How to speak stone-age bullshit

This is a guest post by Christine Cuskley.

As a general rule, there is much that is very badly written about specialist academic disciplines. From farts curing cancer to hot wet aliens, academic research often isn’t well-represented in popular outlets. Research on language and language evolution are no exception. So, generally, people who spend their working days immersed in language research let such flawed reports flow over them like so many offers to publish their thesis for the small fee of £300. You can’t possibly feel miffed at every one or you would explode and get nothing done, and there’s already so much on the internet to distract me even the most focused linguist.

But today I’ve seen something so utterly cringeworthy that it simply shall not pass. In a recent article for The Daily Mail, a man named Christopher Stevens For The Daily Mail adapts an excerpt from a book called Written in Stone, which is by Christopher Stevens (presumably) For Himself. I will direct all criticism towards his nom de plume, on the off chance that he originally submitted a clear, well-thought out, and accurate excerpt from his excellent book which was then mercilessly butchered by an ignorant editor. As an up-front disclaimer, I haven’t actually read the book itself, and I am very unlikely to. Assuming this adapted excerpt is any indication, the book is a mess.

At the heart of it, it seems quite unlikely that this is simply a zealous editor gone mad, since the piece claims to be adapted directly from the book. The adaptation is called “How to talk like a stone-age man: A fascinating new book reveals how our ancient ancestors spoke. And you’ll be astonished by how familiar it all sounds“. If you’re still awake by the end of that rousing mouth-full of a title, don’t worry. It immediately gets worse1:

Here’s how to talk like a stone-age man: say the word ‘pu’. Your mouth is pursed, your nose is narrowed. You are blowing out a breath, as if to dispel a bad smell. In the Stone Age, the sound pu meant exactly what it means today. This is how language began. The earliest words in English date back at least 8,000 years – and they describe themselves: we can work out what the words meant by the shapes our lips form when we say them.

Oh sweet, sweet, Christopher Stevens For The Daily Mail, this is already so, so wrong, and we’re only a few sentences in. Did you even ask anyone about this, even the internet? Let me fix it for you:

Here’s how to talk like a stone-age man if he had lived somewhere in Europe: say the word ‘pu’. Your mouth is pursed, your nose is narrowed. You are blowing out a breath, as if to dispel a bad smell. In the Stone Age, the sounds in the word pu meant exactly what it they means today if you spoke the language considered to be the distant parent of English and other European languages. This is how the language that eventually gave rise to English began. The earliest words in English the language family Indo European – to which English belongs – are estimated to date back at least 8,000 years – and they describe themselves: we can work out what the words meant by the shapes our lips form when we say them.

This is now at least somewhere in the neighborhood of accurate considering what we know about English and its great-grandparent, Proto Indo-European (PIE). And let me just hammer home: PIE is not the mother of all languages – it represents a single family. A lot of people may speak languages which are descendants of PIE, but a) the Indo-European family represents a fraction of the diversity of human language, and b) Indo European and its descendants are not the same thing. My changes may make it sound less sensational, but it is likely no less interesting for your average reader unaware of PIE, or of the existence of language families at all.

Christopher Stevens For The Daily Mail is not talking about all “stone age language” (whatever that even is), but only Indo-European. PIE can say nothing about what *pu may or may not have meant in Australia or South America circa 6000 B.C.E. This is not “how language began”; this is a very limited insight into a stage of a single language family (there are over 200 language families, by the way). To suggest a few reconstructed words from PIE show “how language began” is like holding up a baby lion and suggesting it explains the origin of all multicellular life.


These are big, meaningful differences. I’m not opposed to a little artistic licence with the suggestion that *pu means poo because we get to make a face when we say it. I doubt there is specific empirical evidence to back up this claim; but sure, why not, very cute. But to say that words in PIE describe themselves is, first of all, confusing and circular. I think what Christopher Stevens For The Daily Mail actually means to say is that the forms of words in PIE describe the meanings of the words themselves to a hearer who has no knowledge of the language.

This turns out to be beside the point because it’s inaccurate anyway. The word inaccurate, by the way, comes from the latin root ad-(to) and curare (take care); as in, this is seriously careless. At best, this claim is subjectively true of a handful of PIE words, just as it is true of a handful of English words (e.g., hiss). If you take the breadth of the claim at face value it is just wrong: we could not magically understand someone speaking PIE as long as we stared out their mouth intently enough. As linguists go, I am unusually receptive to ideas about sound symbolism in language evolution – the idea that sounds can hold meaning intrinisically – but this is very, very lazy. The same idea can easily be more accurately portrayed with no overall loss of interest and wonder2.

It only gets worse from there. Christopher Stevens For The Daily Mail tells a few more just-so stories about a couple more PIE words, and then says “[the language] doesn’t even have a proper name. Archaeologists call it Proto Indo European.” I’m not sure what makes this name improper, but it sounds like a proper name to me. And archaelogists generally don’t care much about PIE unless it’s a bit of a hobby; I suppose he means anthropologists, and he wants to mean linguists. But sure, archaeologists; lets not trouble ourselves with Googling anything.

This is followed by another barrage of 10-15 cute stories about why words got to mean what they do from what was apparently the pre-eminent language among talking apes circa 6000 B.C.E. I have no specific etymological knowledge that can refute any of the histories he tells for various words, but he seems to be trying to point out that words are related across time and within a language.

This fact is inherently interesting, and a fact that isn’t necessarily appreciated by non-linguists; but he somehow takes all of the fun out of it while also barely making it clear. He also fails to point out perhaps the most interesting thing of all about how languages change over time: this is not a special feature of PIE, but feature of all other language families around the world. Language change over time and speciation of words is a fact about how language works. That is interesting, but apparently less so than some random pictures of what appear to be Neanderthals, which is a) a different hominid species than modern humans, and b) extinct roughly 24-34,000 years prior to the “ancient” time period under discussion anyway.

Much like the inclusion of interpretive neanderthal sketches and costuming, some of the stories about PIE words are barely coherent:

Prehistoric man took death seriously. We can decode some ancient funeral rituals from the old words and their modern meanings. ‘Mor’ is the IE root of mourn and also of moan. That implies that grieving was a noisy business.

No, we really can’t decode funeral rituals from word relatedness, and no responsible linguist or anthropologist (and definitely not an archaeologist) would try to do this. Grass is another euphemism for wacky tabacky, does this straightforwardly imply that marijuana enthusiasts are experts in lawn maintenance? Not even remotely. This bizzarre method of trying to “decode” broad rituals from one pair of words ignores yet another interesting area of language change: word meanings shift in different ways – often through abstract metaphors – and this can mask their origins completely.

After several more unrelated etymologies which devolve into nothing more than bullet points, Christopher Stevens For The Daily Mail concludes thusly:

It was August Schleicher, a 19th-century German, who theorised that languages are living organisms that are born, bloom and die. He even discovered the concept of evolution, years before Charles Darwin – but Schleicher applied it to words instead of animals. So think before you speak. The words in your mouth are alive . . . and are as old as time.

August Schleicher was a contemporary of Darwin, and in any event, neither of them discovered the concept of evolution. If Christopher Stevens For The Daily Mail had bothered to Google “etymology of evolution”, he would have immediately seen that its use in the conceptual sense of “growth to maturity and development” is attested as early as the 17th century. I can’t find a specific citation that time is older than PIE, but I think that’s because it’s a trivial fact once you know what PIE is. In the end, the whole piece gives the impression that Christopher Stevens For the Daily Mail actually never bothered to Google PIE, though he somehow happened across a few reconstructions.

Finally, there are a few major other major errors of omission. A major fact about PIE is totally glossed over, if not willfully misrepresented: since we don’t have a written record of PIE, we don’t actually know what it was like. What we have are reconstructions that are inferences based on patterns in the modern descendants of PIE. These reconstructions rely on well-founded assumptions and careful study, but they are still reconstructions. We know the word for mother in, for example, French, Italian, and German, in a way that we can never know what words were in PIE. This is why first year linguistics students learn that reconstructions are presented with a ‘*’ to denote uncertainty, as in *pu vs poo. The certainty with which Christopher Stevens For The Daily Mail proclaims the meanings of “stone age” words completely conceals this important fact.

It also represents another missed opportunity: the fact that linguists have found ways to make partial inferential reconstructions of long extinct languages is exciting – there is no good reason to leave this out; it is simply a lack of adcurare. I won’t even get into the fact that linguistics generally has a historical problem of focusing way too much attention on Indo European languages; a fact which one can attribute partially to availability of data, but is undeniably related to a (thankfully waning) European tendency to assume the centre of the universe is somewhere in Europe. It bears reiterating that Christopher Stevens For The Daily Mail seems woefully late to the party in realising there are several other continets and hundreds of other language families.

Overall this adaptation is meandering, ranging from factually questionable to blatantly inaccurate and misleading. I suppose it shows enthusiasm, but that’s about it. It makes me want to stay as far away from the book as possible, and recommend others do the same. If this “adaptation” of the book has you interested in historical linguistics, etymology, or the history of language, I’m glad I guess, welcome to The Nerdery! But I’m also so, so, sorry. This is like your first trip to the US being a layover inside of Newark Airport, and it has somehow miraculously piqued your interest in visiting the rest of the country. I’m so happy you’re interested, but it’s really too bad this had to be your first experience. Actual historical linguistics can offer so much more.

1Note that in these quotations I have removed the original article’s eclectic formatting in which every sentence is its own special paragraph

2Full disclosure: I was interviewed for the New Scientist article linked.

If you’re looking for a good window into the history of English, I recommend:

Our Magnificent Bastard Tongue by John McWhorter

Investigations in Sociohistorical Linguistics: Stories of Colonisation and Contact by Peter Trudgill

The Story of English in 100 Words by David Crystal


Displaying me.jpg

Christine Cuskley is a linguist/psychologist/nerd type who researches the evolution of language and communication at the Institute for Complex Systems in Rome, Italy and the Institute for Scientific Interchange in Turin, Italy. In her spare time she enjoys quilting, comestibles, Quora, and occasional long rants on her personal blog, Mean and Pregnant.

6 thoughts on “How to speak stone-age bullshit”

  1. Well said and many thanks for the post.

    I’d like to add “The English Language: A Historical Introduction” by Charles Barber to your recommended books, though it appears to be depressingly absent from amazon…

  2. Thanks for this great post! – Judging from the excerpts on GoogleBooks, Christopher Stevens For Himself is even worse than Christopher Stevens For The Daily Mail: – e.g. “This prehistoric language followed strict rules of grammar that have also lived on. When words changed tense, for instance, they changed their ending: in the past tense, a ‘d’ was added, just as in English.” This is just so wrong…
    But I really like it how the Daily Mail article is illustrated with pictures and videoclips from a cheap 1960s movie – seems like someone noticed how ridiculous this text is and wanted to mark it as some kind of satire 😉

  3. I didn’t even see that little gem about the past tense (something I’ve recently worked on! ) – that’s almost worse than the moaning.

    I thought about including a clip from another classic documentary on ancient language, Encino Man, but I wanted to keep it concise.

  4. What a great commentary! Too bad I read this after I bought the book! But please help me, I have been trying to talk to a linguist about this; Our species spent over 200,000 years eradicating competitive hominids who preceded us. Every one of them retained the heavy brow ridge off our ape ancestors (well, common ancestors), and every one of them, except Neanderthal, had smaller brains therefore no forehead. I call them “Low-Brows”. Think about it, there must have been a word in every proto-language for low brows. I would guess that that word is at the root of some words for inferior races or people (?); stupid, dumb, idiot, moron, and what? Its there any indication of this that you know of?

    I was working on a summary of all the books I could find on our deep history, when I realized this. If you’re interested, let’s talk or communicate.


Leave a Reply