Where the 80 Percent Comes From: The Arithmetic Behind the Herero Death Toll

Eighty percent. The figure appears in museum texts, encyclopaedia entries, parliamentary speeches and the German government’s own 2021 statement recognising the genocide against the Herero and Nama. It is stated the way a measurement is stated. It is not a measurement. It is a subtraction, and both numbers in it deserve a closer look than they usually get.

One thing first, so that nothing below is misread. Between 1904 and 1908 the German colonial army in South West Africa, today’s Namibia, drove the Herero into the Omaheke sandveld after the battle at the Waterberg, sealed the desert edge against their return, and put the survivors into camps where, by the army’s own count, close to half died. General von Trotha’s proclamation of 2 October 1904 ordered every Herero inside the German border shot. That is a genocide by any definition in use, and Werner Hillebrecht, the former head of the National Archives of Namibia, made the point plainly in 2007: no percentage makes it one and no recount unmakes it. The question here is narrower. Where does the number come from, and how good is it?

The top of the fraction: 80,000

The pre-war figure goes back to the Blue Book, the report the South African administration published in 1918 after taking the colony from Germany. Its authors proposed a minimum of 80,000 Herero before the war, drawn from the estimates of three men who had known the territory: the Cape official Palgrave, the German governor Leutwein, and the officer Schwabe. They then argued that natural increase should push the real figure to 100,000.

Eighty thousand was the round number in circulation before the war, not a count. The geographer Karl Dove, writing in 1903, called it the generally used estimate. The missionary Irle repeated it in 1906. The German governor himself was more cautious. In February 1904, weeks into the uprising, Theodor Leutwein wrote to Berlin arguing against the hardliners who wanted the Herero destroyed, and put it this way: „a people of 60,000 or 70,000 is not so easy to annihilate.“ I quote the sentence because it is the one pre-war figure that comes from the man responsible for the colony, written when he had no reason to inflate it. Elsewhere he is recorded at 70,000 to 80,000. A missionary named Bernsmann, in 1906, arrived at 35,000, an outlier nobody else matched.

Horst Drechsler, whose 1966 study from the East Berlin archives established the genocide thesis in German scholarship, took the Blue Book’s 80,000 and then, in a footnote, subtracted some 10,000 Herero who had died in the typhoid epidemic of 1898. His working figure was therefore 70,000. Helmut Bley used 60,000 to 80,000. Hillebrecht, surveying all of it, settles on somewhere between 60,000 and 80,000 as the honest range.

None of these estimates rested on a census. They rested on what missionaries and officials knew of the chieftainships, which, because Herero pastoralism moved with the cattle, was reasonably complete but never exact. Sixty thousand and eighty thousand are both defensible. They are a third apart.

The bottom of the fraction: 15,130

The post-war figure is more specific and, for that reason, more misleading. Drechsler used 15,130, which the Blue Book presented on its page 35 as the result of the official German census of 1911. Drechsler noted that he could find no 1911 census figures in the Reichskolonialamt files and had to rely on the Blue Book for the number.

Nobody, Hillebrecht observes, seems to have checked the census reports themselves, which sit in the National Archives of Namibia. He did. The German administration’s own population statistics record 19,423 Herero on 1 January 1911, within a total of 70,537 „natives“ in the territory. That is roughly 4,300 more than the Blue Book’s figure, and the Blue Book, published by a wartime enemy of Germany, does not explain where its lower number came from.

To the survivors inside the colony must be added those outside it: a few thousand Herero who reached British Bechuanaland with Samuel Maharero, some who reached the Ovambo kingdoms, and small groups Maria Fisch later traced along other escape routes. Three thousand is the figure usually given for Bechuanaland. It is, again, an estimate.

What the division actually yields

Put the numbers side by side and the percentage is no longer one number.

Eighty thousand before and 15,130 after gives a loss of 81 percent. That is the textbook figure. Seventy thousand before, Drechsler’s own working number, and 19,423 counted plus 3,000 in exile gives 68 percent. Sixty thousand before and the same survivors gives 63 percent. Hillebrecht’s own framing of the established view is 70 to 80 percent, which already carries the uncertainty the round figure hides.

Every one of those results describes the destruction of most of a people in four years. Two-thirds is not a smaller crime than four-fifths. But two-thirds and four-fifths are different claims, and only one of them is printed.

Why the numbers were fought over, and why that fight matters here

The reason nobody wanted to open this arithmetic is that someone already had, in bad faith. In 1989 the Namibian archivist Brigitte Lau published an essay arguing that the pre-war numbers were inflated, the post-war numbers too low, and that the word genocide should therefore be dropped. Gert Sudholt, whose 1975 book supplied her framework, went on to a career in far-right publishing. Walter Nuhn proposed 40,000 Herero before the war so that „only“ 24,000 could have died. Lau’s critics, Tilman Dedering in 1993 and Hillebrecht in 2007, pointed out that this is the shape of Holocaust denial: shrink the before, inflate the after, and pretend the verdict follows from the remainder.

That history explains why the 80 percent has been left alone, and why Hillebrecht, when he did open the census file, was careful to say that the higher survivor count changes nothing about the verdict. It doesn’t. The extermination order exists. The camp mortality, 7,682 dead out of roughly 17,000 Herero and Nama prisoners between October 1904 and March 1907, or 45 percent, is in a German army report, not a British one. The expropriation of all Herero land and cattle is in two German ordinances of 1906 and 1907.

But leaving the arithmetic alone has a cost. A figure that cannot be examined without suspicion of denial stops being a finding and becomes a password. And a password can be taken from you by anyone who does the subtraction in public.

What doesn’t add up

An estimate divided by a count is presented as a count. The 80,000 is a pre-war guess with a spread of at least 20,000; the 15,130 is a census figure that turns out not to match the census. The resulting 80 percent is then cited to one significant figure as if someone had measured it. The claim reaches further than either input allows. What would settle the top of the fraction: nothing now, since no pre-war enumeration was ever made. What would settle the bottom: the archive already has. Hillebrecht’s 19,423 has been in print since 2007, with a file reference, and has not displaced the Blue Book’s number in a single major reference work I could find.

A chain of citation with a published correction nobody picked up. Drechsler cited the Blue Book because he could not find the census in Berlin. Everyone since has cited Drechsler, or the Blue Book through him, or the encyclopaedias that cite both. The 1911 report in Windhoek was there the whole time. It is not that the correction is hidden; Hillebrecht’s article is on the open web. It is that the 15,130 had already become the number, and a number that has become the number stops being checked. What would move it: a single standard reference, the German government’s own 2021 text for instance, adopting the range instead of the point figure, with the archival source named.

What stays open

Whether the pre-war Herero numbered 60,000 or 80,000 will not be known. The archive that could have answered is the one the Germans never made. What can be said is that the loss lay somewhere between two-thirds and four-fifths of a people, that the German army’s own records document the mechanism, and that the one figure in the sum that was ever counted has been miscopied in the standard literature for a century.

Some of the descendants say the true number was higher than any of these. They may be right; the estimates leave room for it. The point is not to lower the toll. It is to say what is known and what is not, because a genocide documented by its perpetrators‘ own files should not have to lean on a figure its defenders cannot show their work for.

KardiKassin

Writes about Africa’s colonial history and how it gets told, sourced and checked at every step.

More from KardiKassin

Made in Germany: How Your Picture of Africa Was Manufactured

A wrong figure got copied for a century because nobody reopened the file. The free 12-page ebook traces that same kind of drift across a whole continent’s worth of image-making.

Get the free ebook

Sources

• Hillebrecht, Werner: „Certain uncertainties“ or Venturing progressively into colonial apologetics? Journal of Namibian Studies 1 (2007), pp. 73 to 95. https://www.academia.edu/64685679/ (contains the 1911 census figure of 19,423, NAN ZBU 172 A.VI.d.4, and the survey of pre-war estimates)
• Union of South Africa: Report on the Natives of South-West Africa and their Treatment by Germany (the „Blue Book“), London 1918; annotated reprint: Jeremy Silvester and Jan-Bart Gewald, Words Cannot Be Found. Leiden: Brill, 2003
• Drechsler, Horst: Südwestafrika unter deutscher Kolonialherrschaft. Berlin: Akademie-Verlag, 1966 (English: Let Us Die Fighting, London: Zed Press, 1980), p. 252 and fn. 329 on the population figures, p. 251 on camp mortality
• Gewald, Jan-Bart: Herero Heroes. A Socio-Political History of the Herero of Namibia, 1890 to 1923. Oxford: James Currey, 1999