Episode Transcript coming soon

Last time, I introduced the Mughal Emperor Akbar the Great, and talked about how he seems to have conducted an experiment similar to those allegedly conducted by the Pharaoh Psamtik, the Holy Roman Emperor Frederick II, and King James IV of Scotland. I ended with an intriguing question: where did Akbar get the idea from? Before the Renaissance, there was no known Latin translation of Herodotus's Histories, that the story of Psamtik's experiment comes from. There are, however, numerous instances of Medieval writers alluding to things Herodotus says, without necessarily knowing where they originally came from, presumably transmitted either through other authors referring to Herodotus, or through the Byzantine Empire, where scholars did continue to read Herodotus, just in Greek. This gives us a plausible account of where Frederick or Salimbene or whoever came up with Frederick's experiment got the idea from. In the case of James IV, Herodotus's Histories had been translated into Latin by that point, so it's not at all unlikely that James IV knew about the experiment. In the case of Akbar, though, it's a bit more mysterious. I find it unlikely that anyone at his court had read Salimbene's Chronicle, even less Pitscotties Historie and Chronicles of Scotland, which was only published three or four years before he did the experiment, so Herodotus is the most plausible source. Histories is the only one of the three texts that's ended up as a major work of world literature that, educated Europeans (although not Mughals) were at least at one point expected to be familiar with, rather than a moderately obscure Medieval chronicle that you'd only come across if you were, say, researching for a podcast episode about an event described in it. But even then, while many ancient Greek texts were translated into Arabic in the early Middle Ages, and became well-known in the Islamic world, Herodotus's Histories wasn't one of them. It seems to have been entirely unknown in the Islamic World before modern times.

What about the possibility that he heard that story from someone who'd read it in Greek? Akbar's enormous library certainly contained Greek texts, but it's not clear from what I've read exactly what Greek texts Akbar had access to, and as I understand it Ancient Greek was not a language that was well-known in the Mughal Empire.

From what I've read, it seems most likely that when contemporaries talk about 'Greek' texts, what they're actually referring to are Persian translations from Arabic of Greek texts translated to Arabic in the early Middle Ages. I'm also strongly getting the impression that Neoplatonic philosophy and medicine were the main areas of Greek thought that interested the Mughals, and given all this, then the Histories of Herodotus were almost certainly not among the texts in Akbar's library.

There were Jesuit missionaries hanging around Akbar's court in the 1570s, and it's likely they would have been familiar with Herodotus. So I guess there's always the possibility that one of them told him about the story of Psamtik. But the account of the missionary Jeronimo Xavier never brings that up, so while we can't rule it out, there's also no positive evidence for it.

Maybe it's just a trope that's been generally floating around since ancient times, and these are the only examples of it that have survived. In fact, in this miniseries, I've talked about a few other stories and thought experiments with similar themes, even if not exactly fitting this particular formula. There's the Ancient Greek thought experiment in the Dissoi Logoi, about how if a Greek baby is brought up in Persia they'll end up speaking Persian and vice versa. There's a story told in Roman-era Jewish texts about Abraham being hidden in a cave and coming out speaking Hebrew. There was that thought experiment that one of the Modists did, where if two children are raised without exposure to language they'll develop a language by assigning meaning to each other's babbling.

In the last few episodes of this miniseries, I've talked quite a lot about traditions in Christianity and Judaism of ascribing sacred status to some particular language, most often Hebrew. But what about other religions? Ancient North India is the source of three major religions still practiced today - Hinduism, Buddhism and Jainism. All of these had representatives at the Ibadat Khana, Akbar's venue for religious debate, so Akbar would have had at least a basic idea of what these religions involved, and would have had a chance to learn about them from experts. And all of them have ideas about the religious significance of particular languages.

The oldest Hindu scriptures, the collection of Hymns called the Vedas, are really quite remarkably old. The latest Vedas were completed by about 800 BCE at the very latest, and the oldest, the Rigveda, has parts that date back in the ballpark of 3400 years, near the end of India's Bronze Age, slightly older than estimates for the oldest parts of the Old Testament, and quite comfortably beating things like the Homeric Epics. It's about the same age as the oldest parts of the Avesta, the Zoroastrian scriptures. Before the Rosetta Stone was discovered during the Napoleonic Wars and enabled Egyptian Hieroglyphs to be deciphered, I think the Rigveda and the Avesta might actually have been the oldest texts that anyone in the world could understand.

The Vedas are written in a particularly ancient form of Sanskrit, a standardised literary register of the ancient North Indian dialects that ultimately gave rise to most of the modern languages of North India. Vedic Sanskrit is among the oldest Indo-European languages attested in writing, second only to the Hittite language of Ancient Anatolia. Sanskrit is also the subject of some of the very earliest known linguistics. As early as the middle of the 1st-Millennium BCE (the dates aren't known precisely), at about the same time as the Pre-Socratic Philosophers of Ancient Greece, or potentially even earlier, the Indian linguist Pāṇini produced a detailed and remarkably scientific study of Sanskrit grammar.

It shouldn't come as much of a surprise, then, that Hindus have traditionally seen Sanskrit as the original language. This naturally follows from the belief that the ancient Hindu scriptures, the Vedas, were dictated exactly by the Gods, and as such are unchanging and perfect. If the Vedas are perfect and divine, then the language in which they are written must also be.

Buddhism and Jainism, however, are based on more recent (although still very ancient) scriptures, written in Prakrit, a group of languages that developed from Sanskrit around the 5th century BCE. And this leads to different beliefs about the primordial language.

Mahavira, known for codifying Jain religious doctrine, lived in either the 6th or the 5th century BCE, and is known to have been a contemporary of the Buddha. Mahavira lived in roughly the same part of India as the Buddha, too, in the lands just south of the Himalayas that would be conquered during both of their lifetimes by the Magadha Kingdom under King Ajatashatru. Also like the Buddha, Mahavira spoke a form of Prakrit - in this case Ardhamagadhi Prakrit. This is the language of the oldest Jain scriptures, although more recent texts are in Sanskrit. And traditionally, Jains have believed that Ardhamagadhi Prakrit is the original language, from which all others, including Sanskrit and Pali, derive.

Different branches of Buddhism have different positions on the origins of language. Theravada Buddhism, the oldest form of Buddhism, is the first one I'll consider, and there's an interesting insight in a text called the Vibhanga Atthakatha. This is a commentary on the Theravada Buddhist scripture the Vibhanga, traditionally ascribed to the monk Buddhagosa, who lived in the Anuradhapura Kingdom of Sri Lanka in the 5th century CE. In it, the author mentions a thought experiment explaining how Magadhi Prakrit (an ancient North Indian language, traditionally said to be the first language of the Buddha, which in Theravada Buddhist tradition is held to be the same as Pali, a standardised form of Prakrit used in ancient Theravada Buddhist scriptures) was the language that a child would default to if they were never taught any other language.

"Parents place their children when young either on a cot or a chair, and speak different things, or perform different actions. Their words are thus distinctly fixed by their children (on their minds) thinking that such was said by him and such by the other; and in process of time they learn the entire language. The mother is a Damila, the father an Andhaka. Their newly born child, if it hears first the speech of the mother, it will speak the language of the Damilas. If it hears first the speech of the father, it will speak the language of the Andhakas. But if it doesn't hear the speech of either of them, it will speak the language of the Magadhas. Also someone who is born in a big jungle, devoid of villages, where no one else speaks, he too will by his own nature start to produce words and speak this same language of the Magadhas."

Damila is an ancient Pali word for Tamil (cognate with 'Tamil' and also borrowed into Sanskrit as Dravida, hence the South-Indian language family including Tamil is called 'Dravidian'). Andhaka refers to an ancestor of modern Telugu, also a Dravidian language, and is the origin of the name of the Indian state of Andhra Pradesh:

Then the source begins to sound very familiar indeed (or at least it should, if you were paying attention in the last five episodes):

"In hell, among the animals, in the realm of ghosts, in the world of men and in the world of gods, everywhere this same language of the Magadhas is preponderant. The remaining eighteen languages - Otta, Kirata, Andhaka, Yonaka, Damila, etc. - undergo change in these [realms]. Only this language of the Magadhas, rightly called language of Brahma and Aryan language, does not change. Even Buddha who rendered his tepitaka words into doctrines, did so by means of the very Magadhi; and why? Because by doing so it was easy to acquire their true significations."

This is specifically a Theravada Buddhist view though, and there are other branches of Buddhism out there. If the Buddhists in the Ibadat Khana came from Tibet, they would mostly have been from the Vajrayana branch of Mahayana Buddhism, now the largest Buddhist sect, and the principal sect North of the Himalayas. Their scriptures are mostly written in a form of Sanskrit sometimes called Buddhist Hybrid Sanskrit, a bit different from the Sanskrit used in Hindu scriptures. Mahayana Buddhism officially doesn't make any claim that their version of Sanskrit is the original language, although there's evidence that at times some Mahayana Buddhists have actually claimed that. The Buddhists at Akbar's court would certainly have been much less likely than Theravada Buddhists, and less likely than Hindus or Jains, to have a strong position on what the original language was. Although they would have likely known about these ideas.

The similarity between the thought experiment from the Vibhanga Atthakatha and other ideas I've talked about in this miniseries, particularly in the first episode, might not be a coincidence. Ancient India and Greece were hardly hermetically sealed off from each other. Trade in exotic goods from India to the Mediterranean went overland through Persia and by sea across the Arabian Sea to Egypt. Archaeologists have found remains of Buddhist and Hindu shrines in Egypt dating back to the Ptolemaic period. When Alexander the Great marched his army as far as India, he visited the Buddhist intellectual centre of Taxila, near present-day Rawalpindi in Northern Pakistan, and Greek scholars accompanying him met their Buddhist counterparts. And after Alexander went back westwards, he left behind a whole Indo-Greek Kingdom in Bactria to the North of Taxila, with a remarkable hybrid culture. When the Mauryan Emperor Ashoka, having already conquered most of the Indian Subcontinent, including Taxila, announced his dramatic conversion to Buddhism, and his renunciation of aggressive war, he made the announcement on a series of pillars, erected throughout his empire, in multiple languages, one of which was Greek. And later, the Gandharan Kingdom would be conquered by the Indo-Greek Kingdom from Bactria, and would become a centre of a heavily Greek-influenced style of Buddhist art. That section of the Vibhanga Atthakatha that I quoted listed a series of 5 languages - one of them was Yonaka, and that's Greek (Ionian, if you're interested in the etymology, like the Turkish word for Greece is Yunanistan). And although Herodotus (likely another contemporary of the Buddha and Mahavira) was a Greek, he lived in Halicarnassus, now Bodrum on the Aegean coast of Turkey, which was at the time ruled by the Achaemenid Persian Empire. The eastern edge of the Persian Empire directly abutted India, and the Achaemenid Empire, for most of its existence counted both Greeks and Indians among its subjects.

The point of all this is that, throughout antiquity, the Eastern Mediterranean and India were much more closely connected, culturally, politically and economically, than people normally think of them as being, so it shouldn't come as that much of a surprise that similar questions about the origin of language were being asked, and were being approached through similar methods, in Ancient Greece and Ancient India.

While he certainly enjoyed hanging out with followers of other religions, Akbar the Great himself was a Muslim, at least at first, and much of his intellectual environment was heavily influenced by traditions of Islamic scholarship, where, since at least the 8th Century, the origins of language were a significant topic of debate.

In the 9th Century, in the remarkable intellectual environment of Baghdad under the Abbasid Caliph Al-Ma'mun, where Islamic scholars were learning about ancient Greek philosophy through translations into Arabic that the Caliph was commissioning, this debate is usually characterised as a debate between two schools of thought. These are Istilah, or the Conventionalist position, and Tawqif, or the Revelationist position.

Istilah is broadly similar to the position of Hermogenes in Cratylus - the dialogue by Plato that you should be familiar with by now, where Hermogenes and Cratylus discuss whether words have some intrinsic connection to their meanings, or whether words are assigned arbitrarily by collective agreement. The Istilah position is essentially the latter - the meanings of words arise as a matter of social convention. It's also the position of Aristotle, probably the most influential Ancient Philosopher on Medieval thought in both Christian and Muslim worlds.

The other position, Tawqif, is that meanings were assigned to words by God, when he gave Adam the power of language. After all, the Quran says

"He taught Adam the names of all things, then He presented them to the angels and said, 'Tell Me the names of these, if what you say is true?'"

It's interesting to note that both the Tawqif and Istilah positions hold that the meanings of words are arbitrary - you can't try and break words up into their components to reveal things about the things they denote, like Socrates does in Cratylus. You either end up coming back to "this word means what it means because God decided that's what it means" or to "this word means what it means because people collectively decided that's what it means".

The Istilah school is particularly associated with the Mu'tazilites, a school of thought originating in the 8th century and championed by the Abbasid Caliph Al Ma'mun. The Mu'tazilites believed that the Quran is the word of God, created by God, and not eternal. They placed great emphasis on the role of reason in understanding the world, and religion, and interpreting scripture. They believed that, where it seems to be logically incoherent, the Quran must be interpreted metaphorically. For example, it's a logical impossibility for God to have physical attributes, so when the Quran describes God as having, for instance, hands, that must be a metaphorical way of explaining the power of God to act, rather than meaning literal, physical hands. For the most part the revelationist position was associated with the Ash'arites, the opposing faction to the Mu'tazilites, who supported a literal interpretation of the Quran.

These linguistic positions reflect views on the nature of the Quran relative to God. The Ash'arites believed that the Quran is co-eternal with God and is uncreated - the word of God is more like an attribute of God than something God created. That means that there is some sense in which language itself is connected to God. For a Mu'tazilite, on the other hand, nothing, including language can be co-eternal with God. Language must be a created thing.

The Istilah-Tawqif and Mu'tazilite-Ash'arite divides didn't line up perfectly though, and there were a few Revelationist Mu'tazilites, such as Al-Farisi in the 10th century. This makes sense when you consider that Conventionalism doesn't logically follow from Mu'tazilitism, even if it has similar vibes. But they are broadly correlated.

And then you get nuanced positions like that of Ibn Aqil in the late 11th and early 12th centuries. He describes a third position that's a sort of compromise between Tawqif and Istilah. God gave people arbitrary words for some things, but then humans used reason to create words for things they didn't have words for by associating them with things they did have words for.

I can see why this might appeal to people. Lots of words are obviously derived from other words that have undergone semantic drifts, and there are plenty more whose metaphorical origins are less transparent. There's a continual process of metaphors being created, and then being used so much that we stop thinking of them as metaphors and they just become regular words, meaning we need to create new metaphors, and the same process happens again. But where the first words originally came from, the first basic nouns and verbs, that we could then use metaphorically to talk about abstract concepts, is a much harder problem.

Another compromise position is that Fakhir al-Din Al-Razi, in the 12th century, who argued that the meaning of the words in the Quran is distinct from the words used to express that meaning, and the former is an eternal attribute of God while the latter is a created thing, subject to time and space and change and all that.

There was also a Naturalist school, that more closely resembled the view of Cratylus from Cratylus. These argued that the meanings of words were not arbitrary, and words had particular connections to their meanings. This was said to have started with Khalil ibn Ahmad in the 8th century, and we end up with people saying things like that qadama, meaning 'to eat hard things' is different from khadama, meaning 'to eat soft things' because the kh sound in khadama is a 'lighter' sound than the q sound in qadama.

After the twelfth century, as far as I can tell the debate on the question of whether the Quran was created or eternal kind of died out without ever reaching a conclusion, and with it the debate on whether language was revealed or conventional.

Getting closer to the actual point, we should note that the Quran doesn't explicitly say what language Adam spoke, which leaves some space for controversy among Muslims. The Quran says "God taught Adam the names of all things", but doesn't say in what language. The most common Muslim opinion seems to be that Adam spoke Arabic, the language in which God dictated the Quran to Muhammad. This was also the most common opinion historically too, and Medieval Islamic linguists seem to have mostly taken it for granted in practice that Arabic grammar is divinely-inspired. There's also a minority opinion that holds that Adam spoke Syriac, or Aramaic, languages of the Ancient Near East at the time of the Old Testament - and contact with Syriac-speaking Christians had a huge role in the development of early Islam. Another opinion is that God taught Adam all languages, which I guess is another very literal interpretation of "God taught Adam the names of all things".

I haven't seen much information on historical Muslim positions on the specific question of what language children would speak innately. The main modern Muslim position on the innateness of language seems to be that the capacity for language is innate, but that there is not any particular language that is innate. I was actually kind of surprised that when I Googled 'do babies innately speak Arabic', none of the results were particularly relevant. I was bluntly corrected by Google's AI Overview, telling me

"Babies do not innately speak Arabic or any specific language."

and then the results were all sensible-looking articles about teaching children Arabic, a few about the aforementioned Islamic perspective on language acquisition, studies of infant language acquisition in Arabic-speaking populations, and that sort of thing. I was expecting at the very least a post in an obscure forum from someone who seemed to think that was a plausible idea for some reason, but there's absolutely nothing.

So overall, it seems like in terms of Akbar's religious enquiries, the question of 'what language would children speak if left to their own devices' could have come from a couple of directions. That exact question had been raised in Buddhist scholarship over a millennium earlier (and answered with 'Magadhi Prakrit, the first language of the Buddha', although the Tibetan Buddhists at Akbar's court would have been less likely to have a definitive answer). Hindus would have mostly said Sanskrit was the original language, and Muslim views would have been complicated, mostly favouring either Arabic or Syriac as the original language - although in this case the exact question of 'the default language' isn''t really one that's been addressed much in historical Muslim scholarship. And then you've got the Jesuits. I talked about some Christian views on the original language that were thriving at about this time in the episode before last. Guillaume Postel, the somewhat eccentric scholar of eastern languages, who believed like most of his contemporaries that all languages descend from Hebrew, lived not long before this and had actually briefly been a Jesuit (before he got thrown out for claiming that the Age of the Holy Spirit was imminent and would involve the entire world being ruled by the King of France). So the Jesuits would have brought their own ideas about this question as well.

And what you notice from the sources is that that language, nation, and even religion seem to be treated somewhat interchangeably. Xavier says that Akbar said he was trying to find out what language the children would speak so he could follow the laws of that country, and Badayuni actually says Akbar was trying to find out what religion the children would follow. This makes some sense if you consider that he was talking to representatives from all these different religions, with different positions on what the original language was. You've got the Hindus saying Sanskrit, the Christians saying Hebrew, if you remember the last episode Akbar's Jewish subjects, despite being a very small and not at all well-documented community, still had representation at the Ibadat Khana, and they were probably also saying Hebrew, the Muslims saying Arabic, or maybe Syriac, the Jains saying Prakrit, the Buddhists might have said Sanskrit or Prakrit or might not have picked a side. It's easy to see how Akbar might have got the impression that finding out what the default language was was a pressing issue for his lifelong project of ascertaining the one true religion.

Another issue is where Akbar got the idea of doing an experiment from. The paper I've been using as a starting point for a lot of the research in this series (Royal Investigations of the Origin of Language by Robin Campbell and Robert Grieve, in 1981) claims that Akbar's experiment is anomalous because there's little other evidence Akbar the Great (or indeed Mughal civilisation in general) was particularly interested in empirical investigation as a means of finding things out (although they accept that it clearly actually did happen). For starters, I think this experiment itself, and particularly the fact that neither Abul Fazl nor Badayuni seems to regard its empirical nature as particularly novel, demonstrates that Mughal culture at the time of Akbar had a decent idea of what an experiment was. And even if it didn't happen, the fact that contemporary Mughal writers described a scientific experiment, with a fairly sensible setup (even if both the methodology and the question it was asking were insane), whether it was a real experiment or a fictional one, shows that people in Mughal India did in fact understand the concept, and thought it was a reasonable way of answering questions.

Gaining knowledge from empirical observation was certainly understood in Medieval Islamic thought, and that was a system of scholarship that the Mughals would have been broadly aware of.

And then we get Akbar's son and successor, the Mughal Emperor Jahangir. He's well-known for his interest in empirical observation, particularly in zoology. His autobiography, the Jahangirnama, includes detailed descriptions of his observations of various animals. For example, he says of the Urial, the wild sheep of Central Asia and the Karakoram Mountains, that

"I have heard repeatedly from huntsmen and avid hunters that at a certain time a worm develops in a mountain sheep's horn, and from the worm's movement an itching pain develops that makes the sheep fight its own kind. If it cannot locate an adversary of its own kind, it knocks its horn against a tree or rock to relieve the itch. After a search was made the worm was discovered in the horn of a female of the species. Because the female does not fight, it became apparent that the tale had no basis in fact.

It reminds me a bit of Frederick II debunking the Medieval European myth about barnacle geese forming from barnacles.

Then there's another sheep-related incident where he visited a village with a nearby pool that the villagers refused to go into because they'd been told that there were what the translation I'm using calls 'alligators' there (alligators don't live in India, although crocodiles most certainly do). Jahangir, to prove this was nonsense, had a live sheep thrown in (which swam across unharmed), and then got his servant to swim across as well (the servant was also fine). And another example, also involving sheep, had Jahangir comparing the air quality of Mahmudabad and Ahmedabad, two cities in Gujarat. In this case the emperor hung up a sheep carcass in each city, and in Ahmadabad it was rotten after eight hours, while the carcass in Mahmudabad's much healthier air apparently took 14 hours to reach the same condition.

So overall I don't think it's particularly implausible that Akbar might have decided to empirically investigate something - people around him at the time clearly didn't think it was implausible, and his son is known to have done a lot of empirical investigation.

And after Akbar's death, the story was repeated, and repeatedly embellished. An account from the Dabistan e Mahazdeb, a work of comparative religion, written in 1643, gives a similar basic outline to the other accounts, except having the children imprisoned until they were fully fourteen years old, and saying that the fact they didn't learn to speak proved

"that letters and language are not natural to man, that is, cannot be used unless they have been acquired by instruction, and it is then only that the use of conversation becomes possible. From this the conclusion was drawn that the world is very ancient and language of long date."

Adding further confusion to the problem of what exactly the experiment was supposed to show.

Then we get the Venetian traveller Niccolao Manucci, who travelled to India on an English ship as a teenager, and had some basic training in Persian and medicine from the Jesuits before he somehow managed, by the time he turned 15 in 1653, to persuade the Mughal court to employ him as a Physician. His experience of the Mughals came well after Akbar's reign, during the reigns of his grandson Shah Jahan, who built the Red Fort and the Taj Mahal, and his great-grandson the religious fanatic Aurangzeb, but the story of Akbar's experiment still seemed to be going strong. Manucci gives a more detailed set of hypotheses, that possibly reflects European concerns about the origin of language at the time.

"This was done, because one set of men asserted that they would speak the natural language, that which was the language of our first parents. Others held that they would speak the Hebrew language; others that they would not speak anything but Chaldean; while the Hindu philosophers and mathematicians asserted that they must infallibly speak the Sanskrit language, which is their Latin."

And states explicitly what the other accounts just left us to guess:

"Each one put questions to the children, and they answered just nothing at all. On the contrary, they were timid, frightened, and fearful, and such they continued to be for the rest of their lives."

And then the French Jesuit priest Francois Catrou, who never went anywhere near India, but had read Manucci's History of the Mughals, paraphrased Manucci in 1705, but also added further details. He said that Akbar had heard that the babies would speak Hebrew, and that he'd found a Jewish interpreter who could tell him whether they actually were speaking Hebrew.

And then Catrou adds that

"They had learned, from their nurse, to talk to each other. They only expressed their thoughts by actions that took the place of words for them."

While there's no particular reason to believe this happened - it first appears in an account written 137 years after the experiment by someone who had no connection to the Mughal court - as a fictional story, it's absolutely lovely. The efforts of a deranged tyrant thwarted by some of the most marginalised people in society, subverting the orders of one of the most powerful men in the entire world to create a human connection to these children in an environment that he had designed with the specific intention of depriving them of exactly that.

I guess it's not particularly implausible that this might have happened. The mute nurses would have their own signing systems, and if they were working together for three years it's likely they would have worked out a common lingua franca to communicate with each other, like how sign languages have always developed in communities of deaf or mute people. Deaf children and hearing children alike, with deaf parents using a sign language generally acquire that sign language as their first language, and sign language acquisition seems to work in pretty much the same way as spoken language acquisition. And if the children emerged from the Gang Mahal signing in this peculiar Gang Mahal Sign Language that had developed among them and the nurses, Akbar and his courtiers would not have acknowledged that as language, because sign languages generally weren't regarded as such at the time. Even Catrou doesn't refer to it as language.

We're getting near the end of this series now. But there are still a few things to look at. First, this series has looked at four experiments, and they have quite a lot in common. This is a sort of genre of story, defined by a few common elements:

One obvious commonality between these stories is that the experiments are all carried out by monarchs. I guess monarchs are usually the only people who have that sort of power over other people, to just get hold of some children and lock them up somewhere without anyone being able to stop them. At least in the case of Frederick, the story is being clearly used to illustrate his tyranny.

Another is that the three of the kings we know about all had a problematic relationship with mainstream religion, especially Frederick and Akbar. We don't really know enough about Psamtik to say anything either way. James IV's religious views don't seem to be anywhere near as unusual, and by all accounts he seems to have been a reasonably pious Catholic, even if he did get excommunicated for going to war against an alliance including the Pope towards the end of his reign. On the other hand, James IV, like Frederick, did keep company with alchemists, and we can see in William Dunbar's poem about John Damian that this raised some eyebrows. Even his interest in medicine would have been suspicious to some people. In the Late Middle Ages and the Early Modern Period, the medical profession was often associated with Atheism. There's a certain logic to it. A doctor has to treat the body like a naturalistic physical system, and act on the assumption that God isn't going to step in and cure a patient miraculously - that's kind of the whole point. And it's easy to imagine, like lots of people did, that there might be a slippery slope from there to just straight-up not believing in God at all.

And that brings us nicely to another closely related similarity. James and Frederick at least were interested in science to an extent that was unusual at the time. Even Psamtik might have been, if we take Herodotus's account of his measurements of the depth of the Nile as having any connection at all with reality. With Frederick especially, we can clearly see that the rumour that he performed a language deprivation experiment was being used as part of an attempt to associate his scientific interests with impiety and deranged brutality. In the experiment with the guy in the barrel especially, Salimbene is switching effortlessly between talking about the brutal methodology and the fact that the hypothesis - that there is no immortal soul - was blasphemous and heretical. Salimbene seems to have regarded the two as pretty much interchangeable. Obviously someone who believes that there's no immortal soul is going to have no problem with engaging in the sort of twisted behaviour that Salimbene accuses Frederick of.

While these experiments could obviously never be replicated, it's often suggested that we can get some insights from real-life incidents where something kind of similar to these experiments has happened.

Throughout history, there have been repeated incidents in which children have been found, having seemingly grown up in isolation from the rest of humanity. As an example, this is one of the most well-documented and definitely real such incidents.

In January 1800, in the village of Saint-Sernin sur Rance, in the rugged and mountainous region Aveyron in Occitania in Southern France, a boy wandered out of the woods. He was about 12 years old, naked despite the cold winter, and seemingly completely unable to speak. He was looked after by the local Abbot and biologist Pierre Joseph Bonaterre, who eventually wrote a book about him. Six months later, he was taken 400 miles north to Paris, and studied by several psychologists. Eventually he was taken in by one of them, by the name of Jean-Marc Itard.

Itard, a Physician at a school for deaf children, set about trying to teach the boy, who he called Victor, to speak. Noticing that he seemed particularly interested in the sound o, he would repeatedly say the word eau - 'water' - when he gave the boy water. This didn't work, but for whatever reason a similar approach succeeded at getting Victor to say the word lait - 'milk'. This was, however, as far as he got. Victor was never successfully taught any more of the French language.

The problem with this, and with pretty much all the other feral children who are sometimes used as examples of real-life language deprivation, is that I don't think it's even particularly plausible that Victor had never been exposed to language. The circumstances in which he was abandoned are lost to history, but he couldn't have survived on his own as a literal newborn baby, so he must have had some time within human society during which he was presumably exposed to language. I reckon it's likely he was just autistic and non-verbal. It's also noteworthy that Itard, once he'd figured out that there was no physiological reason Victor shouldn't be able to speak, was fixated entirely on teaching him spoken language. He never attempted to teach Victor any form of sign language. I suspect he might have had more success in that regard.

Other potential datapoints come from several cases of children kept virtually isolated from exposure to language, not as deliberate linguistic experiments, but simply through extreme abuse and neglect, sometimes in conditions not too different to those that would have been produced in these deliberate language deprivation experiments (assuming any of them actually occurred).

Probably the most famous is a girl known only by the pseudonym Genie. She was found in 1970 in Los Angeles, at the age of 13, in a small room where her father had kept her constrained in unimaginably appalling conditions since she was one year old. And for those twelve years, it's likely she never heard any meaningful amount of language. She certainly hadn't learned to speak.

This was at a time when the idea of a critical period of language acquisition - a phase of early childhood in which language can be acquired from scratch, and after which it can't be - was fairly new, and linguists were excited about the idea of being able to test it - but what they found with Genie was pretty inconclusive. At first, it looked like Genie wouldn't be able to learn to speak, confirming the theory. But gradually, over the next few years, with the help of a team of therapists, and the linguist Susan Curtiss, she was able to learn to use language to at least a rudimentary level. She learned in quite a different way to how infants normally acquire language - she relatively quickly developed a large vocabulary, learning the names for objects and actions, but had much more difficulty actually assembling those words into coherent sentences, and with the actual physical act of making sounds.

There are also quite a few factors that complicate using cases like that of Genie as if they were real-life, modern versions of the experiments these episodes talked about, that mean these cases aren't totally comparable.

To start with, Genie, and a few other victims of similar crimes, were kept in conditions worse than those that were likely experienced in the experiments conducted by anyone except maybe Frederick. Frederick's supposed experiment is the only one where the children are explicitly mentioned as being deprived not only of language but also of all forms on non-verbal communication. The children in these experiments seem to have been released earlier - Herodotus specified that Psamtik kept the children confined for just 2 years, and Akbar's experiment lasted for 3 or 4 years. The descriptions of James and Frederick's experiments don't specify exact amounts of time, but it's unlikely that (if the experiments took place at all) the children would have remained in captivity beyond maybe about 5 or something, when they would normally be expected to have acquired a first language.

Also, the reported experiments don't seem to have involved keeping the children locked in a single room, or anything like that. The conditions aren't completely clear, but James seems to have given his experimental subjects an entire island, and Akbar gave them a building. These children weren't spending years on end tied up in a dark room. In a sense this would make these experiments more useful than the real examples. They would have shown something closer to how children would develop without exposure to language under something as close as possible to otherwise normal circumstances. Genie was generally so psychologically disturbed by what had been done to her that it was impossible to tell what aspects of her difficulty in learning language were specifically a result of language deprivation, rather than psychological or even physical consequences of the abuse she had experienced. For instance there was some evidence that she had experienced some kind of damage to her left brain hemisphere, that normally controls tasks to do with grammar.

And on top of that, like Victor, Genie was not deprived of language literally from birth. For the first year of her life, she was exposed to human society in a relatively normal way. So again this isn't really a case of total language deprivation like in the experiments in this series.

One of the most important differences is that in these cases the children were completely alone, while all the experiments described involve more than one child. There seems to have been an understanding that the process of language acquisition is inherently communicative. You might remember Siger of Brabant, who I talked about in the second episode, who argued that two babies isolated from the rest of the world would invent a language by assigning meanings to each other's babbling. This seems like the only plausible way any kind of innate linguistic framework, if it exists, would actually lead to the development of actual language in a total language deprivation scenario. Nowadays even the most hardcore universal grammar believers don't think there's some fully-formed language just sitting latent in the minds of babies, that they'd end up speaking if it wasn't overridden by the languages they hear around them. And in fact you can see that babies do seem to like making random noises at each other as if they're sort of practicing having a conversation for when they know about words.

Probably the real-world situations that actually come closest to the scenarios described in these last few episodes, have kind of been sitting there in plain sight this entire time, and aren't anywhere near as rare as feral children. I'm talking about deaf children raised by hearing parents who don't use a sign language, and as such who might go a long time before experiencing language. Basically, the research that's been done in these situations shows that the longer it takes before children are exposed to language the harder it is for them to then acquire native-level fluency, and the more likely they are to acquire permanent language disorders. Which I don't think is really a particularly surprising result. But what's really interesting in these situations is that it's common for children to develop their own homesign systems, that aren't necessarily as sophisticated as fully-developed languages, that develop among larger speech communities, but go far beyond mere gesturing, where, for example, signs can be put together into sentences, and signs even develop that are conventional, rather than iconic - that is to say, they don't have any obvious resemblance to the thing they denote.

Of course, these systems develop through being used to communicate with people who have learned language, and who are presumably able to use their own existing understanding of the concept of language to understand the signs and who also can be seen using gestures when they speak. So you could argue that even this isn't the kind of pure language deprivation that the monarchs Psamtik, Frederick, James and Akbar intended to create. Maybe even this input is enough to wipe out a child's innate propensity to speak Phrygian.

I think a lot of the attraction of language deprivation experiment stories is that it's often regarded as one of those experiments that would convincingly settle a critical linguistic debate, but would be so wildly unethical that it could never actually be conducted.

The thing is though, that examples like Victor and Genie show how unlikely it is that a depraved tyrant could even find out anything useful by doing one of these experiments. The traumatic conditions they'd have to inflict in order to ensure anything approaching total language deprivation would preclude any chance of normal development apart from language, so the proof that language is entirely learned would never be conclusive. Despite cases like those of Genie and Victor of Aveyron, there is still little in the way of a firm consensus on the existence or not of a critical period of language acquisition, or anything like that. They didn't actually conclusively settle anything.

There are really two questions that the accounts of these experiments, and the work of other scholars I've talked about in this series, throughout history, have a tendency to conflate - and they aren't the same, but they are closely related. There's not just the question of how children acquire language, but also the question of how humans originally came to acquire language.

In the past, at least, the second question seems to have been taken as a far more important one. The experiments discussed in this series seem to be focussed on the first question - how children acquire language. But they're mostly using it as a proxy for the second question (Abul Fazl's account of Akbar's experiment, where he actually is concerned with whether Language is innate, is a notable exception). There's an assumption in most of them that the language the children would speak is in some sense the first, or at least the best, language. And after the time of Akbar, that question, of what the original language is, persisted as one that intellectuals attempted to answer.

In the episode before last, I talked about how some people in the 16th century made strange claims about what modern language is closest to the Adamic language. I mentioned Johannes Goropius Becanus, who advanced a series of spurious etymologies to claim this honour for his native Antwerp dialect of Dutch. William Camden, in the early 17th century, made a similar claim for Welsh, and there's also, a bit later in the 17th century, Wojciech Dębołęcki, who claimed that Adam spoke nothing else but an early form of his own native language of Polish.

There were also people who believed in a language spoken by God that's nothing to do with any current human language (and who often claimed to have had it revealed to them). In Episode 2 I talked about the 12th-century German Abbess Hildegard of Bingen, who wrote about a Lingua Ignota, or 'Unknown Language', that had been revealed to her by God, and getting up to Elizabethan and England, you get the alchemists, astrologers and general sorcerers John Dee and Edward Kelley, who both wrote in a constructed language that they called Angelical or Enochian, and that they claimed to have been taught by Angels.

Various religious movements since have made similar claims to these Elizabethan wizards. Obviously this includes the Mormons in the 19th century, who created their own version of the Heber myth. God let Jared keep speaking the mysterious sacred language of Adam at the tower of Babel. Jared's descendants migrated to America (it's apparently controversial whether they crossed the Atlantic or Pacific Ocean), and then God destroyed their civilisation. Joseph Smith, the 19th Century founder of Mormonism, claimed that this sacred language was partly revealed to him. Another example is Medefaidrin, that emerged in a church in South-Eastern Nigeria in the 1930s. Its creators claimed it was revealed by the Holy Spirit. Apparently the Church is still using it.

There have also been plenty of ideas that, at least to a person immersed in a modern empirical world, sound much more reasonable.

As early as the late 16th Century, Joseph Justus Scaliger rejected the spurious etymological arguments popular at his time, and the claim that Hebrew or any particular language is the original language. He went as far as to group the languages of Europe into 11 groups (Latin, Germanic, Slavic, Greek, Albanian, Hungarian, Turkic, Finnish, Basque, Irish, Brittonic), and to say that these groups represent multiple distinct origins of language and are not related to each other at all.

And then by the second half of the 17th century, in England, guys like Robert Boyle and John Locke, profoundly dull and unimaginative, soyfacing over lame nerd shit like 'empirical evidence', started claiming that Biblical Hebrew didn't actually seem to give any particularly profound insight into the nature of things, and in fact seemed just as arbitrary as any other language. The German polymath Gottfried von Leibniz had a similar view, and rejected both the traditional view that all languages descend from Hebrew and Scaliger's view that there are several completely unrelated families. Instead, he suggested that the original language was a now-extinct language that you theoretically ought to be able to learn things about by comparing words from languages that we know.

As we got to more modern times, and people became more thoroughly aware of all of the world's languages, the idea that at one point everyone spoke Hebrew became less tenable. From the 16th Century onwards, European contact with India increased, first missionaries like Jeronimo Xavier, and a few small trading operations, and then larger-scale commercial enterprises, culminating in the British East India Company's outright conquest of India in the late 18th Century. And many Europeans visiting India, and learning Indian languages, had come to notice that the languages of Europe and India really have quite a lot in common. A few people suggested they shared a common ancestor, and in 1786 William Jones, a British judge working for the East India Company in Calcutta, brought the idea firmly into the mainstream.

In the early 19th century, linguists started to try and work out the relationships between these languages more systematically than they had before. I talked a bit in Episode 3 about Grimm's Law, identified by Rasmus Rask and then by Jakob Grimm (probably better known for the folktales him and his brother collected), showing how various sounds in Germanic languages consistently correspond with other sounds in other Indo-European languages (for example Proto-Indo-European p corresponds to Germanic f). If you can work out rules like these, then you should be able to have a go at actually reconstructing the long-dead common ancestors of groups of modern languages. In the 1870s, August Schleicher made a start on reconstructing Proto-Indo-European, that common ancestor of Greek, Latin, German and Sanskrit, hypothesised by William Jones among others, and Schleicher even wrote a story in reconstructed Proto-Indo-European: *Avis akvāsas ka - 'the sheep and the horses' - versions using more up-to-date reconstructions of Proto-Indo-European call it something like *H₂óu̯is h₁éḱu̯ōs-kʷe, which should be recognisable if you know any Latin. Indo-European languages are the most studied and were the first to be studied like this, but you can do similar things with other language families.

But you can only go back so far like this, before the relationships between languages become too distant to notice, and the earliest times we can reach through even the most ambitious feats of comparative linguistics are a fraction of the time for which humans seem to have been using fully-formed languages, as sophisticated as any spoken today. If you wanted to go before that, and particularly if you wanted to look at when and how people first started using language, all you had was pretty much just speculation.

Epicurean ideas about the origin of language that I mentioned in the first episode, where language emerged purely as a matter of convention, became popular in the 16th and 17th centuries. In the Renaissance, Humanists had started to get into Roman Epicurean Lucretius - whose materialistic worldview was usually regarded as blatantly heretical (In Part 2 I mentioned the moral panic in Medieval Europe about a supposed Atheistic sect of 'Epicureans'). Humanists, though, would say they were just reading to appreciate the quality of his Latin verse. In the 15th century, people started suggesting maybe there were some things that could be learned from Lucretius's ethics, and by the 17th century, people were starting to seriously consider his approach to how the physical world was organised, including his approach to the origin of language. Pierre Gassendi, a Catholic priest in the first half of the 17th century, interested in the project of reconciling a naturalistic worldview with Christianity, gave the Epicurean account of the origin of language as an example of how a phenomenon could be explained in naturalistic terms.

Into the 18th century, these more secular and naturalistic accounts began to dominate. Jean-Jacques Rousseau and E.B. de Condillac, both in the middle of the 18th century, both reckoned language had originated through a mixture of gestures imitating things, and fairly random noises. The noises became associated with the gestures, and there was a period when noises and gestures each carried different parts of meaning, but since noises were more efficient than gestures for communicating, the gestures were dropped, and the noises took on the whole meaning. Later in the century Johann Gottfried Herder would attribute the origins of words to people imitating the sounds of things and turning them into the words for those things. And this sort of back and forth of guesswork would continue for the next few decades.

Summarising the state that this accumulation of guesses was in in 1861, the linguist Max Müller listed five basic categories into which theories of the origins of language could be grouped. There was the idea put forward by Herder, that early humans created the first languages from imitations of animal noises. Müller called this the Bow-Wow Theory, and it's supported by the fact that a lot of words obviously do come from onomatopoeias (I particularly like the Mandarin māo, the Thai /mɛːw˧/, or the Ancient Egyptian /mi(ʀ)juw/, all meaning 'cat'). The problem is that most words don't have obvious onomatopoeic origins, and it's difficult to explain the origins of words for things that don't make a noise. Then there was the idea, which ultimately goes all the way back to the Presocratic Atomist Democritus, that language developed from emotional exclamations - the Pooh-Pooh Theory (this generally seems like an unusual way for words to develop. 'Yucky' and 'yummy', maybe 'ick'). The next one is one that I find significantly weirder and at the same time very interesting - that is, the idea that things have a sort of mysterious inherent association with particular sorts of sound - even if they don't actually make a noise. Müller called this one the Ding-Dong Theory, because it was explained in quite esoteric terms involving things having a mysterious sort of resonance that humans picked up on. And then there's the idea that the first language began with humans using rhythmic sounds to coordinate manual labour. Muller called this one the Yo-He-Ho Theory.

At this point these were all just guesses, and there was no way of actually checking empirically which, if any, of these ideas was correct. In 1866 the Linguistic Society of Paris decided all this unverifiable speculation was a waste of time and totally unscientific, and banned all discussion of the origins of language from their meetings. The London Philological Society followed in 1872, and it became quite an unfashionable topic for the next few decades.

By the second half of the 20th century, science was in a much better position to start making progress again on the question of the origin of language. For example, we had a much better understanding of the general chronology of human evolution. Tools like Carbon Dating now meant you could get a pretty good idea of how old a lot of stuff was. On the anatomical side, people had a much better understanding of how the brain works, which helps to make educated guesses about the linguistic capabilities of early humans from their skulls. We also have various ways to observe the brain activity of currently living modern humans, and learn things about the cognitive processes involved in language that way. This sort of thing has allowed some new progress to be made on the old theories. For example, the yo-he-ho theory is now generally considered implausible because humans seem to have been using language long before there's any evidence of them doing things that would have required significant rhythmic coordination of manual labour.

Another example is the fact that there are experiments that have given some support to the Ding-Dong Theory (if you ignore the bit about the resonances). Young children, across all sorts of languages, when shown a pointy shape and a blobby shape and asked which is kiki and which is bouba, seem to prefer to call the pointy shape kiki and the blobby shape bouba (or takete and maluma in the earliest version of the experiment, but it should be obvious which corresponds to which shape). There seems to be a sort of synaesthetic association where particular sounds seem pointy or blobby. Shown the Turkish words küçük and büyük, even if you know absolutely no Turkish you can probably have a pretty good guess which one means 'big' and which one means 'small'. Likewise Russian bolshoy and malenkiy, Italian grande and piccolo, Indonesian besar and kecil. Small words have small sounds like k and ch where you make your mouth small. Big words have sounds like b, where you move your lips apart to make a bigger hole.

A new question that began to be addressed in the second half of the 20th century, and one that's been pre-empted a little bit a few times by people I talked about in this miniseries - including Akbar the Great, in Abul Fazl's account of his experiment - is whether there is a Universal Grammar. Essentially, Universal Grammar is an attempt to explain a few weird things about language. Children seem to be able to very quickly, and without being deliberately taught, grasp complex grammatical structures despite only being exposed to a tiny fraction of their native language, and seem to end up knowing more about their native language than could have come purely from the inputs they have received. And children speaking the same language are all exposed to different ranges of inputs, but end up learning exactly the same set of grammatical concepts. And on top of that, there seems to be quite a lot that all languages do in the same way.

The best explanation, according to Universal Grammar believers, is to say that the concept of language, and some basic ideas about how it works, must be an innate feature of the human brain. Computers were an exciting new thing, and it seemed natural to suggest that language worked like a computer program. The human brain came with a program installed that would take inputs - nouns-phrases and verb-phrases - and do something with them to make sentences with meanings. The problem was, that linguists, studying an ever more diverse array of languages, were uncovering more and more different ways of organising sentences, categorising parts of speech, that sort of thing, different ways of doing language, that were steadily chipping away at what Universal Grammar believers could claim were universal features of the language program installed on all human brains, rather than just inputs specific to a person's experiences.

In 2002, Noam Chomsky, historically the most significant proponent of Universal Grammar, published a paper claiming just one linguistic feature as an innate part of human linguistic programming - that is, recursion. This is the ability to nest clauses within other clauses and still be understood, theoretically as many times as you like. I can say something like "I think that you know that I would have liked you to have seen that man that told me about that horse that was walking through the custard factory that had recently been the site of a custard powder explosion that investigators believe was orchestrated by the factory owner in order to destroy documents that proved that she had embezzled over £400,000 from a charity that she claimed was set up in order to deliver custard to those who had been worst affected by the custard shortages that occurred in 2023 because the custard supply chain had been disrupted by an incident in which former British Prime Minister Liz Truss hired a team of mercenaries in order to hijack a supertanker filled with custard powder while it was travelling through the Gulf of Aden in order to reach the Suez Canal.", and in theory you should be able to completely understand me.

It's arguable that the universals that Universal Grammar predicts should be common to all languages have become so vague that it's not even falsifiable. Usually one of the criteria for a theory to be considered scientific is that it should be falsifiable - that is to say, there must be something that, if it was observed, would prove it false. Bertrand Russell gave a good example of an unfalsifiable theory, intended as an argument against the existence of God: What if I said there's a teapot orbiting the Sun, between Earth and Mars. If you object that nobody's seen it, I could just say it's smaller than the resolution of the very best telescopes can detect. If they make better telescopes, I can just keep saying it's smaller. At some point it's obviously on me to provide evidence that the teapot does exist. Universal Grammar kind of has similar vibes, with its proposed linguistic universals becoming increasingly vague as counterexamples are found. On the other hand, defenders of Space Teapot Theory here might say that Russell's Teapot argument only works if you don't have a good reason to believe there is a teapot there, and that if nobody's come up with a better explanation for the things the teapot explains then it's still reasonable to say that the teapot's there and just too small to detect. Likewise Universal Grammar Believers could say that Universal Grammar has enough explanatory power that it can stand up to quite a few null results.

At the same time, the problem of the poverty of the stimulus - that children don't hear enough language to explain all the rules they seem to learn, the main purported problem with language acquisition, to which Universal Grammar is the purported solution, has come to be less of a problem.

For example, there's quite a lot of evidence that children are able to do a hell of a lot of the work of acquiring language, including identifying grammatical structures, and even things as basic as where one word starts and another one ends, by spotting patterns in how frequently different sounds occur in particular combinations. For example, if a particular group of syllables keeps occurring together in a particular order, there's a pretty good chance that's a word, and if you say those syllables together in that order then that will mean something.

Stephen Mithen, in The Language Puzzle, suggests how the emergence of this capability in early humans might relate to the origin of language itself. The idea is that early humans developed a capacity for statistical inference, which is generally useful for plenty of things other than language. They started from a system of calls that each had a totally self-contained meaning, but contained elements with sort of kiki-bouba-style associations with the things they meant. Early humans used statistical inference to gradually break up these ape noises into 'words' that consistently referred to specific objects and actions and could be combined in different ways to say different things about those objects and actions and their relation to each other.

In fact, if the kiki-bouba effect played this much of a role in the emergence of language, maybe Cratylus actually kind of had a point. Maybe words do, in some kind of distant etymological sense, ultimately end up having a connection to the things they mean. It actually seems much more reasonable than our early ancestors just kind of assigning noises to objects completely at random and expecting all the other early humans to understand what the bloody hell they were on about.

Overally, nowadays Universal Grammar isn't a particularly popular idea. Even Chompers himself doesn't really believe in it anymore.

The Origin of Language is an interesting thing to think about because it's still an open problem. So much of the time, in the History of Science, you're looking at people trying to answer questions that nowadays we've basically figured out. But here, we're still a long way off a consensus. Sure, there are things we can pretty confidently rule out as incompatible with a modern understanding of how things work. Nobody thinks humans are born with an innate capacity to speak Hebrew any more (at least, no respected linguists - I'd be surprised if there wasn't a forum somewhere full of people who believe it). But, for instance, Universal Grammar, is still going. It's been losing favour for the last few decades, but it's still just about within the realm of respectable ideas.

The idea that a particular language is the oldest language, or the most important language, or the language spoken by God, or the default human language is an old one, and in popular culture oddly enduring, even to the present day. It's not difficult to find people on the internet making wild claims about the antiquity of some particular language, how all other languages derive from Hebrew, or Arabic, or Sanskrit, or Tamil, or Albanian or whatever else, which presumably spontaneously emerged, fully-formed at some point in deep prehistory. The antiquity of a language is also closely tied up with ideas of the pre-eminence of a particular nation - As an example, it's generally Albanian nationalists who are the ones claiming all languages descend from Albanian. This seems to go all the way back at least to the oldest of these stories. Psamtik was not simply trying to find out whether Egyptian or Phrygian was an older language, he was in fact using this as a proxy for which of the Phrygians or Egyptians was an older nation. This connection also seems to be assumed by Akbar, at least if we take his account of his motives at face value. He wanted to find out what language the children would speak so he could follow that nation's laws.

I don't think it should be that surprising that this kind of story has kept its appeal over such a great span of space and time. Obviously language is an intriguing, very important, and kind of weird phenomenon. Every single human society uses language, but they all use different ones. Why do people from a hundred miles in that direction speak the same language as me, while people a hundred miles in that direction speak a language that sounds kind of weird but I can mostly understand, and the people a hundred miles in that direction speak a language that has nothing in common at all with mine?

Language feels like it's intimately connected with the process of rational thought, too. I, for one, feel like I understand an idea much more coherently if I can express it using language. If you want to pick one thing that makes humans stand out as special among the animals, language would be a pretty good contender. And why is it that an adult takes months of intense study to be competent in a second language, while babies, who are generally kind of stupid, seem to manage it effortlessly, figuring out rules that cause endless problems for linguists and language learners without even realising they're rules at all?

And at the same time you have something that seems at first glance like it might solve some of these problems, but that's too ethically despicable to consider actually attempting. It's ideal if you want to paint someone as a sort of mad scientist willing to put aside any and all ethical scruples in the pursuit of knowledge, even if in practice it probably wouldn't actually settle any of these questions.

And that's it for this miniseries. I hope you've found it as interesting to listen to as I've found it to research. The issues of the history of linguistics and the history of experimental science that these experiments bring up are huge topics, and there's so much more I could have included that I didn't reasonably have time for. Please subscribe to the Science: A Peculiar History on whatever podcast platform you use, like and comment on the episodes, share the episodes wherever you can, and follow the podcast on Facebook, Twitter, Instagram, and Bluesky. You can find episode transcripts, pictures, sources, etc. on the website scienceapeculiarhistory.co.uk. If you have any questions, comments, corrections or suggestions, you can message any of the social media channels, email admin@scienceapeculiarhistory.co.uk, or use the contact form on the website.

Thank you for listening. Come back next time for a completely new miniseries.