Conventional human signs 40,000 BP
« previous post | next post »
Painstaking research by Christian Benz, Ewa Dutkewicz, et al.
Significance
Abstract
As humans, we store and share information. This allows us to distribute knowledge necessary for survival and to coordinate large groups. Our hominin ancestors harnessed the surfaces of mobile artifacts and cave walls as information carriers since the Paleolithic time period. Theories abound as to the meaning and function of these Paleolithic signs. However, very little is known about their basic, measurable properties. We here analyze a corpus of more than 200 mobile objects of a 43,000 to 34,000 y old Aurignacian culture—associated with the first modern humans to settle in Central Europe. These objects are adorned with several thousand geometric signs. We apply classification algorithms and statistical models to capture their quantitative properties. First, our analyses illustrate that these sign sequences are clearly distinguishable from modern day writing. Second, however, their statistical properties are comparable to sign sequences on the earliest protocuneiform tablets. Third, Paleolithic signs were systematically applied to yield higher information density on certain types of objects, e.g. ivory figurines compared to tools. These results cannot be taken to strictly prove that Aurignacian sign sequences encoded numero-ideographic information as in the case of protocuneiform. However, they prove that the first hunter-gatherers arriving in Europe already applied sign sequences of comparable complexity in a deliberate, systematic, and conventional manner—several ten thousand years before the advent of genuine writing.
I like the first paragraph of the paper proper, especially the first sentence:
As humans, we store information outside of our minds. This capacity for information encoding beyond the here and now of spoken words is nowadays reflected in symbols and written language. It is a hallmark of human cognitive evolution and underlies artificial computing systems. Without it, the information age would not be possible. Archaeologists have identified reflections of this capacity in excavation layers dating back to the Paleolithic period of 400,000 to 15,000 thousand years ago.
With beautifully clear photographs (Fig. 1 and Fig. 2),
As comparative samples we employ material from the earliest periods of protocuneiform in ancient Mesopotamia: Uruk V (approx. 3500–3350 BC), Uruk IV (approx. 3350–3200 BC), and Uruk III (approx. 3200–3000 BC). See Fig. 2 for example tablets. We also harness the Text Data Diversity (TeDDi) sample of 89 different languages written in 16 scripts (59) as a modern day counterpart.
This continues the research of Genevieve von Petzinger, for which see "Selected Readings" below.
Selected readings
- "The origins of graphic communication, pt. 2" (2/3/17)
- "The origins of graphic communication " (11/21/15) — must read; ongoing work on the Europe-wide Paleolithic symbol systems of Genevieve von Petzinger, the granddaughter of a World War II codebreaker at Bletchley Park in England
[Thanks to Ted McClure]
Chas Belov said,
May 21, 2026 @ 10:08 pm
¿Not "…since several hundred thousand years ago"? ¿Has this usage changed?
HS said,
May 21, 2026 @ 10:57 pm
¿Has this usage changed?
Presumably not if, like the authors, you are German. But your "since several hundred thousand years ago" still sounds stilted to me – "for several hundred thousand years" sounds like more natural English to my ear.
This proves that the first hunter-gatherers arriving in Europe already developed a system of intentional and conventional signs on mobile artifacts.
This also sounds wrong to me – I would say "had already developed". But there seem to be significant differences between British English and American English in the use of the perfective aspect in cases like this.
However, none of this grammatical nit-picking in any way alters the fact that this looks like a fascinating paper!
Peter Grubtal said,
May 22, 2026 @ 2:06 am
I've heard the riposte from a German to nit-picking like this "it's international English!".
David Marjanović said,
May 22, 2026 @ 5:35 am
Indeed, the German way to say "I've been doing this for 3 weeks" is "I'm doing this since 3 weeks" – plain present tense, no "since"/"for" distinction.
"several ten thousand years before the advent of genuine writing" (at the end of the abstract) looks German to me, too; zehntausend is not taken apart to form the plural zehntausende "tens of thousands".
Chris Button said,
May 23, 2026 @ 2:40 pm
English might be the odd one out here.
French, Portuguese, Spanish would all use the plain present tense with "since" too. I'd be curious to hear from speakers of other related languages.
Yves Rehbein said,
May 23, 2026 @ 3:19 pm
I am not sure we can say that with confidence. It has the advantage of being terse in the face of length restrictions on articles. I don't remember when we last discussed "the oldest" (or similar) but it was the kind of petty nitpicking that misses the point, as am I.