What if language is actually a living organism, a parasite? A brief exploration of the connections between linguistics and biology, and the ideas of the Leiden school.
Note: at 14:06 there’s a small audio glitch I couldn’t fix. The name that’s missing is “Gerald Edelman”, so it should be: “They point to Gerald Edelman’s theory of neural group selection”. Sorry about that!
Timestamps:
00:00 – 01:20 : Some fiction
01:22 – 12:03 : Some parallels between linguistics and biology
12:05 – 18:24 : Some of the Leiden school’s ideas
18:25 – 26:07 : Some broader considerations
26:08 – 27:19 : Some housekeeping
Still very much a work in progress, but a huge thanks to all the people making tutorials out there for stuff like Blender, DAWs, etc. If this video looks like a cheap rip-off of a @BobbyBroccoli video, it’s because it is… ^^ A big thank you to him for making a public tutorial on how he makes his videos.
Bibliography:
– KOERNER, E.F.K. 1980. Pilot and Parasite Disciplines in the Development of Linguistic Sciences. Folia Linguistics Historica, 1(1).
– KOERNER, E.F.K. and ASHER, R.E. 1995. Concise History of the Languages Sciences: From the Sumerians to the Cognitivists. Kidlington, Oxford, United Kingdom: Pergamon (Elsevier Science Ltd.).
– KORTLANDT, F. 1983. A Parasitological View of Non-constructible Sets. Studia linguistica diachronica et synchronica. Berlin, Germany: Mouton, 1985.
– KORTLANDT, F. 2002. The Origin and Nature of the Linguistic Parasite. Language in Time and Space. Berlin, Germany: Mouton, 2003.
– MCCARTHY, C. 2017. The Kekulé Problem – Where Did Language Come From? Nautilus, No.19. New York: NautilusThink Inc, 2017.
– REDONDO, P. 2021. Filippo Sassetti and Thomas Stephens in the beginnings of Indo-European linguistics. Academia Letters, Article 2158.
– SALVERDA, R. Is Language A Virus? Reflections on the Use of Biological Metaphors in the Study of Language. In: 1998. Productivity and Creativity. De Gruyter.
– VAN DRIEM, G. 2001. Languages of the Himalayas. An Ethnolinguistic Handbook of the Greater Himalayas Region. Leiden: Brill.
– VAN DRIEM, G. 2005. The Language Organism: The Leiden Theory of Language Evolution. In: Language Acquisition, Change and Emergence: Essays in Evolutionary Linguistics. Hong Kong: City University of Hong Kong Press.
– VAN DRIEM, G. The Language Organism: Parasite or Mutualist? In: Evidence and Counter-Evidence, Festschrift Frederik Kortlandt, vol. 2. Amsterdam – New York: Rodopi, 2008.
– VAN DRIEM, G. Symbiosism, Symbolism and the Perils of Memetic Management. In: Language and Culture in Northeast India and Beyond. Canberra: Asia-Pacific Linguistics, 2015.
At the end of a long journey through the expanses
of space, a tiny parasite, still in its larval form, landed on a nondescript fleck of a planet.
There, it found a home in the brain of a local hominid, and rapidly spread across the entire
species. It gave its hosts unprecedented powers of communication and abstraction, for you see, this
was a language parasite. Those who wielded those powers invented tools, weapons and religions,
roads, railways and skyscrapers. It never occurred to them that the parasite was dictating
their actions in pursuit of its own goals. In fact it never occurred to them that the parasite
was even there. They built radios and rockets, making their presence in the universe ever more
noticeable, completely oblivious to the silent threats that lurked in the distance. At every step
of the way, the hosts told themselves “this is progress”. Then one day, the skies shattered. In a
mere instant, all life on the planet was swallowed whole. None of the hosts survived. But the
parasite thrived. The stomach of the behemoth that had just eaten the planet was exactly where it had
intended to go all along. It was where it needed to be to start the next stage of its reproductive
cycle. Well, it’d make for some fun sci fi anyway. Science fiction and related genres is probably
where I’ve encountered this idea the most, this idea of language as a parasite, or a virus, or more broadly as some kind of organism. The
Ticket That Exploded by William Burroughs, Snow Crash by Neal Stephenson, though I guess in
that one it’s more of a computer virus really, Pontypool changes everything by Tony Burguess,
and Hot Skull by Mert Baykal, to name a few. But as it turns out, it’s not just
the remit of fiction. There are of course some superficial parallels in the
words we use to describe language. Like, we might say words have roots and stems
and they can fossilise. We might speak of living languages and dead languages.
But the parallels run deeper than that, and many of them have been around pretty much
from the start of linguistics as a modern science. A science doesn’t just pop out of thin air. Poof.
Physics. Tada. It’s more of a gradual process. So if you ask when did linguistics start as a
modern science, you might get different answers depending on who you ask. Some might point to the
works of Ferdinand de Saussure published in the early 20th century. Others might trace it back
further, to William Jones and the beginning of Indo-European studies in the late 18th century.
But overall, I think almost everyone would agree that the 19th century was pivotal to the rise
of modern linguistics. It’s a period where the study of language started to transition towards
something more like a modern science. The crown achievement of 19th century linguistics was the
construction of the Indo-European language family, through efforts that were driven primarily by
German scholars. Essentially, they took a bunch of languages from India to Western Europe,
found that they were related to one another, and traced back their origin all the way
back to a hypothetical common ancestor, called Proto-Indo-European. It’s not at the same
scale, but it’s a bit like figuring out that all mammals are related and that they all descend
from a common ancestor, some kind of cynodont. The basic impulse was nothing new. Basically:
“These two things look alike; I wonder if they might be related somehow”. Before the 19th
century, that idea pops up here and there, but it never gains enough momentum to really take
off. For example, in the 1580s, Thomas Stephens, an English Jesuit who was among the first
from England to travel to India, noticed some similarities between some of the languages
of India and Latin and Greek. A few years later, a Florentine merchant by the name of Filippo
Sassetti, who had also travelled to India, noticed some similarities between Sanskrit
and Italian vocabulary. But that was it. Neither of them dug any deeper into the
relationship between these languages. You also find some speculation about a possible
parent language, a first language that the rest descended from. A main contender for a long
time was Hebrew, based on Biblical exegesis. Later on there were others, like Scythian, but
none of those speculations led to much. And none of them brought the kind of theoretical and
empirical rigour we associate with modern science. And that’s exactly what these scholars
brought to the table starting around 1800. They clarified concepts, developed new methods,
and introduced a more rigorous approach than there had previously been. Conventionally, the
first falsifiable prediction in linguistics is attributed to the Neogrammarians, a German school
of linguistics active from the mid-1870s onwards. By then, linguistics was well on its way to
becoming a quote-on-quote proper science. So in the span of a few decades, from 1800
to 1870, they had advanced by leaps and bounds towards making the study of language an
autonomous science, with its own standards and its own methods. How did they pull it off? A
lot of things had to come together to make it possible. For example, there’s the fact after the
Napoleonic wars, the leaders of the various German states were incentivised to invest in higher
education, and this gave language scholars access to resources they might not otherwise have had. Or
there’s the publication of a number of works that compiled data on many different languages. Data
that used to be scattered across many different sources were now available in a single place,
which made it easier to compare them. But what really matters for us here is the influence that
other sciences had on the birth of linguistics. I mean, yeah, if you want to build a new
science, one of the things you might do is look for guidance and inspiration from other,
more mature sciences. And they did just that. One of the places they looked was
compared anatomy. In the early 1800s, Georges Cuvier introduced the concept of
“correlation of forms”. Basically, he was saying that each body structure is related to and
dependent on all of the other body structures. There’s an interrelationship between the parts
and the whole. This idea influenced language scholars like Friedrich Schlegel and Franz Bopp,
who both started to put a lot more focus on the structure of all the bits and pieces that make
up language and how those bits and pieces relate to one another. Bopp called his method of
language analysis “compared dissection”. Schlegel called for a “compared grammar”
as a direct parallel to “compared anatomy”. Another place they looked was botany. For
example, Carl Linnaeus, the Swedish naturalist who laid the groundwork for modern taxonomy,
became an exemplar of scientific rigour and language scholars cited and sought to emulate
his work. And one of the leading figures of 19th century linguistics, August Schleicher,
was also a passionate botanist. In his work on language, he sometimes referenced the works of
contemporary botanists, like Mathias Schleiden. Yet another place they looked was evolutionary
biology. An easy way to see this influence is through tree diagrams. You know the type: the
descendants are represented on the branches, and where those branches meet you get the
common ancestor they descended from. The first tree diagram to chart evolution seems to
have been made by Lamarck in 1809. Later on, Charles Darwin famously used tree
diagrams for his own theory of evolution. One of the main proponents of Darwinism in
Germany was Ernst Haeckel, who made dozens and dozens of ornate tree diagrams of his own. It so
happens that Haeckel was a close friend of August Schleicher. Schleicher was making tree diagrams
of his own, not to chart the evolution of species, but rather to chart the evolution of languages.
Schleicher’s trees actually predate Darwin’s and it seems like they arrived at the idea
independently. That being said, when Haeckel encouraged Schleicher to read Darwin, which he
did, he saw the close parallels between Darwin’s work on biology and his own work on languages,
and drew from that to bolster his own ideas. Not only did these language scholars
draw from these specific disciplines, but more generally a lot of them were
also influenced by natural organicism, a philosophical view that basically just says
this: if you want to understand the world, you should think of it like an organism. And
so a lot of these scholars thought of language as in some sense organic, at the very least
something that is subject to growth and decay. For example, Friedrich von Schlegel
described language as a “living tissue”. Wilhelm von Humboldt
wrote of a “language organism”. For them the connection between language
and organism was most likely metaphorical, but for others it seems to have been
quite literal. Apparently this was the case of August Schleicher. In his view,
languages were quite literally organisms: “Languages are natural organisms
that, regardless of man’s will, appear, grow and develop following
certain laws, and in turn they age and die. The series of phenomena that are usually
understood as “life” also apply to languages.” Whether they meant it literally or as
a metaphor, the idea that language is an organism played a crucial role in the birth
of linguistics. According to Reinier Salverda, the idea of language as an organism provided a
unifying principle for the study of language, a conceptual basis for them
to adopt methods from biology, and in time this led to the rise of linguistics
as an autonomous science. So yeah, the idea of a language organism has been around from pretty
much the beginning of modern linguistics. But starting with the Neogrammarians in the 1870s,
the idea of a language organism gradually lost its appeal. But other biological metaphors
would pop up later on in the 20th century, and they too would play a role in scientific
discourse. As Salverda describes, criticising old metaphors and offering new ones can be
part of how a science progresses. For example, in the 1950s, Noam Chomsky coined the metaphor of
language as a “mental organ”. What he was trying to capture with this metaphor is the idea that
language is something we’re already born with. He was arguing against the idea that we
“learn” language through experience. You know, like a maze. Each time you run into a dead end,
you learn from that experience. Don’t go this way. So you try another way. Rince and repeat until
you find the exit, assuming there is one. So, for Chomsky, we don’t actually learn language
that way. Language is innately there and grows, like an organ. There’s one small problem with that
metaphor though: you can grow up alone with a pack of wolves and your spleen is gonna grow just
fine. But you won’t get language. Unlike any other organ, for language to grow, you need
social interaction with other human beings. So you can criticise Chomsky’s metaphor on the
grounds that it doesn’t capture that fact about language. And then you can propose an alternative
metaphor that does capture that fact. Like, say, the virus metaphor, which is what someone like
Terrence Deacon uses. To quote Salverda again, “language, just like a virus, does not exist
and develop without contact between its bearers; its reproduction depends on interaction with
other humans.” So, occasionally biological metaphors like that do pop up again and
they continue to structure how we think about language. The metaphors themselves don’t
matter all that much, but it’s more what they represent. They’re a kind of stand-in for
much larger sets of ideas and arguments. But those are metaphors. In the 20th century,
you don’t really find anyone saying that language is actually an organism like August
Schleicher did. With one notable exception: in the mid 1980s, a small group of scholars from
the University of Leiden, in the Netherlands, boldly claimed that language is in fact a
parasite, literally. The founding figure of this school of thought is Frederick Kortlandt.
His ideas were later elaborated on by others, especially by Georges van Driem. Since
they hailed from the University of Leiden, their school of thought was
dubbed the Leiden school. What is meaning? That’s a question that
has stumped scholars for a very long time and there are multiple models to attempt to
explain it. The Leiden school uses the one introduced in the early 20th century by Genevan
linguist Ferdinand de Saussure. What Saussure proposed was what he called a linguistic
sign. The sign is made up of two parts. On one side you have the signifier.
In the case of spoken languages, the signifier is the string of sounds, like, say,
the sounds you need to make to pronounce the word “cat”. You could also call this the linguistic
form. And then on the other side, you have the signified. And that’s the mental concept
that is associated with the string of sounds, so in this case it’s the mental concept of a cat.
You could also call that the meaning. And the two, signifier and signified, or, form and meaning, are
glued together. They’re both part of this larger, indivisible whole that Saussure calls a linguistic
sign. Think of it like a sheet of paper. On one side you have the form, and on the other the
meaning. A sign can be a word, like cat, but it can also be something bigger or smaller than
a word. The smallest linguistic sign is called the morpheme. It’s the smallest unit that conveys
any meaning. For example, if I say “I missed you”, that small sound “ta” at the end of “missed”
is a morpheme, and here it indicates the past. So from that perspective, language ends up being
one big system where all these signs combine in multiple ways. It’s a bit like DNA. With DNA, you
have your four nucleotide bases. These bases can combine in different ways to produce larger units.
The morphemes of a language are a bit like the nucleotide bases in DNA, except that there’s a lot
more than just four of them. And these morphemes can combine to produce larger units, like words,
which can then combine to produce larger units, like phrases, which can then combine to produce
larger units, like sentences, and so on. For the Leiden school, all of these signs
have neural correlates, groups of neurons they’re associated with. They point to Gerald
Edelman’s “theory of neural group selection”, which posits that neurons are subject
to Darwinian selection processes, just like species. The neurons
compete for cortical space, and the linguistic signs they’re associated
with compete for mental representation space. But that’s just inside one person’s brain.
To spread from one person to another, linguistic meaning has to rely on linguistic
form, the string of sounds. To quote Kortlandt, “the sensorimotor channels of speaking
and hearing can be regarded as the male and female sex organs of the linguistic
parasite”. Essentially the idea is that, when you say something, what’s really happening
is that there’s a neural construct in your brain that is trying to replicate itself in another
person’s brain. This is how the linguistic parasite reproduces. And yes, they think
of it as actual, biological reproduction. According to Kortlandt, this approach to language
has more explanatory power than other approaches. There are two main approaches that have
dominated most of 20th century linguistics, namely functionalism and formalism.
Kortlandt defines functionalism as the view that “linguistic forms are instruments
used to convey meaningful elements.” Basically, functionalists look at language as a kind of tool
that we use to express ourselves and communicate with one another. For Kortlandt, the problem with
functionalism is that it doesn’t account for the harmful effects of language. Specifically, he
thinks language stunts our perceptual abilities. “Language invades the left hemisphere, diminishing
the perceptual capacity of the brain. As a result, man’s major capability to change his environment
is matched by a minor capability to gain insight from direct observation.” For Kortlandt, human
behavior like war, child labour, totalitarianism, and so on, really only make sense if we have
this combination of being driven by beliefs and ideas while at the same time being unable
to perceive the world as it is. If language is just a tool as the functionalists believe, then
you wouldn’t expect language to have this kind of harmful effect on us. But if language
is a parasite, then it might make sense. The other approach that dominated 20th-century
linguistics is called formalism. Kortlandt defines formalism as the view that “linguistic forms
are abstract structures which can be filled with meaningful elements.” Basically, formalists are
looking at language as a kind of abstract system that you can make sense of through formal logic
and mathematics. Kortlandt argues that language can’t be analysed that way. In his view,
linguistic meaning has the properties of a non-constructible set. There is no a priori way
to deduce the applications of meaning from its implications. The argument is probably a bit too
technical to parse in a YouTube video, so let’s just fall back on some of van Driem’s examples
to illustrate. You have a door that is closed but unlocked. I can say that the door is closed. But
I can also say that the door is open. You know, someone knocks and I reply “it’s open”. So you
have the same thing, a door that is closed but unlocked, and I can refer to it both by saying
it’s closed, and, it’s open. In both cases I’m telling the truth, but in conventional logic that
would be a contradiction. It can’t be both open and closed. So, in conventional logic to resolve
that contradiction, you’d probably have to argue that you’re not actually referring to the same
thing. Like, maybe when you say “it’s open”, you’re referring to the lock, and when you say
“it’s closed”, you’re referring to the wooden slab that makes up the door. Or something
like that anyway. Van Driem says: look, you don’t need those kinds of explanations.
It’s just a contradiction, and that’s fine, because linguistic meaning doesn’t follow
those rules of conventional logic. Instead, meaning is determined by its applications. What
that means is this, paraphrasing van Driem again: if I say “this is a house”, it’s a house because
I decided to call it one. If I want to call a cardboard box a house, it becomes a house. For
you the listener, when I say “this is a house”, what I’m really saying is “I want you to think
that I think that this is a house”. And whether or not that idea survives in your brain will have
nothing to do with truth and logic, but rather with Darwinian selection processes, the same
that drive the evolution of biological species. So there you have it. It’s an arresting
idea, right? Language as a parasite. Most of the accounts I’ve seen say that cultural and
linguistic evolution piggyback on biological evolution. But if language is a parasite, then
it collapses everything down into just one thing: biological evolution. There’s not really
any dividing line between evolution before language and after language. Not really any
dividing line between inheriting a physical trait from your parents or adopting a new word
from a friend. It’s all biological. To situate this idea in the broader context of linguistic
science, I should stress that this is very much a minority view. Most linguists, at least the
ones I know, would probably dismiss it. Well, to be honest most linguists have probably never
even heard of it. It’s not an idea that has all that much traction. But as a metaphor, I
think many would acknowledge that there are some interesting parallels between languages
and organisms, and even parasites specifically. It’s when you say you mean it literally that
people start to look at you funny. Some of that might be for good reason, but some of it might
have to do with the assumptions we’re making, which might not necessarily line up with
what the Leiden school is actually saying. Like, if you tell me something is a parasite,
a living organism, one of the assumptions I’m gonna make is that it reproduces through
genetic means. To me that’s an essential part of how we define what organisms are. If
the Leiden school was saying that the language parasite reproduces through genetic means,
then it’d be pretty bonkers, right? I mean, with a virus, we might talk about how it spreads
through physical contact or through the air, but we all know there’s actually something
tangible there: a virion that carries genetic information from one host to the other. And it’d
be a pretty wild idea to suggest that that’s how language works. Just take signed languages or even
reading. You could put a solid window pane between me and the person signing or me and a book,
and the language parasite would still spread, which shouldn’t be possible if it was something
like a virion. If that were the debate, it’d be pretty much an open and shut case
just because empirically it’s kinda bonkers. But that’s not what they’re saying. They’re saying
the parasite spreads entirely through non-genetic means. Patterns in sound and light waves are
enough for it to get from one host to another. And so then what you end up with is less bonkers but
also a lot more abstract. The question becomes: does something like that qualify as an organism?
That’s honestly not my wheel house. I know that the word “organism” is notoriously difficult
to define, and there are boundary cases like RNA viruses and whatnot, but I suspect most
biologists would think you’re stretching the definition of organism pretty damn far if
you start to include things like language. I won’t make too much of a point of it here, but
it’s worth remembering that that kind of question is also part of science, broadly speaking. From
the outside in, what we usually see of science is a picture that is focused mostly on empirical
findings. And empirical findings are of course very important. Empirical evidence is the linch
pin that holds everything together. But there are other parts to the process, like arguments
over theoretical consistency or concept validity, and these sometimes might get minimised a
bit in science communication. I guess this is a bit of a lead-in for future videos,
because it’s gonna come up a few times, both to illustrate how, when we as lay people
ignore those issues, we can sometimes get ourselves into trouble and think we’ve managed
to explain more than the science actually has, but also to illustrate how sometimes a scientific
disagreement isn’t necessarily about the evidence per se, but about how to interpret that
evidence in light of competing theories. Remember how we were talking about the metaphor of
language as a mental organ vs language as a virus? Salverda had this to say, and it encapsulates what
I’m trying to get at: “These metaphors (so the metaphor of language as an organ and the metaphor
of language as a virus) these metaphors are linked to an extensive framework of theoretical
assumptions, arguments and empirical evidence. As a consequence, we are faced here with a
complex whole of theoretical oppositions, which project very different perspectives on
the central problem of linguistic creativity.” See, when two conflicting theories collide,
sometimes what you get isn’t this sanitised image of science where all you need is evidence
to settle the debate. Sometimes it can get messy, bafflingly so. In one of the next videos I’m
gonna really nerd out on a famous example that illustrates just how messy it can get,
namely the Piraha controversy. So if you think you might be interested in that
kind of thing, feel free to stick around. I’ll end on more of a big-picture kind of point,
and that’s just that it can be pretty interesting to look at the broader implications this theory
has, the kind of world view we can arrive at if we assume language is actually a parasite.
In the Leiden school’s use of the term, “parasite” is neutral. It could be good or
bad or both or neither. It just refers to an organism that can’t survive without latching
on to another organism. But is language a net positive for us or is it a net negative? Or to
use the more colloquial definition of the terms, is language a parasite or is it a symbiote?
Kortlandt leans more towards the more pessimistic view. By stunting our perceptual abilities,
language might turn out to be our undoing. And if we want to find our way out from awful things
like child labor, totalitarianism, war and so on, then we have to start focusing less on linguistic
abstractions and more on direct observation, immediate experience. Van Driem on the
other hand thinks it’s probably more of a net positive. But he adds a twist: a net
positive for who? If language is a parasite, it’s tempting to locate our own identity within
the host, and treat language as something foreign, external to us. But language is so central to
the human experience, so crucial to who we are, that in a sense, we are the parasite. Or rather,
to be human is to be both the host and the parasite. So even if language was in some sense
detrimental to the host, that still wouldn’t be enough to say it’s a net negative for us, since
we’re not just the host, we’re also the parasite. Personally, when I stumbled on the Leiden school,
I thought it offered a nice answer to the Kekulé problem that the late Cormac McCarthy brought
up in his 2017 essay. The question is this: why do scientists sometimes report that they made
discoveries through dreams with visual imagery, symbols and metaphors? Kekulé claimed that
he discovered the molecular structure of benzene, which looks like a ring, when he fell
asleep and dreamt of a snake eating its own tail, the Ouroboros of mythology. And so what
McCarthy asks is: why couldn’t Kekulé’s mind have just said, using words, “Kekulé, it’s
a bloody ring”? Why did his unconscious mind use visual imagery when simple language could have
done the trick? McCarthy speculates that it’s probably because, evolutionarily, language
just hasn’t been around for all that long, so our unconscious mind tends to fall back on
the older tools it’s more comfortable with. But if you bring in the ideas of
the Leiden school, you could put a darker twist on it and imagine that there’s
tension between the host and the parasite. Kekulé’s mind used visual imagery instead of
language because the host doesn’t trust the parasite. It knows it has been invaded by
a foreign organism and is trying to fight back. Obviously that’s just wild speculation.
But, well, it’s fun to think about anyway. So, is language a parasite? I suppose that among
the many things language could turn out to be, a parasite might be one of them. In my book
it’s not one of the more likely candidates. In fact I find it very unlikely. But nonetheless,
it might be worth entertaining the idea, perhaps as a statement of fact as the Leiden school
does, but if not, then at least as a metaphor. Well that was fun. Fun to make anyway. If ever
you want to stick around, I have two videos coming up that are probably going to be shorter, more
pragmatic ones. One is going to be about value capture in language learning, looking at some of
the issues around metrics like Duolingo streaks or number of Anki cards and whatnot. The other one
is going to be looking at some practical things interpreters do that might be useful for language
learners. So those are going to be more practical topics. Then I have two longer ones where I’m
really going to nerd out on the linguistic side of things, one is going to be looking at the
Piraha controversy, and the other at the question “what is the world’s hardest language?”. Pretty
classic topics on this platform, but hopefully I’ll be able to add something new that hasn’t
been addressed in other videos. And then a last one about subtitles vs dubbing, so I’ll be
talking about stuff like the McGurk effect, how public policies affect preferences, and so on.
No idea when they’ll come out, nor in what order, but that’s what I’ve got in the works anyway.
Alright, thanks for watching and, toodles? OK bye.