0

Accusing a regime or other political organization of mass killing can be a fairly effective way of riling people up. But a lot of the relevant statistics seem lacking in direct support; we don't often have the combination of official records-keeping and demographic analysis that explains where e.g. our understanding of the scale of Nazi democide in the vernichtungslager comes from. So, for example, anti-Soviet polemics sometimes attributed an incredibly large number of deaths to the gulag network, with numbers running into the tens of millions (R. J. Rummel has the most meticulous, if deranged, samples of these kinds of estimates).

Still, the Soviet government was known for both internal and external duplicity, so I wonder if the newer figure of approximately 1,500,000, which is said to be grounded in NKVD archives, is so much more reasonable.

More pointedly, is there a sort of moral dilemma of evidentialism in this kind of context? Where we confront the following horns:

  1. A commitment to precision/accuracy, proportioned to available evidence. This commitment favors lower numbers insofar as fewer specific deaths can be reliably confirmed.
  2. A commitment to honoring the memory of the victims. Apologists for the US war in Vietnam, for example, e.g. Guenter Lewy, seem to disrespect both the physical facts as well as the probability that the number of people killed modulo those facts was much higher than can be determined by reviewing hospital records or body counts. (When many of the victims were incinerated or vaporized, and then not while anyone was looking very closely, it seems offensive to trust that the statistics provided by the aggressors are anywhere near the reality.)

Is this really a moral dilemma, or is this problem indicative more that democide statistics aren't actually as important as various polemicists make them out to be? If saving one life is like saving the whole world, then taking even one life unjustly might be thought akin to murdering an entire world, in which case we would judge any pair of regimes equally malignant if both were guilty of even a single murder, rather than try to rank regimes in a list of increasingly large democide tolls.

Kristian Berry
  • 9,561
  • 1
  • 12
  • 33
  • 1
    If you replace 'moral' by 'religious' it should be clear enough: Hitler is the secularized form of Satan. But the replacement fails to provide a working moral compass: If Stalin killed twice as many as Hitler is Stalin = 2xSatan? Whether in theology morality or math, finite = infinite is a bigger source of error than 1=0. – Rushi Jun 28 '23 at 09:49
  • @Rusi with respect to the Nazis, we should also keep in mind the rate of killing once they got going, and their apparent plans. Stalin had decades to do what Hitler did in a few short years, so we can project that the Nazi slaughter would have far and away eclipsed the Soviet one had it not ended when it did. Or then we can also consider that the Chinese nationalist regime at that time massacred more people by a single concrete action than any other regime is known to have (blowing the Yellow River dam in 1938). I don't know about comparing Hitler or Chiang Kai-shek to Satan, though. – Kristian Berry Jun 28 '23 at 15:56

2 Answers2

1

"Is this really a moral dilemma, or is this problem indicative more that democide statistics aren't actually as important as various polemicists make them out to be?"

Why are those the only choices?

Consider the Armenian Genocide, or the Holodomor. Death tolls and the reasons for the deaths are vigorously disputed. What then are you suggesting is the response? Ignore the significance of these events?

I personally found it fascinating that the death toll of the Three Kingdoms War in China early in the first millennium kill more people than any war until at least WW1 and based on average estimates not until WW2 (the Chinese numbers are based on census data and excess deaths, and are pretty solid as I understand it). The worlds probably second most deadly war, and do Europeans even learn that it happened? I don't think Chinese culture and politics can be understood without reckoning with the scale of that war.

A partisan attempt to tell a story about how politics should be done based on regime death tolls, is obviously about bad-faith rhetoric, generally with no real concern for the humans involved, only a pursuit of the means to berate partisan opponents. So sure, skip that.

But is pursuit of better estimates of death tolls pointless or counterproductive? There are still Holocaust deniers claiming it was just a typhus outbreak, despite all the zyklon-b and books bound with human skin. The weight of evidence forced a reckoning in Germany, that led to profound cultural and political change that hadn't happened after WW1. Japan didn't truly have such a reckoning, their foreign policy still involves denials and evasions, and has proven still susceptible to nationalist militaristic rhetoric. Russia didn't reckon with their causing the Holodomo, and governments acknowledgement of it as genocide is still split around the world by how friendly a nation is towards Russia, and arguably that contributed to the current war. Better statistics, better evidence, can bring lasting political changes, that's why gathering information about Bucha and other Russian massacres is so important for when the reckoning comes, and hopefully the trials for crimes against humanity.

Death tolls are part of the story of humanity. They are subject to misuse, abuse, and lies. That does not make them unimportant, it only reveals the importance of who shapes the narrative about the past.

See this answer for how good history comes from constructive contention over how we narrate the past: Do historians have responsibility in how they decide to depict something?

"I wonder if the newer figure of approximately 1,500,000, which is said to be grounded in NKVD archives, is so much more reasonable"

So go look at the methodology. Good historians tend to be cautious, but also try to be fair, and to be pursuasive. People have really mined Soviet information to try to get the best estimate. Personally I respect that attempt to assess what happened as best we can. Think of the challenge we will have with North Korea, when that falls, and none of the people left alive will have an interest in the world knowing how they stayed that way.

CriglCragl
  • 19,444
  • 4
  • 23
  • 65
  • I used to study democide statistics in excruciating detail, cross-reference the endless citations that Rummel, Matthew White, Chomsky, and various others provide. And so yes, where orders of magnitude can be supported, I think we can be sensitive to historical accuracy; but I take issue with e.g. Rummel's artificial precision. And the further back in time a campaign of atrocities is, I suspect that we simply have lost out on a chance at much of a solid analysis. – Kristian Berry Jun 28 '23 at 15:40
  • My own efforts have been focused on the death toll wrought by US forces during the Vietnam War. On the level of sheer number-juggling, I consolidated an estimate that seemed a probable order-of-magnitude of 1M to 2M in South Vietnam and ~1M in Cambodia, Laos, and North Vietnam. But I still have trouble showing how the 1.5M in SV was caused (I need a huge subset in 1967, and there seems like a possible mechanism for it in the details of that year, but then in the back of my head I wonder if my interpretation is well-grounded enough). – Kristian Berry Jun 28 '23 at 15:43
  • 1
    So I guess a third/fourth option might be: "Instead of specific numbers, should we focus on ranges of orders-of-magnitude?" – Kristian Berry Jun 28 '23 at 15:45
  • Incidentally, was the Three Kingdoms War deadlier than the Taiping Rebellion? But I think demographic shortfalls in Chinese antiquity have to be approached delicately when used as estimates of deaths in some period (and I think some Western, or non-Chinese, analysts latch on to the demographics to shore up a picture of Chinese history as especially grisly, when China should not be regarded as a singular nation like that, over all the involved centuries, in the first place). – Kristian Berry Jun 29 '23 at 15:42
  • @KristianBerry: Taiping 20-30m over a shorter period, 3KW 36-40m. China conducted very thorough censuses, & had an exceptionally able beuracracy, for the era. I don't get why you think scholars don't just want to get the most accurate answer they can, taking account of the evidence. Pick your scholars & scholarship, because that is the verifiable track record of concern of the good ones. Chinese history is grisly, famines & floods have always had huge impacts, as they've stayed at higher population densities. It's about telling the story, & since Qin the conviction was China is one place. – CriglCragl Jun 29 '23 at 16:51
  • I have spoken with Chinese citizens who think the unity of China has been exaggerated when we look back further in time, as if we treated the Holy Roman Empire as a more consistent/stable "object" than it was (or Europe more generally, for that matter). And very few analysts (or none that I know of, actually) think census deficits are one-to-one with death tolls. Moreover, as a matter of epistemic principle, I really do think that the longer-ago something was, the less reason we have to strongly believe specifics about those times. We can know that an asteroid hit the Earth ~65M years ago... – Kristian Berry Jun 29 '23 at 17:30
  • ... but if people claimed to have good grounds for which day that happened, I would raise an eyebrow, to say the least. Re: Taiping, I've seen figures of 50M and even 100M. Some estimates for the Khanate invasion go from 35M to 60M. So it's hard to be overly confident in such numbers. – Kristian Berry Jun 29 '23 at 17:31
  • And if you want stronger reasons to doubt the accuracy of official statistics there, [accusations that Chang Hsien-Chung killed 600M Chinese people in the 1600s](https://en.wikipedia.org/wiki/Zhang_Xianzhong) are a good start. – Kristian Berry Jun 29 '23 at 17:42
  • @KristianBerry: But, Ancient Roman records are dramatically better than those in the whole millenium after the fall of the Western Roman Empire. Census data is not the same as a victor-written propaganda claim. Censuses generally related to how much tax you could expect, & had strong motivation to be accurate. Bottom line, is I think scholars who specialise in the subject, are better place to draw conclusions than us amateurs. Where there is dispute, I look to criteria of commitment to impartiality. – CriglCragl Jun 29 '23 at 18:16
  • I would not claim to be an expert on this topic *per se* but I would like to point out that I've been studying this topic for about twenty years, and as far as China goes, I've even read obscure books like Ping-ti Ho's *Studies on the Population of China*. And I'm not familiar with the level of consensus you are indicating anyway (expert interpretations of the An Lushan death toll have fluctuated dramatically over the last 10 years or so, for example). – Kristian Berry Jun 29 '23 at 18:28
0

I think it is meaningless to try to rank examples of evil in such a way, and arguments about, for example, exactly how many millions of people Stalin had murdered, are perverse. Democide is deplorable regardless of the numbers involved. There is certainly no moral dilemma of the sort you imagine. Why would you feel a compulsion to minimise the apparent scale of atrocities by scrupulously acknowledging only those deaths for which you have incontrovertible evidence?

Marco Ocram
  • 8,686
  • 1
  • 8
  • 28
  • 1
    One reason for being cautious about larger numbers is that promoting larger numbers can often undermine the outward plausibility of the moral argument being made. It is silly to say that the gulag cost 30,000,000 people their lives, and when anti-communist polemicists say such things, they undermine their critique. Oddly, a Christian writer recently tried justifying claims that Christianity has resulted in an extraordinary number of deaths; as far as his critique was aimed at the RCC, though, he too then undermines the seriousness of his argument. – Kristian Berry Jun 28 '23 at 15:48