War Losses (Austria-Hungary)

Introduction

The exact number of Austro-Hungarian military and civilian war losses during the First World War remains unknown to the present day as the successor states did not cooperate with one another to arrive at reliable figures after the war. With respect to military losses, some work has been done to calculate the total number of losses and to classify those losses in categories such as home region, age, branch of service and so on. Much of this work was completed during the war by a number of military statistical institutions.

Civilian losses, however, remained nearly undiscussed. This was not only a problem of missing data. The authors of the Austrian sections of the Carnegie Series “Social and Economic History of the World War” were ministers or high-ranking officials during the war and attempted to justify their own decisions during the war. Essentially, each author strove to prove that the collapse and dissolution of Austria-Hungary was not his fault. One of the few exceptions was Wilhelm Winkler (1884-1984) who was a population-statistician in a military office during the war and worked with civilian losses after the war.¹ His work is more than problematic. Given that the war reduced births to about half of the peacetime average, Winkler counted the non-begotten (he used the term “unborn”) as losses.² A non-begotten baby cannot be a loss, but Winkler was thinking from the perspective of a “Leviathan” whom the individual life has to serve. For Winkler, wartime human losses were not suffering humans but rather the reduction of the human power of this “Leviathan,” whether of the state or another collective like the nation.

Military Losses

During the war, two independent institutions registered military losses: the first was the “List of Losses” (Verlustliste), compiled by the Department X/VL (Verlustlistengruppe – Group of the List of Losses) of the Ministry of War (Kriegsministerium), which in August 1917 was transferred to the Office of War Statistics (Kriegsstatistisches Büro) attached to the War Archives (Kriegsarchiv). Independently, the Austro-Hungarian High Command (k.u.k.Armeeoberkommando, or AOK) also recorded losses. Both institutions counted differently and came to different figures: in its last report on losses up to September 1918, the AOK registered 499,203 deaths whereas the Kriegsstatistische Büro noted 363,144 deaths at the front and 324,590 deaths in hospitals, altogether 627,534 deaths for the same time. In both cases, the total dead were calculated to include an unknown number of dead who were classified as missing. At the end of the war about 1 million people were missing.

After the war, three people were active in calculating war losses. The first was Winkler. During the war he had been a member of the Scientific Committee of War Economy (Wissenschaftliches Komitee für Kriegswirtschaft, or WKKW) which was part of the Kriegsministerium. After the war, he worked in the Ministry of Military Affairs (Staatsamt für Heerwesen), but joined the Office of Statistics (Bundesamt für Statistik) in 1921. His publication on war dead in 1919 was based on works of the WKKW undertaken in 1917/18.³ The WKKW did not try to fix an exact number of deaths. Instead, it estimated the number of deaths at 1.2 million by the end of 1917. A sample of 120,000 deaths from eight different time periods was classified by home regions (sub-categorized by nationality) as well as age and then grossed up to 1.2 million. This work was marked by a German-national view and, in particular, the classification of the fallen soldiers into national groups is tendentious in its derivation and in its conclusion. For Winkler, however, these results could be used to scale the heroism of the different nations of the monarchy: the higher the losses per capita, the “tougher” the nation. Winkler failed to take into account many factors influencing the intensity of losses and his conclusions are highly flawed. For example, according to Winkler, Bosnia had the fewest losses per capita. Yet, during the first years of war Bosnia faced the worst supply situation of the monarchy and its recruiting results were much below average. This was also a time of high losses as shown in table 2. If there are fewer Bosnian soldiers in times of high losses and Bosnia had the least losses per capita, this figure says nothing about the heroism of Bosnian soldiers.

In addition to Winkler, Gaston Bodart (1867-1940), a Viennese statistician and military historian, dealt with war casualties even before the First World War.⁴ During the war he worked at the Kriegsstatistische Büro. Afterwards, he tried to calculate as closely as possible the real number of casualties from each year and on each front. In around 1921 he finished his manuscript on Austro-Hungarian war losses, though it was never published. According to Bodart, 1,046,893 soldiers were dead, while 332,950 were still missing. Bodart counted half of these as dead, thus arriving at the number of 1,213,368, rounded to 1.2 million. This figure did not include prisoner of war deaths.

Edmund Glaise-Horstenau (1882-1946), a regular officer throughout First World War and an active Nazi between 1938 and 1945, headed a staff to publicize an official Austrian history of the First World War, which appeared in 1930-1938 in seven volumes.⁵ This work includes tables of losses. The “losses at the front” differs slightly from the results of the lists of the AOK and Bodart (539,633 instead of 499,203 fixed by the AOK, and of 521,146 given by Bodart). Including rear areas, “Österreich-Ungarns letzter Krieg” gives a total of 1,016,200 dead, whereas Bodart counted 1,046,893. But only Bodart took into account the estimated number of deaths of those still missing in 1921.

The number of deceased prisoners of war (POWs) can be estimated roughly at 450,000: 385,000 died in Russian captivity; 35,000 in Italian POW camps; 30,000 in Serbia, especially during the great retreat of the Serbian army in autumn 1915; and 3,000 died in Romania. With exception of Italy, the data was very crude, especially that of Russia which only ranged until the beginning of November 1917.⁶

Table 1: Total Deaths of the Austro-Hungarian Armed Forces⁷

Table 2: Reported Deaths at the Fronts by Year (according to Bodart)⁸

Table 3: Deaths by Home Region up to the End of 1917 (according to Winkler)⁹

Civilian Dead

Direct Losses

There are four groups of direct civilian deaths. The first and smallest group includes those who served in the army as civilians. According to the Verlustliste, 107 were killed, 269 were wounded, 374 went missing, and 1,123 were taken as prisoners of war. The second group includes the population of war zones. An unknown number died during the fighting, especially during the first months of the war in Galicia; these losses were neither counted nor estimated during war. Even the statistic about the movement of population (births and deaths) for Galicia, Bucovina and Dalmatia was not compiled after 1913.

The third group includes those murdered or – if one accepts the legal distinction – executed by soldiers or judicial organs, whether military or civil. The Austro-Hungarian army was extremely brutal in Galicia and Bosnia and civilians were executed often without any legal procedures. Estimates during the war ranged up to 80,000 victims.¹⁰ In addition to spontaneous massacres by soldiers, the highest military ranks were also involved in war crimes. During the reoccupation of Galicia in 1915 the commander of the Fourth Army, Joseph Ferdinand, Archduke of Austria (1872-1942), ordered that civilians be hung as a warning to the population that was allegedly cooperating with the enemy.¹¹ In fact, these civilians died because Franz Conrad von Hötzendorf (1852-1925), Chief of AOK 1906-1911 and 1912-1917, needed scapegoats for his erroneous decisions, especially in 1914 on the Eastern Front. In his (and his staff’s) place, his sub-commanders, Czech soldiers, and the population (Ukrainians and Poles) were made responsible for military disasters.

Refugees (including the evacuated) were not considered a group of losses. But during flight and shelter in the hinterland, they suffered more than others due to undersupply and health epidemics. During war, there permanently were about 500,000 Austrian refugees; Bosnia-Herzegovina had about 50,000 refugees in 1915.¹²

Table 4: Population of the Territory of Austria-Hungary 1910 and 1920/21, highlighting Front Areas¹³

Indirect Losses

Indirect losses are caused by war but not military actions, resulting mainly from famine, cold, and health epidemics. Thanks to its dual political structure, Austria-Hungary had three rather than one unified civilian statistics. The Hungarian Statistical Central Office published its data up to 1915 during the war and from 1916 to 1918 after the war, covering the old Hungarian territory with the exception of Croatia-Slavonia.¹⁴ The Austrian Statistical Central Commission (k.k. Statistische Zentralkommission) published during the war all available and allowed numbers.¹⁵ Depending on the topic, this data covered more or less the whole Austrian part of the monarchy and included the period from 1914 to 1917. Like some other successor states, Austrian Republic’s statistics tried to connect the old and the new republican statistics starting in 1919. But these “link-statistics” only reflected the area of the later Austrian Republic. Data on the Bosnian-Herzegovinan population ended in 1915.¹⁶

The total population of Austria, Hungary, and Bosnia-Herzegovina during war is completely unknown. Therefore, calculating a death rate is impossible. Data exists for the natural population movement (births and deaths) of Hungary (with 1914 borders, excluding Croatia), that of the area of the later Republic of Austria (without Burgenland) and Bohemia, Moravia, and Silesia. The Hungarian annual data is subdivided by month, that of Bohemia, Moravia, and Silesia by quarter and the area of the Austrian Republic by quarter only for the years 1917 and 1918.¹⁷

To get meaningful information about indirect civilian war losses from the total number of deaths, one has to exclude deaths of babies from the total. About 15 to 20 percent of all children died during their first year before the war. In peacetime, this was about 20 to 25 percent of all deaths. Nine months after the beginning of war, births were rapidly decreasing to half or less. The shrinking birth rate was reducing the total number of deaths as shown in table 5. Therefore, deceased babies distorted the death rate. Only the statistics of Hungary and the Austrian Republic reported the babies that died throughout the war, in both cases by month.

Table 5: Deaths in the Area of Austrian Republic and Hungary 1914-1918¹⁸

Food supply contributed to the number of indirect deaths in wartime Austria-Hungary. A self-supplier before war, Austria-Hungary had problems with food supply from the very beginning of war. However, the situation differed drastically in Austria and Hungary. Hungary was strong in agricultural production. During the war, the harvest of wheat and rye (for Hungary without Croatia) decreased from 5.4 million tons in 1913 to 3.4 million tons in 1918 (62.9 percent of 1913). The harvests were also uneven: those of 1914, 1916 and 1918 were worse, while those of 1915 and 1917 better than the previous year. Furthermore, the Hungarian net export of grain and flour to Austria continuously decreased from 2.1 million tons in 1913 to about 50,000 tons in 1917 (the figure increased in 1918). Statistically, Hungary had more grain per year of war and capita than before the war. With the exception of December 1914 to May 1915, when the wartime system of grain distribution was established, and the time of the Spanish flu in the autumn of 1918, the number of deaths remained at the peacetime level.

Austria faced a completely different situation. Austria lost her main grain producing area, Galicia, quickly and for the entire war. Furthermore, the statistically recorded Austrian harvest showed a shrinking of crop year by year, from 3.1 million tons of wheat and rye in 1913 to less than 1.4 million tons in 1917, 45 percent of 1913.¹⁹ In January 1918 the Office of National Nutrition (Amt für Volksernährung) had to reduce the insufficient daily rations of flour per capita from 200 grams to about 160 grams and even this could not be distributed completely. The amount of flour available went down to less than half. Instead of 104,000 tons needed in July 1918 only 43,000 tons could be distributed.²⁰

In contrast to Austria, the Hungarian crop of 1917 was at 81 percent of 1913. Neither a continuous decrease in the Austrian harvest nor the difference in reduction in Hungary seems to be plausible, as both areas suffered similarly under wartime conditions. In fact, the Austrian statistical data did not reflect the decreasing harvest itself but rather the increasing incompetence of the Austrian administration to record production. Hungarian statistics, which differed from the Austrian ones, produced much more reliable results, but they were compiled too late to control the harvest trade.²¹ Especially in 1916/17 and again in 1917/18 Austrian food production was shifting from the controlled and legal sphere into an illegal and uncontrolled one and was bartered on black markets.²²

Compared to the base figure of 100 in 1914, deaths increased to 117 inside the area of the later Austrian Republic in 1915, 119 in 1916, 132 in 1917 and 151 in 1918. This was one of the worse-supplied regions of Austria. The increase of deaths in 1918, however, was compiled only the last quarter when the Spanish flu had struck. During the previous quarters, deaths were stagnating for the first time during the war. The same result is shown by nearly all death-statistics of larger Austrian cities.²³

Indirect war losses were also caused by the cold, at its worst in February. In Hungary there was an extremely high death rate in the first quarter of 1916: 30 percent above the average of the following three quarters. In peacetime, the figure was only 10 percent above the average. The first quarter of 1917 also had a high death-rate, 20 percent above the average. In both cases, the following second quarter was also significantly above the average. Austria’s (the later Republic) quarterly data for 1914 to 1916 is missing. 1917’s rate was similar to Hungary’s, though the absolute number was higher in terms of percentage. The data for 1916 – published during war, including infant mortality – showed high death rates during the first two quarters of 1916.

In general, Austria-Hungary seems to have had enough heating material – mainly coal – for the population. However, in 1915 coal production reached its lowest point as the demands of the war industry were growing rapidly. A second crisis occurred in February/March 1917. Beginning with the Brussilov Offensive in June 1916, all main railway lines were closed for hinterland traffic for more than half a year. Romania’s declaration of war followed in August. Afterwards, the Central Powers began an offensive to conquer Romania that lasted until January 1917. In February/March, when a very hard winter hit Europe, Austro-Hungarian coal stocks were empty, though this seems to have been a question of distribution and not one of quantity.

With the exception of the very end of the war, medical progress prevented the spread of health epidemics. Most illnesses such as typhus, typhoid fever, smallpox, and cholera had their peak in 1915. Dysentery spread in 1917, decreasing rapidly in 1918. The most famous epidemic was the global pandemic called the Spanish flu in 1918. In Austria-Hungary nearly all monthly statistical data of larger administrative areas or cities indicates a slight increase in September, a massive increase in October and November, and a decrease in December, though still much above the normal level. From October to December 1918 in Hungary the average of quarterly deaths of the preceding nine months was exceeded by 100,000 persons (80,000 to 180,000). This number corresponds exactly with the increase of pneumonia (+47,000) and of flu (+53,000) from 1917 to 1918. Within the area of the later Republic of Austria in the last quarter 1918, some 26,000 persons died above the average of the preceding three quarters (30,000 to 56,000). There was an increase of pneumonia (+10,000) and of “other infections” (+15,000). During the last three months of 1918 in Austria-Hungary, the Spanish flu approximately doubled the deaths of a “normal” quarter.

In wartime Hungary (until September 1918), about 70,000 civilians died above the average of the years before the First World War, which was 22.6 percent of all civilian deaths in 1914. This figure includes deaths caused by famine and cold as well as all epidemics, but excluding the Spanish flu. In the area of the later Republic of Austria, the number was 90,000 civilians or 100 percent of all civilian deaths in 1914. If we suppose a similar death rate of babies in Bohemia, Moravia, and Silesia as in the area of the Austrian Republic, these lands had 80,000 civilian deaths above average, which was 60 percent of the deaths in 1914. During the last quarter of 1918 about 50,000 persons died above the average of the preceding quarters (+113 percent).

Conclusion

To produce a vague estimate for the indirect civilian losses of Austria-Hungary, we should assume a similar supply situation in Croatia as in Hungary and a similar situation in the remaining Austrian lands and Bosnia as in the area of the later Austrian Republic. Taking the deaths of 1914 minus 20 percent of births (estimated deaths of babies) we get an approximate minimum of 465,000 civilian indirect war deaths (351,000 from Austria, 82,000 from Hungary²⁴ and 32,000 from Bosnia). This is very close to Winkler’s result of 467,000 deaths, though his figure included the fallout from the Spanish flu.²⁵ If we also suppose a similar effect of Spanish flu as in the area of the Austrian Republic and Hungary (one-quarter of the deaths in the latter years of the war), this pandemic produced about 260,000 victims: 135,000 from Austria, 115,000 from Hungary, and 10,000 from Bosnia. Altogether the losses of Austria-Hungary can be estimated at 2 million (excluding direct civilian war deaths), in addition to deceased prisoners of war, in all with about 2.4 million or 46.1 per thousand of the population in 1910.

*Military Losses according to Winkler.

Table 6: Estimated total Austro-Hungarian War Deaths excluding prisoners of war (which cannot be divided by region) and direct civilian losses

Anatol Schmied-Kowarzik, Austrian Academy of Science