I was wondering which year in history was the bloodiest? Is it one of the last two years of WWII? Or are there any other such bloody time in history?

I only wish to include non-natural causes deaths such as from war, disease, or even from accidents.

The number alone is not helpful, so I am looking for a ratio of total number of such deaths to the total human population of given year.

  • 5
    If you believe the bible is true, and want percentages, it had to be the day of the Flood. If you want pure numbers, it almost had to be 2015, since there were more living on the earth than ever before and therefore, more death.... Now that I've gotten my daily snark out of the way, you'll have to define your request a bit more... death by what? war? famine? natural causes?
    – CGCampbell
    Mar 25, 2016 at 13:06
  • 6
    This may be a more interesting question than it appears. There is a British historian who specializes in measuring violence through history. I believe he concluded that total violence decreases over time, which is counterintuitive. Alas, I don't recall his name. And the one year measurement is going to void the results - statistically there will be outliers and 2015 is the optimal bet - the number of deaths will be proportional to the number of people alive.
    – MCW
    Mar 25, 2016 at 13:34
  • 1
    Just a guess but I would say 1916. It's also quite possible that one of the years during the Holocaust or Mao's Great Leap forward. It could even be 1994 - the Rwandan civil War. I think something like 4 million people died. There are a few candidates, but I think due to the nature of these atrocities, the exact data isn't forthcoming. A
    – Anaryl
    Mar 25, 2016 at 13:48
  • 2
    @T.E.D oh yes Spanish flu - I forgot about that
    – Anaryl
    Mar 25, 2016 at 14:05
  • 17
    I find curious the definition of "disease" as "non-natural" (unless you count a -yet to happen- GM disease).
    – SJuan76
    Mar 25, 2016 at 14:09

3 Answers 3


The actual year would be in prehistory, when the human population were more concentrated.

In terms of recorded history, the initial outbreak of the Plague of Justinian in 541 had an estimated death toll of 25 million. Most estimates of world population gives about ~200 million for the 6th century. The plague thus killed roughly 10-13% of the global population.

For reference, the 1918 flu pandemic killed perhaps 50-100 million. The global population at the time was about 1,860 million, however, so the ratio works out to be 2.7-5.4% of the total population.

A.D. 541 also have additional "advantages" in terms of higher infant, maternal, and child mortality were far higher in some parts of the world, relative to the industrialised nations of 1918.

  • 2
    the Plague of Justinian in 541 had an estimated death toll of 25 million -- what is the source of this claim that 25M died in the single year of 541 from the plague? Right now the linked Wikipedia page says "deaths of an estimated 15–100 million people during two centuries of recurrence", but I'm unable to find any source that claims that 25M died in the single year of 541.
    – user54367
    Apr 1, 2023 at 4:50

70,000 BC

Analysis of the human genome has suggested that there was a choke-point in human history where the number of humans was drastically reduced.

Some believe that this was due to a single volcanic eruption occurring 70,000 years ago give or take.

According to this theory human population may have been reduced to just a few thousand breeding pairs.

Not sure if this meets your definition of 'non-natural' or of 'history' but no other event will have made the same ratio.

  • 10
    History and prehistory; that said, this such a good answer that I can't bring myself to cavil about a difference of a few tens of thousands of years.
    – MCW
    Mar 25, 2016 at 18:56
  • I remember hearing about this. But wasn't that just for one line? IIRC, you get very different findings when you check the Y Chromosome (inherited only from father to son) and Mitochondrial DNA (inherited only from the maternal line).
    – T.E.D.
    Mar 26, 2016 at 15:45
  • 4
    I'm pretty sure that volcanoes going off is a natural thing. On the other hand, the question classes disease as non-natural so who knows? Mar 27, 2016 at 23:10
  • Recent criticism about this theory: bbc.com/news/science-environment-22355515
    – Evargalo
    Feb 28, 2019 at 13:10

A bit of quick research tells me the worst in absolute terms almost has to be 1918.

There was a worldwide influenza epidemic that killed about 40 million people that year. That's the single biggest recorded death toll for a global pandemic in a single year in human history1.

There also happened to be one of the bloodiest wars in human history (in terms of sheer number of combat deaths) going on at the time. For most years, roughly 3 million combat deaths would be a huge deal. That year it wasn't even 10% of what the flu did.

Barring some new fast-moving pandemic, or a disaster of extinction-event proportions, that is liable to be the worst the human race will see for quite some time2. The number of deaths due to war is actually decreasing globally.

1 - The Black Death probably killed more people; perhaps as many as 200 million in Eurasia. But it took 6 years to work its way around Europe.
2 - Everyone reading this please go find some wood to knock on.

  • 2
    As for some putative European contact year, the problem with that is even by the most ridiculously generous estimates there were no more than 100 million pre-Columbian Americans. For it to beat 1918, those would have to be correct, and most of the deaths would have had to happen in a single year. Communications and travel weren't that fast on that continent at that time. For example the medieval black death took 6 whole years to work its way through all of Europe.
    – T.E.D.
    Mar 25, 2016 at 14:11
  • 2
    If the Black Death killed 200 million people in 6 years, you have an average of 33 million/year if you assume equal distribution. So it's at least conceivable there's a year close to the end that had more than 40 million. And of course, the percentage of total population was much higher at that time. Mar 26, 2016 at 5:46
  • 2
    @GuntramBlohm - Yes, but you're taking the highest estimate (which is quite likely to be an overestimate), and then giving it a boost on one year, for a supportable reason, but mostly because it helps arrive at a number you'd like it to. One should avoid doing that with multiple pieces of information.This is the exact kind of logic that Columbus used to argue Japan was 3,000 miles west of the Canaries, when it was more like 20,000. Far better to use averages so wrong estimates even out.
    – T.E.D.
    Mar 26, 2016 at 15:32
  • 3
    If 200m people died of Black Death over 6 years then there must be at least one year when at least 33m people died of Black Death. Assuming equal distributions would lessen the worst year. Taking a lower estimate of 75m for the Black Death over 6 years with population at 450m shows its worst year killed at least 2.7% of the world population, c.f. 2.2% for the flu using 40m / 1.8b. So the Black Death answers the question in terms of ratio. And whether it breaches 40m in a single year is probably known as well.
    – djechlin
    Mar 26, 2016 at 23:53
  • 1
    The book referred to above regarding the historical trend of violence is "The Better Angels of our Nature" by Steven Pinker. A weighty tome but very interesting appraisal Apr 1, 2016 at 3:31

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.