Hacker News Reader

Comment by pyth0 | Hacker News Reader

Please tell me by what metric and to what other conflicts are you comparing this to where over 35,000 people (over 15,000 of which are children) killed constitutes the "lowest civilian casualty rate in modern history". The smugness with which you treat the tragedy that is currently unfolding is disgusting.

View on HN

Not parent, but let's take each side's numbers at face value:

The Gaza Ministry of Health says as of today that 35,562 people have been killed [0]. The Israeli Ministry of Defense in March said it has killed 13,000 Hamas operatives [1].

Leaving aside the two month gap between these figures, the civilian casualty ratio is 1:1.7.

I tried to find a source for what a "typical" casualty ratio is in urban conflicts. This source [2] claims that 90% of overall casualties is a typical number. That would be a ratio of 1:9.

John Spencer, who chairs the Modern Warfare Institute at USMA, and seems to be an authority on the subject, has a tweet addressing this specifically [3], in which he cites the Battles of Mosul, and Manila as having casualty rates of 1:2.5, 1:6 respectively.

I don't think proving the negative of "lowest civilian casualty rate in modern history" is feasible, but a nearly 5x improvement in civilian casualties compared to the assumed norm, and lower civilian casualties than Spencer's comparisons seems to indicate that the claim is not without merit.

[0] https://www.aljazeera.com/news/longform/2023/10/9/israel-ham... [1] https://www.timesofisrael.com/liveblog_entry/hamas-says-gaza... [2] https://civiliansinconflict.org/our-work/conflict-trends/urb... [3] https://x.com/SpencerGuard/status/1786612914117349769

It’s hard to believe those numbers when (according to anonymous Israeli military officers) the Israelis are willing to routinely accept civilian casualties of 20 to 1. https://www.972mag.com/lavender-ai-israeli-army-gaza/

Neither the Gaza Ministry of Health, nor the armed wing of Hamas have released the number of combatant casualties, leaving the IDF's number as the only estimate available. If that estimate is incorrect, they could and should challenge it by putting forth their own.

The source you provided says "for every junior Hamas operative that Lavender marked, it was permissible to kill up to 15 or 20 civilians". This is clearly an upper bound, and makes no claim about how close to this they actually come.

In the same way, I might offer my customers an SLA of two nines in their contract, but never drop below three in practice. Part of effective planning is describing the worst case scenario, but that doesn't imply it will actually happen.

The battle of Manila stands out as one of the most horrifying in modern history, hardly a good standard to be targeting.

Modern History is already quite good compared to the entirety of history. The bog standard Siege of La Rochelle ended up starving the civilian population of 27,000 to 5,000. War is brutal, and when you dealing with brutal enemies like Hamas it's never going to an orderly affair.

The IDF counts every grown man that they kill as a "Hamas operative"

Hamas counts every Hamas operative younger than 18 who dies as a child.

This is an outrageous and horrifying accusation, can you share your source for it?

> 35,000 people (over 15,000 of which are children)

For what is worth, the UN estimates are significantly lower with less than 8’000 children (and 5’000 women) out of the 25’000 identified casualties. Maybe there are indeed 10’000 additional victims as Hamas claims (the UN take that number at face value, Israel estimated are slightly lower) but it seems unlikely that 75% of them are children. It’s not physically imposible though.

The number of dead will be higher than reported, as there are many still under the rubble of thousands of homes.

And there will still be many more starving to death each day, thanks for aid trucks being attacked and burned (with IDF collusion apparently).

https://www.theguardian.com/world/article/2024/may/21/israel...

Reply on news.ycombinator.com