The Origins of the French Language by Olivier Robichaud

Today’s special episode is by Olivier Robichaud. Olivier is the news editor for Métro, a daily newspaper based in Montreal, Canada. His main specialties are local news, municipal politics and the law. Olivier has also studied sociolinguistics, specifically the variation of French across time and space. In this talk Olivier will explain the origins of Old French and what impact the Latin, Gallic and Frankish languages had upon it.

Olivier: Hi everyone, my name is Olivier Robichaud, and I’m here to talk to you about the evolution of the French language.

To start this off, just a little word about myself. I’m a journalist by trade, I work for a daily paper in Montreal, Canada. But I’ve also studied linguistics, specifically the variation of French in time and space. I’m also a huge history buff and as such I’m a big fan of Gary’s podcast. So when he asked me to do an episode, I was thrilled.

The reason I’m doing this is because A) I’m a nerd and B) there was a question that kept creeping subtly into Gary’s narrative and I thought needed to be addressed more thoroughly. That is the question of how the French language came to be. Did it come from Frankish? From Latin? How much hybridization actually occurred in West Frankia in the early Middle Ages? And what, if anything, is left of the celtic language that preceded the Roman invasion?

So I pestered Gary with my thoughts on the issue, and he kindly offered to let me share them with you on his show. So here we are.

Now, before I start, I have to put in this little diclaimer. Although I studied linguistics in college, I lack the impeccable credentials of some of the other guest hosts that have come on the show. I’m no researcher or academic. I’m a journalist. So I’m not in a position to discuss the various disagreements or give credence to this or that alternative view. So take what follows as a relatively traditional, relatively consensual view of what happened, among linguists. And keep in mind that there are other views concerning some of the specifics.

Let’s answer one of those questions really quickly: French came from Latin. But what I want to explore is exactly how we went from a Celtic pre-Roman Gaul to a Latin imperial Gaul to a kingdom that speaks a new language called French. How does that happen?

Don’t worry, I won’t go through the whole history again, Gary already did that quite thoroughly. And if you enjoyed it as much as I did, then we’ve all spent a very good time together.

Now, the first question, the one about passing from celtic to latin, is both the hardest and the easiest to answer. Hardest because the Gauls famously refused to write their language down. Even when they were literate, they wrote in Latin. So we have to tease out whether or not people still spoke Celtic from passing comments from Latin authors. And those comments can be particularly hard to come by in some periods. I mean, remember the «lost centuries» that Gary told us about? Yeah…

Now, if you’re thinking the Romans just killed everyone and settled Romans in the empty space, that’s one of the few options that we can say definitely did not happen. There is plenty of evidence of social mixing between Gauls and Romans after Cesar’s conquest. So what happened? In short, the Gauls probably just preferred to use Latin for a number of social and political reasons. And eventually, a critical mass of them just didn’t see the point of passing on the celtic language to their kids. It took a while, in fact it took until a few decades after the fall of the Western Roman Empire, but it happened.

I say it’s also the easiest question to answer because I can basically say «we don’t know Jack» and move on to more recent examples that can explain what might have happened.

How and why such a phenomenon occurs, where an entire country abandons one language in favor of another, is one of the focusses of a field of study called sociolinguistics. That is the study of how society and politics influence a language and vice-versa.

I’ll give you an example. Those of you who live in linguistically diverse areas might have noticed that people use one language in one context and another language in another context. In other words, languages can have differentiated social functions. People in Arab countries use their local dialect for daily communication, but Classical Arabic at the mosque. Haitians use Creole in their daily lives, but French in government. In America and Britain, recent immigrants will use their native language in the home, but English when speaking to other members of the community.

So what, you say. Well, those social functions are not innocent. They reflect the power dynamics within the society and they determine to a large degree what might happen to each language in play in the future. There are ALWAYS power dynamics between different languages used within a given society. Heck, there are power dynamics between the different linguistic styles or variant between social groups who speak the SAME language. But that’s a totally different scenario that has nothing to do with Latin vs Frankish. Anyway, to get back to the point, in most cases, the language of the majority is dominant in a given society and will tend to assimilate other linguistic groups. But in many cases, a minority group has a disproportionate influence on the rest because they have obtained political, economic and cultural domination in that particular society. When this happens, the minority language can have a huge influence on the majority language. It can change it in many different ways, and it can even replace it.

Take Ireland. It is a fine example of how state policy and social stigma can bring a language to its knees. Dispossession, exclusion, political and economic marginalization, and a brutal, brutal famine brought many people to the conclusion that if they wanted to get buy, they had to learn English. And that all happened quite recently, so it’s all documented and you can learn about it in any good history book about Ireland.

In other cases, you don’t even need active discouragement for people to abandon their language. The city of Brussels is a very interesting example. For a long time, French was the main language used in the public sphere in Belgium. Work, politics, higher education, the law: everything important for social mobility was essentially off-limits for the Flemmish inhabitants of Belgium, who speak a version of Dutch. This, by the way, was a period of active discouragement of Dutch. But after years of domination by the francophone part of Belgium, the Dutch-speaking region, called Flanders, got a large amount of autonomy through successive legislation from the 1920s through the 1960s, notably on linguistic issues. Flanders was able to reinvigorate the Dutch language inside its borders. But the capital Brussels, which is situated well into Flemmish territory, was given a bilingual status. So while Flanders could legislate to make sure, for example, that its inhabitants learned Dutch in school rather than French, all schools of all languages were open to everyone inside the city of Brussels. And so were many other services.

Well… after all that political, economic and cultural domination in previous decades, French was still the prestige language. A lot of business was conducted in French, and the French schools were more reputable. So guess what happened. That’s right, the Dutch-speaking inhabitants who wanted to give their kids a leg up in the world encouraged their kids to learn French, while almost none of the francophones did the opposite. Those Flemmish kids kept using French in the workplace when they grew up and they taught French to their kids. Now, by the time the 1960s came and went, francophones were already the majority in Brussels, but Dutch-speakers still accounted for about a quarter of the population. But even after Beligum got rid of most of the barriers and stigma attached to the Dutch language, that process continued. After a couple generations of linguistic laissez-faire, today they are down to 5%.

Those two examples, Ireland and Brussels, illustrate some of the forces that were probably at work in Gaul during the Roman Empire. In both cases, we find a less prestigious group abandoning its language in favour of that of a more prestigious group, for reasons that are specific to the local context.

How does that translate to Gaul? Again, it’s hard to tell. But we do know that Celtic elites had romanized quite rapidly, so they knew Latin. We also know that the incoming Franks were learning Latin as well so that they could more easily invest the institutions and power structures that were vacated by the collapsing Roman Empire. So there are two prestigious groups right there with Latin as a common language. It’s possible that Celtic was simply seen as having no future. And it disappeared from most places during the 500s.

So that’s how we got from a Celtic Gaul to a romanized Latin Gaul. Now how did we get from Latin to French?

To answer this question, we need both traditional linguistics (that is, the study of the language itself) and add on a touch of sociolinguistics.

The first thing you need to note is that all languages change over time. You probably noticed that the first time you came across Shakespeare in school, and that is still Modern English. Have you ever tried reading Chaucer in the original Middle English, or Beowolf in Old English? Complete gibberish to modern ears. Our Language has changed a lot, and Latin also changed.

Also, keep in mind that the Roman Empire was big. Like, really big. So it’s no surprise that local styles of speech popped up in different places. Now, as long as the central state was strong and Rome was culturally vibrant, that wasn’t a problem. There was still a common reference point for everyone, a baseline Latin language that everyone could agree was shared and the little deviations from it were just that: small deviations. It was like chips and fries in Britain and the US. Americans and Canadians might laugh at Brits for calling a fry a chip, but we all understand and it’s all still English.

But when the central roman state collapsed during the 5th century, and probably a bit before that, that common reference point also collapsed. With trade routes closing, literacy falling and Rome being just a shadow of its former self, there was no common reference point. People looked to their immediate community. So those small deviations took on a life of their own.

With these two things in mind, lets explore some of the most fundamental differences between Latin and French and how those differences arose.

If you’ve ever studied Latin (which I have NOT, by the way, I only have a superficial knowledge of certain rules), you might have had difficulty with the grammar because the words change so much. In modern Romance languages like French, the function and caracteristics of the word in a sentence is determined by its position. Just like in English, the subject comes first, then the verb, then the object. And if we switch that order around for effect, we have to add in other words that explain that change. Like if I want to go from «She ate the apple» to «The apple that she ate», there is the extra word «that» which tell us that I’ve moved the object to the front of the sentence. Classical Latin doesn’t do that. The position of the words matters not one bit: the function and caracteristics of each word are given by prefixes and suffixes. So the word itself is changed rather than its place in the sentence.

I’ll give you an example. If I want to say «Peter sees Paul» in Latin, I’ll say «Petrus videt Paulum». Or I can say «Paulum videt Petrus». It’s the same thing. Whether Petrus comes first or last in the sentence, we know that he is the subject because of the «us» tacked on to the end of his name. If I wanted to say «Paul sees Peter», I would say «Paulus videt Petrum». Or «Petrum videt Paulus». In all cases, the subject is determined by the person that gets «us» tacked on to their name, and the object is the one that gets «um».

Well, as languages change, certain letters are more unstable than others. Which letters are unstable varies between languages and contexts, but the last letter of a word is often very likely to get axed.

So in spoken Latin, people just stopped pronouncing the «s» and «m» and the end of «Petrus» and «Paulum». So it became «Petru» and «Paulu» in all cases.

And now all of a sudden, word position becomes super important. Because the word itself doesn’t tell you which one is the subject and which one is the object anymore.

For American listeners who know a bit of Spanish, it’s like when you talk to someone from certain South American countries and you realize they’re not pronouncing the «s»es at the end of words. It’s not too complicated when it’s just the s of the plural form that isn’t being pronounced. You can figure out that there’s more that one «taco» even if he didn’t say «tacos». But sometimes you can’t figure out who the heck they’re talking about because they keep omitting the «s» at the end of a verb. Spanish verb endings are supposed to tell you who’s doing the action because Spanish people tend to omit the pronoun. But if there’s no «s» at the end you can’t tell if it’s supposed to be second person singular or third person singular. In other words, you can’t tell if their talking about you or about some other person.

When such confusion enters a language, the speakers of that language develop new ways to communicate what they want to say. So late Latin speakers used other words and phrases to place help place the nouns and verbs in relation to one another. That’s one of the reasons why French has so many articles and prepositions: those relationships constantly have to be expressed.

By the way, Latin had a bunch of those word endings for all sorts of functions and caracterstics, including determining gender. And if you’re trying to learn French and it bugs the hell out of you that you don’t know why the cup you’re holding is masculine and the beer inside is feminine, here is the answer. Latin used to have a «neutral» gender, as well as masculine and feminine. So if you were to add an adjective to describe your cup or your beer, you wouldn’t have to make it sound like it’s a boy or a girl. Which is, you know, normal. But again, the last letters fell in late Latin. So in adjectives, the gender neutral «um» and the masculine «us» both became «u». So now there was no way to have a neutral gender for the adjectives of things like cups and beer. Everything was either a boy or a girl. And the choice was completely, utterly arbitrary.

So that’s what was happening in the spoken Latin of the late Empire. In fact, some of this was happening much earlier, as the Republic was turning into the Empire in the last decades before Christ.

And when the Empire split, it became worst. The styles of Latin spoken in Gaul, Iberia and Italy (among other places), they all drifted away from each other. Part of it was because of influence from German occupyers, including the Franks. But a large part was just more of what we just saw: changes occurring because of the internal dynamics of the language.

So where do we draw the line between Latin and French? Well this is where sociolinguistics come in. Because the line drawn between one language and another has nothing to do with the languages themselves. It has everything to do with society and politics.

In the case of Latin, for a long time the little changes over time, they didn’t matter all that much. Latin was changing, but it was still recognizable as Latin. There are letters from bishops in the 400s and 500s telling their priests to use a «rustic style» when preaching to their flocks. Something like «Don’t go stressing your “us” and your “um” or you’ll sound too highfalutin».

Things are murky in the 600s and early 700s. But one thing we know for sure is that by the time of Charlemagne, the latin spoken in Gaul was no longer understood by people who had learned latin as second language using books from Antiquity. No longer did we talk about a «rustic style», no. Charlemagne and his court spoke of the people’s language as a «lingua romana rustica», or the «rustic roman language».

That was in fact one of the motivations behind Charlemagne’s educational reforms. He wanted people to speak «proper» latin. But that mission failed, and near the end of his reign in 813, he signed a document asking priests to attend to preach in that «lingua romana rustica» instead of classical latin. That document is actually the first known use of that phrase. And the «romana» part of that term is why all languages and dialects that evolved out of latin are called «romance languages».

Now here’s the crux of the issue. The people writing those kinds of admonitions were very often the ruling Franks who spoke Frankish. They were learning Latin as a second language, and to do that, they used grammar manuals written in the 300s AD. And many of those manuals were based on texts written by people like Cicero and Cesar 300 to 400 years before that! So you can see why there was a disconnect, the language had lots of time to evolve and that evolution wasn’t taken into account. It would be kind of like if Russia invaded the US and forced everyone to talk like Chaucer. Imagine that.

Another group of finger waggers were clerics. Sometimes Frankish, but in many cases, they were native Latin speakers. But the Latin that they learned for church using those same manuals was so valued by the church that they came to disregard their own vernacular language as being less valuable.

These two groups didn’t recognize the Latin of the people they ruled over or preached to as being Latin. And they were in a social position that made them able to impose their views on the majority. They had all the political power and all the cultural prestige, which meant they could call the shots linguistically.

Case in point. Greek has an even longer history than Latin and it’s changed at least as much as Latin since the time of Pericles. So why do we still call it Greek? Quite simple: because no one from the outside had the nerve to tell Greek people that they weren’t speaking Greek.

So that’s how we went from Latin to French. But what about the name of the language? If it came from Latin, why do we use a word that is clearly derived from the name of the Franks?

That’s another interesting piece of history, and of sociolinguistics.

The name of the kingdom was originally Frankia, and then West Frankia (meaning Gaul) when Charlemagne’s empire split. The West part was rendered useless when the other parts of the Empire took different names, so it was just Frankia. During the late Empire and early Middle Ages, the K sound became more like a TS sound, so even in Charlemagne’s time it was probably pronounced more like Fran-tsya. And then eventually France during the High Middle Ages.

Well, there were many different dialects spoken in France in the Middle Ages. And when people wanted to stress that they were talking about one or several of the local vernacular languages, they might use the term «langue françoyse» or «langage françoys», in opposition to Latin. But it didn’t mean that there was something specific called «LE langage françoys», or THE French language. It just meant that they weren’t speaking latin, so it could be any one of the numerous dialects spoken at that time in France.

So in 1539, when the king of France decided that his laws would be written in the vernacular language instead of Latin, and also to allow his subjects to use their own mother tongues in the courts, he signed a decree saying that justice should be rendered «en langage maternel françoys». Litteraly, «in a native speech of France». So anyone should be able to use their own dialect in the courts instead of Latin.

Problem is, the king spoke the Parisian dialect, so that was how his laws were written. And most lawyers and judges were either born in or trained in Paris. So they used the Parisian dialect in the courts. And because they, and of course the king, had the prestige to call the linguistic shots, their idea of what French was became THE idea of what French was and that specific Parisian dialect took for itself the name «françoys». And that later became «français». Or in English, French.

To be fair, this is ONE interpretation of what happened based on what we think the king meant when he said «en langage maternel françoys». An older interpretation has the king chosing the Parisian dialect as the only real «langage maternel françoys» and making it the language of the laws and the courts, but that interpretation has fallen out of favour.

By the way, I didn’t give you the king’s name because it would have been confusing. His name was François, another variation on the Frankish name.

So in essence, and to paraphrase a famous French linguist: French IS Latin as spoken in Gaul in the 21st century. The reason we call it French has nothing to do with how different the language became from Classical Latin and much more to do with the negative perceptions of outside linguistic groups, as well as the snootyism of Parisian lawyers.

So what about the Franks? Did they have no influence at all on French? Not quite. They did have an influence, and it was a stronger influence than what you might find in other Romance languages like Italian or Spanish. But most of that influence comes in the form of imported words, as well as some changes in the phonetics of the language.

Those two elements of language, vocabulary and phonetics, are the two most superficial elements of language. They change a lot, all the time. And that is where you find most of the Frankish heritage in French.

To be fair, there were some grammatical influences. The biggest one was a preference in word order, so the subject-verb-object order would be replaced by something else in some contexts. But that was later canceled out by other forces, so it’s A) a little difficult to explain in English and B) kind of a moot point…

Anyway, as far as the vocabulary is concerned, linguists have calculated that there are a bit more than 500 words of Frankish origin in Modern French. Back in Charlemagne’s time there would have been more because some were dropped over the centuries. So let’s guesstimate about 1000 words. That’s a fair amount. In any case, it’s much more than the 150 or so words that come from the original celtic language. But compare that with the 10 000 words that passed from French into English after the Norman conquest in 1066. And I’m only talking about a period of a few hundred years between 1066 and 1399, never mind the words that came into English after that.

How big is that number? Well, any given language has about 30 000 core words that are necessary for proper daily communication. Each individual speaker knows about twice as many words, but most are specialty words related to a trade or a region or something like that. But of those 30 000 core words in English, fully one third come from Norman French. And yet we still call English a Germanic language, not a Romance language.

I’m telling you this because I want to illustrate that although French did evolve under the influence of Frankish to some extent, it is not a hybrid language. It’s Latin, with some outside influence.

I’ll spare you the phonetic influence because there’s only a couple noteworthy elements and this podcast episode is already overly long.

So to sum it all up: French is a Latin language, with notable Frankish influence but certainly not a high degree of hybridization, and the celtic heritage is practically nill.

That’s all for me. I hope you had as much fun listening to this as I had writing it.