Chapter 9

Table 9.1 presents the estimates, sources, and calculations of this Yugoslavian democide. My first concern in the table is to tally the World War II dead for Yugoslavia. Numerous estimates of this are available in the sources and I organize and consolidate them into military dead (lines 4 to 9 in Table 9.1), Partisan dead (lines 12 to 13), civilian dead (lines 23 to 26), civil war-dead (lines 30 to 31) and total war-dead (lines 34 to 48).

As to the military dead, there is usually no clear distinction between Yugoslavian soldiers that died fighting the Nazi invasion (an exception is line 4) and those Partisans, Chetniks, Ustashi (Croat) soldiers, and Slovenian Home Guards that died fighting each other. Given the twelve days for the Nazi victory and the reasonable estimate of 5,000 Yugoslav dead (line 4), the other estimates of military dead (lines 5 to 9) must include civil war battle-dead and those who died fighting occupation troops. The consolidated low of these estimates (line 10) is 300,000 dead. That this is indeed a lower bound is suggested by it equaling the consolidated low of just the Partisan dead (line 14). However, the mid-value and high for the military is 52,000 and 105,000 more dead, respectively, than that for only the Partisans and thus may be closer to the truth. Presumably, all these dead not only comprise the Yugoslav military, Partisans, and Chetniks that died fighting the German, Italian, Hungarian, Bulgarian, and Rumanian occupation forces, but also those dying in the fighting among Partisan, Chetnik, Croatian, and Slovenian Home Guard forces.

Estimates of civilian dead are scarce and problematical. Ideally, we would have estimates of those that died directly or indirectly from the battles fought by the various armies and guerrillas. We do have estimates of those who died in the civil war (lines 30 to 31), but there is no indication whether these were collateral casualties of battle, or also included those massacred. The higher of these (line 31) is an official figure for the Croatian-Partisan-Chetnik conflict. As will be seen below, this estimate is no different than many estimates of the overall military and civilian dead for the whole war period (e.g., lines 36 and 37).

Estimates of total war-dead (lines 34 to 48) apparently cover military and civilian battle-dead, and presumably democide as well, particularly that by the Nazis. No source that I could find gives a thorough breakdown of these estimates; the most consistent figures (near 1,700,000 killed) is in line with an official post-war total of the Tito government (line 44). Milovan Djilas, who suggests that the true figure may lie between 1,700,000 to 1,800,000, specifically mentions that this range includes the dead from camp, massacre, anti-communist fighting, and communist executions (line 48).

Consolidating all these estimates gives us a tentative range of 1,600,000 to 2,000,000 total war-dead (line 45). This is an preliminary figure until final democide estimates can be determined below.

Turning now to the democide estimates, the first of these (lines 51 to 52) are for the Partisans up to their transformation into a legal Yugoslav government in June, 1944. In the sources there are many indications of their atrocities, massacres, executions of anti-communists, and the killing of prisoners. But few figures, especially of any usable magnitude, are given. Taking all the diverse individual estimates into account, however, my own (line 53) is that from 50,000 to 150,000 people were murdered by the Partisans, probably 100,000. Some may be troubled that this estimate is the same as for the Chetniks (line 185). But this toll is only for the period up to July 1944, after which the Partisans were recognized in effect as a legal government. From July 1944 to the end of the war, they murdered an additional 500,000 people (line 280). However, the 100,000 democide estimate for the Chetniks is for the whole war.

For the period following the legal establishment of the Tito government (lines 57 to 164) there is a large number of democide estimates. I divide these into various categories and consolidate them. Unfortunately, these consolidated estimates usually overlap (e.g., lines 83 and 96) and this has to be taken account of in the final totals. Rather than sum all the consolidated values and then reduce them by some assumed proportion of overlap, they are accumulatively summed and adjusted as each range of consolidated values is added.

The first accumulated sum (line 72) simply adds those killed by the Partisans (line 71) after the "liberation" of Belgrade in October 1944 to the consolidated range of "anti-communists", opponents, and "collaborators" killed in 1945 and after (line 66), since the Belgrade democide predates these figures. Similarly, the estimate of the Cominformists killed form a unique category and is simply added to the other figures (line 76).

The estimates of those killed in the Bleiburg and related massacres (consolidated on line 83) are probably picked up by other categories, particularly those for "anti-communists" and "collaborators" (line 66) and Croats killed (line 96), and I therefore ignore them (line 84). The estimates of the number of Croats murdered (lines 87 to 95) probably contains independent figures, but part or most of the estimates likely overlap with those for "communists" opponents, and "collaborators." I assume this overlap to be 75, 50, and 25 percent for the low, mid, and high consolidated estimates (line 96), and accordingly I add these proportional amounts to the accumulated sum (line 97).

Forced labor and imprisonment for opponents or undesirables was a characteristic of the Tito regime, as it was of other East European, post-war communist governments. The table provides estimates (lines 101 to 111) of the prison and forced labor camp population, and consolidates these into their annual populations for the periods 1945 to 1955 (line 112) and 1956 to 1965. There are only two estimates of how many of these died or were killed (lines 115 and 116). It is clear from the sources that life in these camps was hellish: "starvation, overcrowding, brutality and death-conditions, which make Dachau and Buchenwald mild by comparison."¹ This was particularly true of the Goli Otok prison in which the conformists were held, as the letters from one inmate, Venko Markovski, well attests.² With this in mind I calculated an annual unnatural death toll of 1, 2.5, and 5 percent of the prison/camp/forced labor population for the years 1945 to 1955; .5, 1.25 and 2.5 for the less Stalinist, subsequent years 1956 to 1965 (these are very conservative excess death rates, compared to the literature's generally low estimate of 10 percent for Stalin's camps ³). Together these rates yield a range of 17,000 to 206,000 people who died in the camps or prisons from 1945 to 1965 (lines 117 and 118). These deaths may still overlap with those under previous categories, as for "anti-communists, opponents," and "collaborators." I prudently assume that the percentage overlap is 75, 50, and 25 percent for low, mid, and high values, respectively, and I add these proportional amounts of the forced labor, camp, and prison deaths to the accumulated sum (line 119).

After the war the Tito government tried to expel all ethnic Germans in the country, killing a number before and in the process. The result of calculating this toll elsewhere is given here (line 125). German POWs also were often killed before and after the war, I give some estimates of their number in the table (lines 131 to 135) and consolidate them (line 136): the consolidated low is calculated as the lower estimate of those who surrendered (low 130) times the ratio of the consolidated high to the higher estimate (line 129). No doubt there is some overlap with the consolidated total on line 66 (e.g., line 60), which I assume to be 20, 10, and 5 percent for the low, mid-value, and high, respectively. The result of subtracting this overlap is added to the accumulated sum (line 137).

Italian POWs and civilians were often killed as Tito's forces crossed the border into and occupied part of Italy before and immediately after the war. Based on one estimate (line 140), I assume that the number killed (independent of the occupation of Trieste and vicinity--lines 157 to 159) was 5,000 and add this to the accumulative sum (line 142).

I also add to this (line 150) the estimated range for Moslems and Albanians killed (line 149), and for Chetniks (lines 153 and 154). Although the sources frequently mention the killing out of hand of all captured Chetniks during the civil war and the execution of those caught after the war, I could not find any overall totals. I estimate that independent of other totals (such as on line 66) at least 10,000 Chetniks were killed. To this sum (lines 161 and 166) I add the consolidated ranges for those Italians and others killed during the brief occupation of Trieste and vicinity (160) and also add some miscellaneous estimates (line 165).

Now we can look at those estimates of the total killed by the Tito regime that I list in the table (lines 169 to 171). Under these is given the accumulative sum of all the previous estimates (line 172). Two of the estimates (lines 170 and 171) cover only part of the killing, and the other one (line 169) gives no basis. In any case, all are less than even the accumulative estimate's low of 585,000 dead. Given the diverse estimates involved in the accumulative sum, however, I will keep it as a preliminary democide for the Tito regime until I can check the overall total of all democide against some demographic calculations.

Based on the estimates of "anti-communists," opponents, and "collaborators" killed (lines 57 to 65a) mostly during the war, and those killed in Belgrade (lines 69 and 70), I estimate the Tito government's democide as 585,000 to 2,130,000 (line 174).

Next to consider is the democide by the Chetniks. They massacred Croatian and Moslem civilians, showed no mercy to any Ustashi they caught, and killed Partisan sympathizers and prisoners, but no overall figures are available. I list some estimates of specific massacres in the table (lines 177 to 184), and based on the sources I also provide my estimate of what seems a reasonable range of Chetnik democide (line 185). Now, many Chetnik units revengefully sought out and killed whatever Ustashi and, in some cases, simple Croats in reach (for the Ustashi slaughter estimated below). Nonetheless, I am troubled by the estimate of Omrcanin (line 181a) that 500,000 Croatians were killed. He neither justifies nor explains the estimate, and nowhere can I find other figures that are even close. Yet, given the Serbian hatred of and vengefulness toward the Croatians, the estimate is possible, even if improbable. I therefore treat it as a high, while making the mid-value only 20 percent of this number.

No count of democide during this period can ignore the massacres by the German puppet State of Croatia. Estimates of these are divided in the table and consolidated for the Jews (lines 189 to 193), Gypsies (lines 196 to 201), and Serbs, for which I separate the partial estimates (lines 205 to 219) from the overall estimates (lines 222 to 237), although as can be seen there is not much difference in their magnitudes. Some estimates of the number killed in Croatian concentration camps are also listed (lines 240 to 242), but these dead are surely included in the estimates for the Jews, Gypsies, and Serb deaths, and therefore not consolidated. I conclude that the total democide by the Croatians is 242,000 to 1,088,000 people (line 245).

Finally is the democide by the various occupation armies. I give elsewhere calculations for the Nazis democide and list the total here (line 248). Based on the sources, particularly the work by Fotitch,⁴ I give an overall minimal estimate for the Bulgarians of 10,000 Yugoslavs killed. A variety of estimates are available and consolidated for the Hungarian massacres of Serbs (lines 255 to 261) and Jews (lines 266 to 269), and totaled (line 272). Finally, there is the democide by the Italians, which I estimate as a minimal 5,000 (line 275).

Now I can calculate the overall democide in Yugoslavia. In the table I pull together the democide totals for the Chetniks, Partisans, and so on (lines 278 to 286). I divide that for the Tito government into democide during and after the war (lines 280 and 281). The total democide of all parties in Yugoslavia turns out to be 1,515,000 to 4,805,000 people, of which 1,230,000 to 3,425,000 of them were killed during the war (lines 287 and 288).

To assess these totals, I also show the battle-dead totals (line 289) from the beginning of the table and then I add this range to that of democide during the war (line 290). This sum can now be compared to the overall war dead total (line 291). As can be seen, while the low of the battle plus democide total is within the range of overall war-dead, the mid and high values are much larger. This does not automatically imply that the democide figures are too high, for there is no way of knowing whether the estimates upon which the war-dead are based did in fact take into account the variety of killing included here, especially in considering that the central figure of 1,700,000 killed is nearly the same as that given by the Tito government itself (line 44).

To further evaluate the battle-dead plus democide total, I calculated population deficits for the years 1941 to 1950 (lines 292 and 327) and 1941 to 1965 (lines 293a and 328) from population estimates and calculated rates of increase, to be discussed further below. The deficit is more consistent with the total battle-dead and war-time democide, but it is for 1950 and much killing did take place in the post-war period. Therefore, the total war and post-war democide and battle-dead are added together (line 293) for comparison to a population deficit for the years 1941 to 1965, by which year such killing had virtually ended. As can be seen, the range is well below that for the deficit and consistent with it, even considering that the deficit includes the unborn.

Keeping the variety of estimates in mind and the above checks, it seems a prudent conclusion to accept the overall democide total.

Two tasks remain. One is to show the calculations for the two population deficits which are used above, and the other is to calculate the rate of democide. To this end, a number of population estimates and rates of increase are given (lines 296 to 321). From these a range for the 1941 population of Croatia (line 322) is determined.

Of special interest is the percent rate of increase in the Yugoslavian population. That from 1929 to 1931 and 1931 to 1941 was calculated as 1.3 percent per year (lines 305 and 310). The formula used for all such calculations here is

growth rate = (later population/earlier population)years - 1.

For comparison are shown rates of increase for certain post-war periods (lines 314, 316, and 320), which over the years 1962 to 1966 settled down to about 1.1 percent. From all these rates a range of population increases for 1941 to 1950 (line 323) and to 1965 (line 324) are determined. These now enable Yugoslavia's population to be projected to 1950 and 1965 based on the 1941 population, and by subtracting these projections from the actual population estimates (lines 311 and 317) to determine the population deficits used previously (lines 327 and 328). This deficit reflects not only the number of unnatural deaths but also those consequently unborn.

Three sets of democide rates are calculated. The first is for the State of Croatia (lines 332 to 333). The second is for the Tito government during the war and up to 1965, by which democide had largely ended. The third is for the near full life of the Tito government up to 1987 (the cut-off year for this comparative study). No rates could be determined for the wartime Partisan or Tito government, I could not find estimates of the population under their control.



* From the pre-publisher edited manuscript of Chapter 9 in R.J. Rummel, Statistics of Democide, 1997. For full reference to Statistics of Democide, the list of its contents, figures, and tables, and the text of its preface, click book.

1. Quote from British representative Frank Waddams (Beloff, 1985, p. 134).

2. Markovski (1984).

3. Rummel (1990, Table 1B, line 113).

4. Fotitch (1948, pp. 137-139).

---

For citations see the Statistics of Democide REFERENCES

---

You are the visitor since 11/25/02

Go to top of document