How Do We Know How Languages Are Related?

When two languages share some things in common, they might be related, but they also might not be. How do we know? Produced for Mental Floss

[@rzeka]

I like the examples of "heart" and "hound" to show how sounds can change.

[@TomRNZ]

In an extinct Australian Aboriginal language called Mbabaram, their word for _dog_ is _dog._ The language is completely unrelated to English, and the similarity is just pure coincidence.

[@annaireton8816]

Ah yes the floor is made out of floor

[@TheThekingrass]

Really like these videos, great job

[@domception9108]

Excellent as usual. Thank you!

[@HiddenXTube]

One of the best channels about language on youtube!

[@funzjag]

This is a very interesting channel! I've no choice but to subscribe. Peace and love from West Virginia!

[@causilvestrini7000]

I love these videos, they're utterly interesting and both the drawings and the words are chiefly done. Thank you and please keep them coming! :-D

[@inventorofmachines]

You guys are awesome. Nice pictures too.

[@EliGoldfish]

Super informative, concise and the whiteboard visual aids are pretty to look at AND reinforce the concepts well.

[@sararielle]

We can look at commonalities to create what a potential proto-language may have looked like but we learned from the construction of "proto-romance" from various modern romance languages that our reconstruction may have important pieces missing. "Proto-Romance", for instance, was missing the case system we see in Latin (the *actual* proto-romance language) because none of the remaining descendants retain explicit case structures.

[@rzeka]

Isn't Proto-Romance just Latin though? or are you talking about Proto-Italic?

[@sararielle]

That's the point. The proto-Romance language *was* Latin. Unlike many proto- reconstructions, we actually have the proto- language in this case to compare our reconstruction to. So the researchers used the standard methods to do a reconstruction based on modern Romance languages and then compared it to Latin as a way of seeing how good the methods are. i'm trying to track down the original paper. I'll let you know when I find it.

[@rzeka]

Sara Melnick OH okay, I get it now! That makes perfect sense, sorry I misunderstood you. That's actually a really important point that I've never thought about before. I'd love to see the papers please do let me know.

[@gf4670]

Not true. "Latin" was not the proto-Romance language, but there also wasn't exactly a "proto-Romance" to begin with, merely a gradation of dialects that evolved out of colloquial Vulgar Latin over many centuries and geographic extent. For instance, Iberian Romance is largely descended from somewhat earlier dialects since the Latinzation of Iberia was quite early, whereas the Gallic Romance and many Italian forms evolved out of later variants which showed certain innovations that had developed among the more-native Latin population in Italy, and there was of course a Balkan version as well which developed into Romanian. Often it's stated among Romance linguists that Spanish and Portuguese are descended from "Republican" vulgar Latin and Italian and French from "Imperial" vulgar Latin. And this isn't to say that they weren't mutually intelligible or even the same language, they most certainly were, but the strains of influence constituting the "proto" in their development are extremely complex and dynamic across time and space. And this is where the study of Romance is very important is illustrating many of the common subtleties of language change that we've found also apply to other languages like Chinese. Basically what I'm saying is that an actual "proto" language of Romance (and Latin) lay further back in history to probably closer to the Punic Wars, back when Old Latin was a much more unified language community and before the standardized version we call "Latin" had started to be formulated around the 1st Century BC and from which the colloquial Latin of the soldiers and merchants DID NOT develop. Also the existence of the cases in Latin (classical or vulgar) are obvious to linguists even without the ancient texts. Romanian has retained 4 cases and the others have several unique features about them, especially certain preposition peculiarities, which make it clear that what they descended from used grammatical case to denote syntactic relationships.

[@tideghost]

Wouldn't by that logic, Sanskrit should be considered as the Proto-Indo-Aryan language? The way the Indo-Aryan languages arose is quite similar to Vulgar Latin diverging into different languages. Sanskrit was like Latin, not widely spoken, while Prakrits were spoken by commoners and in daily life.

[@torrawel]

tiny note: when the latin alphabet arrived in Mexico, the word became written as "xocolatl" . The Nahuatl word started with a "sh" sound, which, in those days, was still part of the Spanish language and written with the letter x. Later (especially in Spain) the sound shifted to Spanish "j" (jota, as in: Don Quixote, Ximénez &... Mexico) and eventually to "ks". In modern-day Nahuatl it is still used as an "sh" sound. Which means that Mexico (a Nahuatl word by the way), is pronounced "Meshico" (actually: Mēxihco: "meh-she-'ko"). And Nahuatl was THE Aztec language, not "an" Aztec language (an "Aztecan" language would be correct)

[@theblackryvius6613]

Very interesting stuff

[@tideghost]

IPA would help...

[@torrawel]

@@tideghost for sure! But it would also make it more boring :). Aztec glyphs are also still used in Mexico (for example in the national flag). They look a lot better than the Latin alphabet or IPA😉

[@thomashogan16]

His drawings are fantastic! What a talent. How I wish I were still teaching English...I'd have used these all the time.

[@sillysod33]

Wonderful video, so informative, thank you! 🙏🏼

[@rjspangler1488]

Yeah, excellent job!

[@sigutjo]

Brilliant as always!

[@miunone09]

It's "xocolatl"! With an X!

[@CanuckJim]

Kerd and kun for dog reminded me of the Scots Gaelic for dog: cù

[@imokin86]

BennyGesserit, it is indeed related to them, if I remember correctly. Does an N show up if you build a plural or an oblique case of cù?

[@CanuckJim]

Yes! "Coin" is the plural - I believe that's the spelling (my grandad spoke it but I, being a truly dumb child, never thought to learn!)

[@imokin86]

Yeah, that proves it.

[@nromk]

no in Spanish you can not say can for dog, the word you are looking for is canino, which can be implied to all members of the dog family in science, a wild dog or wolf in some cases, and a another way of calling a man a dog

[@Furienna]

@@nromk And yet, "kanin" in Swedish means "rabbit". Go figure...

[@taeyeonsnose7003]

This was well explained and the drawings are amazing! Really well done :)

[@pewp_tickalar]

such a well-constructed video. hours of Wikipedia-reading expressed in just a few minutes

[@sgs6991]

In náhuatl it is Xocolatl

[@yoroshiku137]

Noticed for the first time after countless videos that the guy drawing on the board looks elsewhere for a reference of what to draw. That killed the Magic a little.

[@Stellarainn]

Romance

[@t.k.abrams4720]

That map at the end is so awful

[@lialanikaija7127]

all languages descended form one language the nofastic language.

[@louisxvii2137]

This video is a propaganda lie.

[@americannotamericant1773]

нαρργ моνие is it because they didn't add Russian? Well my friend your language is complete nonsense so it is neither related to Germanic or Latin Romantic languages.

[@americannotamericant1773]

The Idiodyssey he downplays the true legitimate facts that these languages are familiar his notion of the other is obscure and quite ridiculous. His name and his demeanor towards the fact and also his name makes me think he is Russian. Also the more educated Russians can speak English.

[@americannotamericant1773]

The Idiodyssey he downplays the true legitimate facts that these languages are familiar his notion of the other is obscure and quite ridiculous. His name and his demeanor towards the fact and also his name makes me think he is Russian. Also the more educated Russians can speak English.

[@rzeka]

What are you on about? The comparative method is pretty hard to prove wrong. If you're going to try to deny it, the evidence is really stacked against you. And propaganda for who? Who is benefiting from having people think languages are related to each other? If someone is going to make up a lie, why lie about this? and uhh other guy, Anthony, I don't see anything that would suggest he's Russian. The letters in his name are Greek - and even then, they're not even spelling anything in Greek, they're made to look like English. So if anything I would think he's an English speaker. And even if he is Russian why is that relevant?

[@americannotamericant1773]

rzeka well bull shit me lmfao well either way his complexions on the language relationships is totally wrong