This claim has been floating around for awhile, apparently, but a specific example is given in a 2003 NY Times article, excerpted from the book "What Every Person Should Know about War" by Chris Hedges:

Has the world ever been at peace?

Of the past 3,400 years, humans have been entirely at peace for 268 of them, or just 8 percent of recorded history.

Note that the article gives a specific definition for war:

War is defined as an active conflict that has claimed more than 1,000 lives.

Taking that definition, have there really been 268 years in the last 3400 where no active conflict that ended up taking more than 1000 lives was going on? What evidence is this based on?

I found this article criticizing the claim and citing a probable source, but it doesn't really address the evidence.

There was a similar question about the number of days of peace since WWII, but obviously the kind of evidence available would be very different for ancient history.

  • related: skeptics.stackexchange.com/questions/27555/…
    – Sklivvz
    Commented Jan 29, 2016 at 11:50
  • @Sklivvz, thanks, I meant to mention that in my question but forgot.
    – user11522
    Commented Jan 29, 2016 at 11:52
  • 1
    In case anyone is wondering, the question came up here: worldbuilding.stackexchange.com/questions/34599/…
    – user11522
    Commented Jan 29, 2016 at 11:53
  • Apart from this definition apparently not being used according to the current answers below, I would argue that this definition of war is still too vague to produce verifiable statistics - for instance, a series of short and consecutive conflicts might cost a few hundred lives each, so your reliant in turn on a strong definition of "a conflict", and when it starts and ends.
    – IMSoP
    Commented Jan 31, 2016 at 17:26

2 Answers 2


The blog post you found cites a very fun journal article, but doesn't discuss it very well, denying the fun to all of us. The origin of this legend was figured out in 2005 and the results are available for all to read.

The most prominent figure we see in these anecdotes was a conscious hoax by peace activist Norman Cousins in 1953; Cousins claimed to newspaper readers that the "Norwegian Academy of Science" discovered only 292 years of peace since 3600 BC, and 14,513 wars since then. These were meant to be absurdly overspecific figures, and the headline in the newspaper in which his article appeared calls his work an "Imaginary Experiment".

But the idea seems to have existed before Cousins's hoax. In particular, a magazine called the Moskovskie Vedomosti claimed that there had been 227 years of peace, and 268 wars (not years of war; this was possibly misremembered by later authors). But the Vedomosti could not be located by the authors of this study, and in any case there seemed to be several figures floating around. The authors ran across a similar claim by a writer named Bloch, and are pleased at their good fortune when they find him:

For it so happens that Bloch, in a tiny note, refers to his source: a Russian military encyclopedia Encyclopädie der Kriegs- und Marinewissenschaften (St. Petersburg, 1885). We were lucky enough to obtain a copy of the article in the encyclopedia Bloch refers to - after many months of writing to libraries all over the world, to no avail.

In the lemma on war Lieut. General G.A. Leer refers, in his turn, to his source of the war figures: the work of the French philosopher Odysse Barot Lettres sur la Philosophie de l'Histoire (Paris, 1864). We were afraid that this work, too, would refer to another, yet more ancient, source, and that that source would, in its turn, refer to a still more ancient one, in a kind of infinite regression. But again we were lucky enough (after many a month of writing to libraries in France, to no avail), to obtain a copy, and this time we hit the jackpot: Barot's book is indubitably the one and only primordial original Source of sources; totally obscure itself, but immortalized by the myth it helped to create.

The original text is quoted; it presents the author's count of peace treaties that have been signed, and his tabulation of 227 years of peace, which seems (implicitly) to presume that times before peace treaties are signed are times of war.

But what exactly do these figures mean, presuming of course that Barot did not dream them up but actually catalogued and counted all these treaties (which is uncertain as he nowhere presents such a list: we have to believe him on his word)? As may be gathered from the quotation above, what Barot actually counted were peace treaties along with alliance and amity treaties, and NOT WARS. Cousins, as well as all of his predecessors, have drawn the totally unsubstantiated conclusion that the number of peace treaties equals the number of wars, under the assumption that all wars are ended by means of peace treaties. But apart from the volatility of such an assumption, it is not only peace treaties Barot counted, but also treaties of alliance and amity, and these do not necessarily, or not at all, justify the assumption of warlike activities.

The paywalled article that referenced this essay could find no scientific source for claims of two hundred and XX years of peace, and even the French source cited in this essay does not explain his methodology.

  • The links in the answer don't mention the "268" year claim.
    – DavePhD
    Commented Jan 29, 2016 at 12:27
  • Vespasien V. Pella published "268 years" in 1931. When does it say MacArthur came up with it, before or after 1931?
    – DavePhD
    Commented Jan 29, 2016 at 12:38
  • Many sources attribute it to the Society for International Law. books.google.com/…
    – DavePhD
    Commented Jan 29, 2016 at 12:52
  • 2
    The 1932 book "Inevitable War" says "In February, 1932, statistics published in Poland computed history to extend back 3,421 years, during which about 8,000 peace treaties were concluded, and in which time there were only 268 years of peace against 3,153 years of war" and cites to reference 20, but I don't have the full text to see what reference 20 is. books.google.com/…
    – DavePhD
    Commented Jan 29, 2016 at 13:19
  • Sorry for skimming over my sources last night; the Polish branch of this legend seems to be a distortion of an earlier Russian branch.
    – Avery
    Commented Jan 29, 2016 at 21:13

The second article linked in the OP incorrectly states that the origin of the "268 years" figure is a 1968 book. The "268 years" claim is actually much older.

The actual origin is "La criminalité de la guerre et les illusions de la paix [The Crime of War and the Illusions of Peace]" in Revue Internationale de droit penal (Paris) 1931, volume 8, pages 65-75.

An English abstract of which is provided in Social Science Abstracts Volume 4, page 1489.

From 1496 B.C. to 1925, A.D. a period of 3,421 years, there were nearly 3,153 years of war and scarcely 268 years of universal peace.

Only by reading the original could the author's intent be analyzed.

This statement, as alluded to in another answer, is an update of earlier analysis in Lettres sur la philosophie de l'histoire , quoting page 20:

En face de ces 8397 conventions solennelles de paix, d'alliance, d'amitié, on ne peut trouver dans cette longue période de 3357 années, — 1496 avant J. C., 1861, — que 227 années de paix, contre 3130 années de guerre

machine translated:

In front of these 8397 solemn agreements of peace, alliance, of friendship, one can only find in this 3357 years long period, - 1496 BC, 1861, - 227 years of peace, against 3130 years of war

The above is analyzed in "Policy Legends and Folklists: Traditional Beliefs in the Public Sphere" Journal of American Folklore volume 123, pages 150-178:

Do his numbers have any basis? The treaty count has a very weak claim to credence, as he consulted many diplomatic compendia, which he footnotes. However, he gives no rules for what qualifies as a treaty of peace, alliance, or friendship. Should the count include two monarchs agreeing to a royal marriage or coordinating with each other to conquer their neighbor? Perhaps he was unconcerned with these distinctions, seeing any treaty as situated at the positive law pole in his law-versus-force metaphysics. Some of his treaties are imaginary. He drew his earliest, in 1496 B.C., from a list in Histoire des anciens traitez by Jean Barbeyrac, professor of law at Groningen, but Barbeyrac himself assigned the event to “those obscure times when history is mixed with fable” (1739, vol. 1:1). Historians would now call it pure fable and date the Amphic-tyonic Leagues a millennium later. Although Odysse-Barot’s treaty count has negligible scholarly value, it was pioneering in its concept and was perhaps the first datum to embody to such a degree the approach of current statistical peace research—empirical, quantitative, and replicable. Nothing supports his claims for treaty length, war-free years, or deaths. He related his research on the treaty count with gusto: “These investigations which at first glance seem dry imperceptibly acquire an attraction that one would never have suspected; the ponderous and dusty folios become as entertaining as a novel” (1864:95). However, he is silent about his collection of the other numbers.

  • Thanks for your answer. This information seems to contradict Avery's answer. Not being able to read the original French myself, I'm not able to make a definitive judgment here, but overall I found the supporting references provided by Avery more convincing. I appreciate this contribution to the discussion though.
    – user11522
    Commented Jan 31, 2016 at 12:11
  • @dan1111 the one mistake in the other answer is the claim that " a magazine called the Moskovskie Vedomosti claimed ... 268 wars". Instead it is a daily newspaper and claimed "286" wars. The article was by Gustave Valbert, either 30 or 31 March 1894 and it said "286 wars in Europe". See discussion on pages 13 and 14 of the 1911 book "War and Its Alleged Benefits" books.google.com/…
    – DavePhD
    Commented Jan 31, 2016 at 15:28

You must log in to answer this question.