analyzed since 1842. It should be emphasized that these data did not contain statistics on fires in the Kingdom of Poland and the Grand Duchy of Finland.

The first decade of detailed fire statistics (1842-1851) not only demonstrated the main factual, economic and social indicators of the consequences of fire accidents, but also revealed factors assessing their reliability.

The number of fires per year. Starting with the report for 1842, the data were presented in a single form: separately for each province (region, city government), by month of the reporting year, the total amount for the year in question, and an analysis of data for previous years. Undoubtedly, this indicator is extremely important. He not only talked about the number of fires, but also made it possible to identify regions with increased or decreased fire danger, to mark the fire months. However, already in the first statistical review, it was emphasized that these figures reflect only an approximate picture of the fire situation, since they do not include minor fire incidents ^{[12, p. 431]}. At the same time, it is interesting to note that the number of fires that occurred in St. Petersburg and Moscow also includes minor incidents, as indicated by a special clarification: "... the figure 258 assigned here to the St. Petersburg Capital is made up of all fire cases, not excluding those in which the flame at the first so-to-speak the outbreak was prevented by the rapid arrival of the Police or the timely discretion of the householders themselves..." ^{[13, p. 380]}.

The number of fires caused by arson. Criminal liability for arson in the first half of the 19th century provided for various types of punishments depending on the class. For example, deliberate simple arson of buildings was punished for nobles by deprivation of ranks and nobility and eternal exile to penal servitude without being put on the scaffold. For commoners, it was an eternal exile to hard labor, which was preceded by whipping ^{[14, p. 105]}. In the 50s of the XIX century. Criminal cases of "intentional incendiary acts" were considered by Military courts, which indirectly contributed to a decrease in the number of arson attacks ^{[13, p. 384]}. The number of fires caused by deliberate actions was strictly recorded. The statistical descriptions emphasized that these figures included not only considered and proven cases, but also those in which arson was only implied ^{[15, p. 283]}. Thus, the statistics on arson have a high degree of reliability.

The number of fires caused by lightning. Lightning fires were extremely destructive, not only because of their suddenness, but also because of the superstitions prevailing in Russian society. There are many signs associated with a thunderstorm and the fires it produces. For example, it was believed that a fire from a thunderstorm, as a punishment from God, is not filled with water, but only with milk (although such cases of extinguishing are not described) ^{[16, p. 175]}. And on St. Panteleimon's day on July 27 and the Mother of God of Smolensk on July 28, bread and hay are not mowed or harvested, believing that lightning will burn the bread and hay collected on these days ^{[16, p. 109]}. In this regard, fires caused by lightning were remembered by ordinary people for a long time, which was recorded with a high degree of reliability by local police officials when compiling reports.

The number of fires in towns and villages. Despite the inaccuracies in the total number of fires, the division of fire cases into urban and rural is extremely important. Statistical data from the reports convincingly proved that fires occurred more frequently in cities than in villages, but they were less destructive. Starting with the report for 1848, information about specific fires appears in statistical data (indicating the date, location and number of burned buildings. For example, out of 61 fires that destroyed from 100 to 200 houses in 1848, only 8 occurred in cities ^{[17, pp. 24-28]}. In the reports of the governors, such devastating fires even had their own special name – "wonderful fires".

Fires caused by carelessness. In the middle of the 19th century . This category of fires included fires caused by careless handling of fire, from the poor condition of pipes and furnaces, and fire accidents, the cause of which remained unknown. Since 1846 (report for 1845), statistics are published with the specified division. Within the framework of this study, the third group of fires is of the greatest interest (the causes of which have not been identified). This indicator speaks, first of all, about the quality of the work of the local police. And the data presented in the reports of 1845-1851 demonstrate an extremely low percentage of disclosure of the causes of fires (Table 1).

Table 1. Percentage of the total number of fires per year and fires whose causes have not been identified in the Russian Empire in 1845-1851.

Table 1 shows that the proportion of fires that occurred from unknown causes in 1845-1851 from all fires per year ranges from 65 to 77%. Thus, on average, the causes of about 70% of fires can be attributed to both arson and fires from negligence.

The devastation of fires. The degree of devastation of fires in the middle of the 19th century. It was calculated by the number of burnt houses (buildings). Despite the apparent simplicity of the calculation, this indicator, for a number of reasons, had an extremely low degree of reliability. Firstly, the inaccuracy in the number of burnt buildings was reflected in various methods of statistical accounting: the reports included not only completely burnt buildings, but also partially damaged by fire; there were controversial issues about the unit of damaged buildings – to take into account the number of burnt buildings or burnt yards (households) ^{[24, p. 2]}. Secondly, significant errors were also included in the statistics of burnt shops. As it was noted in the very first report for 1842, the statistical data did not include data that did not contain specific figures (for example, "two rows of shops burned down" or "shops burned down in the settlement of Yelane in Atkarsky district") ^{[12, p. 462]}. However, information about burnt-out liturgical, government, public and industrial buildings had a high degree of reliability.

Economic damage from fires. The devastation of fires in the middle of the 19th century required an assessment of the economic damage to each province (region, city government) and to the Russian Empire as a whole. In the 1950s, fire reports formulated the main indicators of the value of losses: the average value of each fire and each farm, the private losses of each province, and the total damage across the empire. For a number of reasons, the assessment of economic damage could not claim a high degree of reliability. Thus, the damage assessment could vary depending on the province (for example, according to average prices for building materials), information provided by homeowners, etc.

It is also important that a significant number of fires remained unappreciated. For example, out of the total number of fires in the Moscow province that occurred in 1849. (219 units) ^{[21, pp. 160-161]}), 160 remained unappreciated ^{[21, p. 186]} (that is, 73%). At the same time, in the Estland province in the same year, only 2 fires (10%) were designated unappreciated ^{[21, pp. 162-163, 186]}. Overall, of the 7,226 fires recorded in 1849, 1,526 remained unappreciated (21%) ^{[21, pp. 162-163, 185-186]}. To form the total amount of damage in the empire, these cases were calculated according to the average indicators (for the province). Table 2 provides information on the amount of economic damage caused by fires in 1842-1851.

Table 2. Economic damage caused by fires in the Russian Empire in 1842-1851.

Table 2 shows that the average losses of one burned owner and the value of one fire differed by 3 or more times. This means that one fire caused damage to several households. The total value of losses from fires per year was estimated in millions of rubles. For comparison, the annual cost of maintaining firefighters in the Russian Empire, including salaries for fire brigadiers and firefighters, as well as uniforms, provisions, and the needs of guards, and the maintenance of fire horses, amounted to 197,356 rubles (excluding Moscow and St. Petersburg) ^{[9, pp. 48-49]}.

Those who died in fires. The statistics of those killed in fires during the decade under review have undergone changes. Initially (the report for 1842), only factual information with paintings on provinces was posted ^{[12, pp. 464-465]}. After 2 years, statistical information about the dead is supplemented by age and gender division ^{[15, pp. 300-302]}. Detailed information about those killed in fires was also posted annually in the special report "Deaths from negligence, calculated throughout Russia" ^{[18, pp. 76-134]}. However, the number of people killed in fires included only those who died directly in the fire, and excluded those who were injured and died some time later ^{[13, p. 413]}. On average, more than 600 people died in fires every year in the Russian Empire.

The human factor also influenced the assessment of the reliability of information about fires. Typos have been noticed in the submitted reports. For example, in the report for 1843, the table containing information on the value of losses mistakenly indicated 1842 ^{[13, pp. 392-393]}. The structure of the report implies an analytical reference to previous years, so these inaccuracies can be corrected by researchers.

An assessment of the reliability of information about fires in the Russian Empire in the middle of the 19th century demonstrated that their objectivity depended on many factors: the personal responsibility of officials and governors, the speed of conducting criminal cases on fires, and the accuracy of assessing material damage.

The publication of statistical information on fires in the Russian Empire, which began in 1843, despite their relative reliability, was extremely important for understanding the scale and frequency of fires and assessing the economic and demographic damage caused by them. Firstly, the frequency of recurrence of fires was directly related to the population and was almost proportional to the number of yards. Secondly, the intensity of fires was determined by climatic factors and the average size of villages. Thirdly, fires occurred more frequently in cities than in villages, but they were less destructive. Fourth, the largest percentage of fires occurred in the autumn months, and the largest percentage of losses occurred in the summer months.

Thus, it can be argued that the experience of the first decade of providing fire statistics in the Russian Empire proved useful, determined the further development of the statistical service and contributed to the complication of statistical analysis.