from the tab "Discussion" at the top of the article you linked to - - although I don't think that any of the links in the original article will work in this text version - - everything that follows the end of this line is verbatim unedited text of a single Copy-and-Paste:
Astonishing Death Toll
The number of people who perished in the purges is subject to hot disputes with death toll estimates ranging from 1 to '''100 million''' people, depending on who counts and what is counted as a purge.
This has got to be a joke. You couldn't convince a 10 year old that a median of 50 million would be executed out of a population of 160 million. Such is unfeasible. Demographic data does not remotely correspond to such allegations. Countless sources have correctly specified the death toll from executions to total 681,692 including the Russian Viktor Zemskov, J.Arch Getty, Robert Thurston and Stephen Wheatcroft. Archival data must be given priority over hearsay, rumours, and crude arithmetic.
Some sources place the number at about 20 million, which includes approximately 5 million kulaks and other peasants killed between 1929 and 1933; 5 million who died during the Ukrainian Holodomor, 5 million executed between 1933 and 1953 (including military personnel executions during the Great Patriotic War), and 5 million dead in gulag camps.
This has been debunked by archival material. It's been revealed that 150,000 kulaks died during 1930-1931; 3 million died during the rural famine of 1932-1933; 786,000 executions from 1930-1953; a 1 million deaths in the GULAG. The total figure is not quite 5 million. Again, we must give priority to archival sources over second-hand estimates. Sources include "Years of Hunger" by Stephen Wheatcroft and RW Davies and "Victims of the Soviet Penal System in the Pre-War Years: A First Approach on the Basis of Archival Evidence" by J.Arch Getty.
Then don't talk about it. Fix the article with your data, adding sources with every claim. Remember: be bold. -Kasreyn 22:02, 19 April 2006 (UTC)
"The main evidence for the gendercidal impact of the "Great Terror" lies in the Soviet census of 1959. In a fascinating addendum to the original edition of his work on the Purge period, The Great Terror, Robert Conquest uses the census figures to argue that the Soviet population "was some 20 million lower than Western observers had expected after making allowance for war losses." "But the main point," he notes, "arises from a consideration of the figures for males and females in the different age groups." He then unveils a striking table indicating that whereas age cohorts up to 25-29 displayed the usual 51-to-49 percent split of women to men, from 30-34 the gap widened to 55 to 45 percent. Thereafter, the disparity became massive, reflecting the generations of males caught up in the purges and the Great Patriotic War. From 35-39, women outnumbered men by 61 to 39 percent; from 40-54, the figure was 62 to 38 percent; in the 55-59 age group, 67 to 33 percent; from 60-69, 65 to 35 percent; and 70 or older, 68 to 32 percent. [...] [...] The estimates are "only approximations," Conquest notes, and "anything like complete accuracy on the casualty figures is probably unattainable." But "it now seems that further examination of the data will not go far from the estimates we now have except, perhaps, to show them to be understated"; and "in any case, the sheer magnitudes of the Stalin holocaust are now beyond doubt." He cites Joseph Berger's remark that the atrocities of Stalin's rule "left the Soviet Union in the condition of 'a country devastated by nuclear warfare.'" (All figures and quotes from Conquest, The Great Terror: A Reassessment, pp. 485-88.)" from Genocide Watch --Dwarf Kirlston 18:47, 6 November 2007 (UTC)
is there a Wikiproject Genocide?--Dwarf Kirlston 18:47, 7 November 2007 (UTC)
> "The main evidence for the gendercidal impact of the "Great Terror" lies in the Soviet census of 1959. In a fascinating addendum to the original edition of his work on the Purge period, The Great Terror, Robert Conquest uses the census figures to argue that the Soviet population "was some 20 million lower than Western observers had expected after making allowance for war losses."
This is pretty weak. For one thing, the opening of archives has revealed that the costs of war were worse than had been assumed previously. Once you begin making this type of alteration, everything else can be thrown out of kilter and it makes no sense to claim to know what the population of 1959 should have been then. This is why Conquest's tracts are pretty meaningless. The archives of the Gulag and the general demographical data have become available. This information can be found in Haynes & Husan, A CENTURY OF STATE MURDER? It does not support claims of "20 million." —Preceding unsigned comment added by 207.69.137.7 (talk) 04:08, 3 October 2008 (UTC)
There are plenty of other reasons why this might have happened. It is a post hoc ergo propter hoc argument to say that the lower population was a result of genocide. To assume that it's related to genocide is drawing a pretty long bow. David Glantz has proposed, with some justification, that the real figure for Soviet war dead was closer to 40 million rather than the figure of 20 million which is usually quoted.Flanker235 (talk) 11:47, 29 September 2011 (UTC)