Is it sad that every time I have to write a paper for class (that involves a fair amount of research) I just want to turn it into a podcast?  I haven’t decided if that’s sad or not.  For many reasons, it’s fitting because of the presentational nature of the data but in the case of a paper the restrictions on length seem to cut a lot of interesting stuff short.  So a podcast is much more appropriate if I want to drag this thing out and explain a bunch of stuff.


Oddly enough, however, this one will flip that logic on its head as this podcast is about a very small portion of the paper I’m writing.  That paper is on the “Imagination Revolution” that took place around 50k years ago during what is commonly called “The Upper Pleistocene.”  What’s that?  Yeah, remember that shit, there will be a quiz later.


Where to begin?  This might be a bit heavy on human evolution so if you don’t want to get an anthropology lecture, this may not be the podcast for you.  That said, I find this utterly fascinating and this topic is critical to understanding our humble origins.  A little backstory on the paper then I promise to teach you something about all of those cool fossils you hear about (like Lucy).


We have a choice of three topics – the other two involve why villages gave up autonomy to become part of chiefdoms and why would hunter/gatherers give up that lifestyle and start food production?  If I’m brutally honest, all three topics are actually of interest to me though this one – the nature of cognition and how we understand the birthplace of culture is too good to pass up.  Mostly because of the enormous speculative nature of it.  It is an impossibly complex process and to try to simply say we evolved advanced brains because of one simple reason is an exercise in total futility.  But I digress.  The actual assignment (the question I chose) reads something like this:

“Was there an ‘imagination revolution’ in the Upper Paleolithic?  Use your additional readings to consider the scholarly positions on the significance, pace and “punctuation” of the origins of fully modern human cognition. Which of these positions do you find the most compelling in your case studies?  How, where, and when would you argue that the transition to fully modern human cognition took place?  How would you address criticisms from supporters of other positions?”


As for the “additional readings” I will not cover them in this podcast but suffice to say the archaeological community is utterly divided on this.  Some claim it to have been a “revolution” where others claim it was “a long, gradual, sometimes recursive process.” (Shea 2011).  John Shea seems to have something that can be described as “both” and it’s effectively the approach I’m going to take.  I know, shock/horror, I think the answer lies somewhere in the middle.  I’m incredibly predictable – did you see that coming?  No?  Then I’m occasionally predictable.  Look, another middle position.


So some fancy words are like to come up that I will explain away as they do.  One you certainly noticed that already came up is “Upper Pleistocene.”  For this, let’s take a short detour because explaining this seemingly fancy-sounding names is kind of hilarious.  The “cene’s” are referring to rock layers and there are six of them in what is called the “Cenozoic” area.  Now, before you run to the hills because I’m about to toss a bunch of Greek words at you, let’s look at the etymology of Cenozoic.  As translated, it means “recent life.”  The six epochs of the Cenozoic I will now read to you in plain English.  These go back to the time directly following the extinction of the dinosaurs about 65 million years ago.


They are… are you ready?  Cuz I find this hilarious.  The “old-new,” then the “dawn-new,” (kinda poetic I guess), then the “few-new” (as in we don’t find many lifeforms in this era), then the “less-new” then the “more-new” and finally, the Pleistocene which is the “most-new.”  Yeah… so the Upper Pleistocene is the “Most Recent Most New.”  Uh huh.  Now go, amaze your friends.


So the Pleistocene roughly coincides with a huge amount of time we call the Ice Age and the “Upper Pleistocene” is the end of that era (thus, the end of the Ice Age).  This age starts about 125k YA and ends at the dawn of agriculture, about 12k YA.  The “imagination revolution” as it is so-named by a few archaeologists takes place around 50k YA – that is, the “Upper” in the “Upper Pleistocene.”  If you’re wondering why they use “upper” or “lower” or are confused about it, its best to think of it in terms of rock layers.  The newest stuff is on top (the upper) and the oldest stuff is on bottom (the lower).  This is in contrast to the other series of terms – late and early – due to these somehow being more confusing for a lot of people.  Or one person.  Let’s call him Jim.  Moving on – let’s get him past the Greek.


Now if this were my paper I’d be going into some frankly boring specifics on Natufian tools and art and you frankly don’t want to hear all of that.  Natufian, by the way, is a name that is only further confusing.  Archaeological naming conventions aren’t based on any specific culture and the name that culture would have used.  In this case, the naming convention is based on the name of the first site found that depicts that culture.  In this case, the first site spotted that involved this era – Natuf. 


Some specifics I will go into, however, is the history of tools and the sudden onset of very specific, highly-individualized tools.  Tool use has been dated back to about 3 million years ago.  Since that time, tools remained relatively unchanged and uniform across locations.  In fact, they don’t evolve much.  A brief primer (I’ll use the Human Origins site of the Smithsonian website to illustrate this succinctly):

Early Stone Age (Early Pleistocene) Tools - known as Oldowan) - date back to about 3 million years.  This are basic tools, individual stones fashioned into individual implements.  Hammer stones, flakes, that sort of thing.  Around 1.76 MYA we saw the onset of more specialized handaxes, known as Acheulian tools. (Smithsonian 2017)

Then.. for over a million fucking years nothing changed or very, very little changed.  Around 200k years ago (now we’re in the Middle Pleistocene), we start to see some sophisticated tools and weapons.  Spears, arrowheads, awls, scrapers (the latter two for hide processing).  Still, while these tools are unique in the grand scheme of things, they’re still relatively similar across sites. By this stage, we are seeing the first Homo sapiens. So from 200k years ago until 50k years ago, tools were sophisticated but again, not unique.  For the most part, they used relatively similar pieces with no variations in say, handles or decoration.  We are now at the dawn of what is being referred to by some archaeologists as “the imagination revolution.”  At this point, “there is no doubt that relatively rapid shifts (within several centuries or a few thousand years) in core reduction strategies as well as bone and antler tool design occurred during the Upper Paleolithic in various regions.” (Bar-Yosef 2002)  He goes on to point out that this changes reflect differences in style without or rarely related to functional needs.


So… to put it quite eloquently (if I don’t say so myself)… what the fuck happened?


There is little archaeological evidence to suggest a slow, gradual change toward this.  Quite the opposite, actually.  There is a massive spike in individuality (including the sudden shaping of bone figures, use of pearls for jewelry, elaborate burials, etc.) after literally millions of years of none.  And this spike seems to have taken place over a relatively short time period.


So my paper focuses on an unique aspect that, so far as I can tell from some admittedly sparse research, is poorly understood.  Namely, the role of language.  But before I get to that (and erupt your brain with some math), let’s talk about someone who is likely to agree with me, John Shea.


Shea wrote a paper in 2011 that is blithely titled “Homo sapiens is as Homo sapiens was.”  His subtitle, “Behavioral Variability versus ‘Behavioral Modernity’ in Paleolithic Archaeology” is an excellent insight into his argument – that modern humans have not evolved behaviorally into modernity but rather are merely behaving different than our ancestors – not necessarily better or worse.  I’m going to supplement this argument with an exploration of language and how powerful the evolution of something as simple as more vowels can be.  But first!  Let’s discuss postcranial robusticity!  I know!  You’ve been sitting there asking yourself “but what about relative densities of femurs?!”  I can’t believe it’s taken me this long, I apologize.  Let’s get started.


This seems unrelated but it’s important to establish a causal relationship between what we call “archaic humans” (that is the first thing we’d call Homo sapiens) and what we call “early modern humans” which were formerly known as Cro-Magnon.  In a paper published in the American Journal of Physical Anthropology, a group of scientists showed that the robustness of the body below the neck (thus, “postcranial robusticity”) has drastically changed in the last 50k years such that the humans from that time period (early modern humans) more closely resemble archaic humans than humans of today.  It is worth pointing out, however, that we have a near-identical genome. (Spencer-Larsen et al 1993).  What does not match up, however, is this change and the change in brain size.  Brain size reaches its maximum several thousand years before this change – suggesting that these skeletal changes in the last 50k years are developmental


So again… what happened?


Well let’s get to the gist of this.  I will briefly add, in passing, that there’s a whole host of factors we are not going to get into.  One is the Younger Dryas, what is commonly called the “Little Ice Age” and the source of bunkery for climate denialists everywhere.  The Younger Dryas and the Ice Age created a lot of relatively short (evolutionarily speaking) climatic changes that forced various wildlife into what are known as “refugia.”  Super-fancy term for “only place they wouldn’t die.”  Another key point is our gross inability to nail down the exact time of the domestication of dogs.  The variability due to the genome and a whole host of other factors puts this range in the 34k-9k year range.  Not tremendously useful but enormous instructive if we are to further understand the development of human modernity.  But all of that is at least fascinating side-research if you want to understand say, why modern humans made it and why Neanderthals did not.  If you want this aspiring anthropologists perspective, I’ll just say “its complicated.”  Though one of the only things I got from attending the last CARTA symposium was that we have actually identified the exact genetic mutation that occurred that caused the human brain to grow like my dad cooks bacon – low and slow.  This happened after we split from chimps and bonobos but before we split from Neanderthals – suggesting they were likely just as smart as us and the difference (they only died out 30k YA) in survival may have been down to a cultural adaptation. (Huttner, et al 2015)


So… the hidden monster in all of this is language and that’s for a multitude of reasons but chief among which has to be that we don’t have any form of written language until 5k YA.  And if the fossil record of teeth is to be considered seriously, we have had speech-specific dental traits evolving since possibly as far back as 10 million years ago – an ancient ancestor of ours known as Ramapithecus. There is a noticeable decrease in size of what were once large canines (many apes still have large, useless canines that are a sexual dimorphism – the bigger canines are found on males and they are a sign of strength for mate selection).  But, there’s a small uptick in canine size during the time of Homo erectus and Neanderthals before a dramatic return to a drop in size starting around… you guessed it, the Upper Pleistocene.  Why is this important?  Among other dental features, a reduced canine is important to reduce phonetic errors.  In short, it makes it harder to annunciate if it’s large.  A reduction suggests a selection for clearer and clearer speech – better and more sophisticated communication.


Okay, dude.  I’ve listened to you drone on about ancient humans and stupid Greek names and how gorillas have sex, when are you going to just fucking tell me about this stupid revolution?  To that I say… you have a filthy mouth and… okay.  We’re here.


Language is arguable the strongest barrier to effective communication.  Without dumbing the argument down too much, all of us should know what it’s like to try to communicate with someone that does not speak your language.  In my case, living in California, it is just a reality that you will occasionally come across a native Spanish speaker who has limited grasp of English and you have to try to simplify your concepts and possibly even use (not-obvious) hand gestures and symbols to reach an understanding.  If we were to imagine a world where this is not just how different cultures speak to each other but how every single living person speaks to each other it becomes rather difficult to imagine us ever getting to a level of complexity that includes computers and automobiles and insane Japanese sex robots.  You know, the basics.


So if we lived in a community of hunter-gatherers 50k YA, we’d want to develop a sophisticated language.  Hell, let’s assume it’s 200k YA first.  We’d still want to share ideas efficiently but what have we been exposed to?  What sounds do we imagine cavemen making?  I’ll give you a simple idea… try to imagine the words that describe what we all do on the most fundamental level.  How many syllables to they have?  Eat.  Sleep.  Live.  Die.  Kill.   Obviously these words are in English, but here they are in Arabic:

تأكل ينام حي موت قتل

(pronounced – from right – “tah-kul,” “yanaam,” “heeya,” “moot,” and “qua-tel.” respectively)

The same goes for Spanish (comer, dormir, vivir, morir, matar), Chinese (che, shay-jeow, shen-guwo, sih, sha – sidenote, I have thoughts on why “sleep” and “live” are longer in Chinese but they have to do with how that culture sees individuals and groups), and Swahili (kula, usingizi, kuishi, kufa, kuua).  Compare these words to something like “government” and we start introducing multiple syllables.  Our language uses only three (with some nuance into the pronunciation combining consonants).  What about the others?  Arabic also has three – hoo-koo-mah (written with five letters).  Chinese doesn’t quite fit my argument but nonetheless it’s somewhat complex (zheng-fu) – only two syllables.  Swahili has four syllables – serikali.  This is an isolated example using language but I hope I’m making something like a point – our oldest words are the simplest to say and our newest words are more complicated (“turnt” notwithstanding).


This begins to make sense when you consider how language would have started and developed.  Let’s consider developing our own language and let’s mathematically consider the possibilities.  Let’s suppose our language only has two vowels and two consonants and all of our words use one of each. “Ma” or “Do” show all four of our letters.  In mathematical terms, we have two places for letters and each place can be occupied by two letters.  If we multiply place one’s number of possible values by place two’s possible number of values, we get the total number of possible words.  In this case, four.  Ma, Moo, Da, Doo.  If we then say that we can flip the vowels and consonants also, we double our vocabulary.  Am, Oom, Ad, Ood.  With just four letters we have created a language of eight words.  But what if we say that words can just use two vowels?  Or just two consonants?  Duh-duh and muh-muh (consider the possibility that the flat “u” sound is just how any consonant sound “finishes”) as well as two vowels with a “glottal stop” – ah-ah or oo-oo, ah-oo or oo-ah.  Now we can just say that we have two letter words with four possible combinations – we’ve again doubled our vocabulary to 16 possible words – again using only four letters.  The mathematical formula here is very, very simple if you knowingly limit the number of possible “places” and don’t combine consonants.  So words like burn or stem or drive are not yet possible.  If the formula is, for instance, consonant-vowel-consonant-vowel (mama), and you have ten consonants in your primitive language and only two vowels, then how many words can you have?


It gets better.  If I say that we can alternate the cadence so now it can be vowel-consonant-vowel-consonant as well – then I have doubled my simple language to 800 words.  That is to say nothing of making a four-syllable-all-vowel word or possibly adding complex double-consonant words (like burn).  That is with a language consisting only of ten consonants and two vowels.  Consider the average modern English speaker has about 25k words in her spoken vocabulary (source: and you start to grasp just how powerful language is. 


I’m proposing that one of the key limiting factors may have just been human beings testing out how much horsepower they have.  Any avid video game player that follows the lives of some of these game consoles knows that the most hardware-challenging titles do not release at launch.  It takes developers years to figure out how to get the most out of the system’s architecture and I see no reason to think ancient humans were any different.  Evolution was supplying the mind with ever more capable pattern-spotting tools, making us better and better at hunting (we began incorporating more meat into our diet 2 MYA – well within the proposed time constraint proposed about “low and slow” brains) (Milton 1999).


The leap to being able to develop more and more sophisticated means of communication could have potentially triggered the explosion in complexity that we see and could help explain why there were several smaller blips on the radar around this time as well.  If we had been using primitive language for a few millennia and the language slowly grew more and more robust (think more consonants and combinations), it is no wonder that a sudden increase in complexity happened as we neared a tipping point in ability.  If we briefly return to the math from above, consider a system that used 12 consonants and 3 vowels (still very simple).  The jump from 3 sounds per word to 4 sounds per word jumps our potential word count from 3375 to 50,625!!! 


By now, we’re just applying power rules.  Without adding additional letters, only adding additional places for letters, we increase our word total by an order of magnitude.  For a visual representation – here is what the graph of possible number of words looks like if we only have ten letters.  The large numbers on the y-axis are possible number of words and the small numbers on the x axis are number of places for letter sounds:

Graph of Language.png

Figure 1 - Where y is the potential number of words in the system, 10 is the given number of letters in a hypothetical primitive language, and x is the number of individual phonetic values in a single word.


As you can hopefully see, there is an enormous leap in the potential for complexity once we make the leap from 4 available “phonemes” to 5.  In terms of human cognitive capacity, the fossil record suggests we hit the ability to do this about 200k years ago (or even earlier!) but the onset of utilizing it in this very specific way was unlikely to happen for a very, very long time.  As you can see, when it did happen, there would be a very obvious added bonus for culturally complex societies – more and better ways to insult your mother-in-law.


What’s the downside to my argument?  There is no fuckin’ way to test this hypothesis.  We can’t bonk an ancient human on the head and ask him about phonemes and syntax.  We can’t even look at an ancient brain – we can only infer brain size based on skull casings.  Turns out soft, fleshy tissue doesn’t last very long.  Who knew?!

Did you enjoy listening to this as much as I enjoyed researching it?  I doubt it!  But if you did, please take a moment to tell a friend and share your enthusiasm.  If you’re a repeat listener/reader then you know I’m not very regular with my releases but if I obviously had a larger following I’d take some time out of my day and possibly turn more of my college paper projects into podcasts (or plan some interviews with intelligent people for that matter).  So if you could hit the subscribe button if you listen on your RSS feed or on iTunes.  Send me an email ( and I’ll add you to an email list if it ever grows large enough to justify one (promise I won’t give Cambridge Analytica your data – spoiler: they probably already have yours if you’re on Facebook).

Also, if you are capable of giving then I obviously encourage that.  I have a fairly bare Patreon page but, as I always say, if it takes off I’ll put significantly more effort into it and create a legitimate platform.  If you’re interested in giving, that link is


Thanks for listening to what is sure to be one of many podcasts on actual human beings!


Works Cited

Bar-Yosef, Ofer. 2002. "The Upper Paleolithic Revolution." Annual Review of Anthropology 1, 363-393.

Christopher B. Ruff, Erik Trinkaus, Alan Walker, and Clark Spencer Larsen. 1993. "Postcranial Robusticity in Homo." American Journal of Physical Anthropology.

History, Smithsonian National Museum of Natural. 2017. Stone Tools. November 8.


Milton, Katharine. 1999. "A Hypothesis to Explain the Role of Meat-Eating in Human Evolution." Evolutionary Anthropology 11-21.

Shea, John J. 2011. "Homo sapiens Is as Homo sapiens Was: Behavioral Variability versus "Behavioral Modernity" in Paleolithic Archaeology." Current Anthropology 1-35.

Sheets, John. 1977. "Homind Dental Evolution and the Origins of Language." Man 518-526.

Various (Unnamed). 2017. November 8.