37

Reading some dialogues from Socrates, it struck me how eloquently the people seemed to speak from those times thousands of years ago. (Although this might be a result of the translation.)

And yet this was a time when philosophy, logic and science was being invented.

It would stand to reason that a language could not just appear fully formed. But gradually grow over time. (e.g. the word "eloquently" is quite a specific concept which is unlikely to have existed in the first proto-languages).

So do we know what time periods correlate to the sophistication of a language. i.e. did Ancient Greeks have a vocubulary similar to modern humans. What about before that?

Or are there any examples of historic languages with far fewer words and being much more basic?

zooby
  • 653
  • 1
  • 5
  • 9
  • 6
    Latin & Greek were well developed for purposes of philosophical discourse. Primitive languages could get by with declaratives, questions, and commands; but advanced languages require explanations, conditionals, qualifications, statements of purpose, etc. Take a look at the oldest hymns in the Rigveda. I find them primitive, but the Upanišads very sophisticated. Sanskrit syntax seems clumsy compared to L&G. – Bert Barrois Dec 16 '19 at 13:46
  • @Bert How could they develop a language without a language to express what they were trying to develop? It seems quite circular. – zooby Dec 16 '19 at 17:50
  • Are you working from the original Greek, or from English translations of the dialogues? – Jon Kiparsky Dec 16 '19 at 20:03
  • @Jon An English translation so, yeah, could be embelshed somewhat. – zooby Dec 16 '19 at 21:01
  • 1
    Probably not "embellished", but certainly rendered in a fashion to make the reasoning clear to the modern reader, ie, in "sophisticated" language. – Jon Kiparsky Dec 16 '19 at 21:20
  • 7
    Does this answer your question? Are some languages more advanced than others? – Sir Cornflakes Dec 17 '19 at 11:54
  • @zooby "How could they develop a language without a language to express what they were trying to develop?" That seems like a good question for a whole post. – Ryan_L Dec 17 '19 at 19:20
  • 3
    I feel like it must be said that Plato's dialogues are probably not representative of how eloquently people spoke in Ancient Greece. – Mason Dec 18 '19 at 01:38
  • 3
    As Mason pointed out, Plato's dialogues are not recordings of how people spoke. They are representations of conversations that Socrates had, written after Socrates was executed. The rhetoric of someone writing and editing and carefully choosing their words is probably going to seem more refined than a spontaneous conversation. – matan-matika Dec 18 '19 at 06:35
  • There was no way to say "iPhone" in Ancient Greek. – Bob Jarvis - Слава Україні Dec 19 '19 at 12:22

4 Answers4

65

Or are there any examples of historic languages with far fewer words and being much more basic?

Interestingly enough, there are not! Nor are there any examples of "more basic" modern languages (*).

In fact, there's a sort of axiom in modern linguistics that all languages with native speakers are equally expressive. Anything that can be expressed in English can also be expressed in Arapaho or Cantonese or Swahili—or Ancient Greek or Ancient Egyptian or Classical Nahuatl, for that matter. You might have to invent new words for concepts like "computers", since the Ancient Egyptians didn't have those, but you can work around that with, say, "a machine that does calculations".

Now, there are some languages that are more basic: "pidgins", which arise naturally when speakers of different languages need to make basic communication work between them. These tend to have extremely simple grammar without much expressiveness. But, even more interestingly, when children grow up speaking these languages, things change—the result is a creole, with full-fledged morphosyntax and as much expressiveness as any other natural language!

This is one of the big pieces of evidence in favor of a "language mechanism" in the human brain. Different theories then go in different directions from there, arguing about how much exactly is innate. But it seems clear that some underpinning of language is, in fact, intrinsic to humans.

So, where did language come from in the first place? Good question—nobody really knows! There are quite a lot of different hypotheses, but very little evidence to test them against. So for now, it remains one of the big unsolved mysteries of linguistics.

(*) There's a very controversial claim that the Pirahã language spoken in South America is actually simpler and more basic than any other. Some linguists support it; others disagree with it. Personally, I think such an extraordinary claim requires extraordinary evidence, and there simply isn't enough evidence to back this one up.

P.S. If you want to investigate this further, John McWhorter's research on creoles would be a good place to start. Some other sources are mentioned in the comments.

Draconis
  • 65,972
  • 3
  • 141
  • 215
  • This seems to suggest that the number of words of a typical language would correspond approimately in some way to the number of different mental states or processes in the brain. i.e. to express what's in our brain we need this number of words. IDK. That's just a feeling. And the language grows to fit this number. – zooby Dec 16 '19 at 02:42
  • 13
    @zooby Possibly, though the idea of "words" is not as well-defined as you might think! Languages also split up the huge space of concepts in various different ways, which don't always align with each other. – Draconis Dec 16 '19 at 02:48
  • 1
    Um, creole formation isn't an argument for innateness, particularly. It could argue for a lot of positions. And no linguist I know claims that Pirahã is actually simpler or more basic (whatever those might mean); merely that it has certain features and lacks certain others, like any other language. It's ex cathedra claims about which features are basic that are the cause of the controversy. – jlawler Dec 16 '19 at 02:49
  • 7
    @jlawler Interesting; I've seen creoles presented as an argument for some sort of language mechanism in the brain, since they show that humans can create these grammatical structures without being taught them (with some particular examples of features arising in creoles that the parent languages didn't have). And I've definitely seen claims that Pirahã is theoretically finite, unlike all other human languages, because it lacks recursion. That's mainly what I'm referring to in the last paragraph. – Draconis Dec 16 '19 at 02:51
  • 2
    Have similar investigations been conducted based on Nicaraguan Sign Language? – Nardog Dec 16 '19 at 05:44
  • @Nardog Very good question! Unfortunately I don't know. – Draconis Dec 16 '19 at 05:51
  • 5
    fun note, computers as a word predates computers as we know them today, rather than a machine that computed though, they were people who computed, so computers was a job role, one requiring math skills. Pythagoras may well have been a computer, just as calculators are named after the job of calculating things. NASA employed computers to calculate things prior to the introduction of IBM machines – Tom J Nowell Dec 16 '19 at 15:25
  • 1
    @zooby, natural languages have synonyms that have same meaning while being different words. So, there can be more words, than "state and processes". Also, single word can contain multiple roots each representing different lexical unit. So, one word can represent multiple "state and processes" at once. – user28434 Dec 16 '19 at 16:17
  • @TomJNowell Asimov's "End of eternity" features this meaning of the word extensively and it caused me no end of confusion until I looked it up. – TimothyAWiseman Dec 16 '19 at 18:27
  • 2
    @user28434 Its even more complicated than that since there are many words with almost completely unrelated meanings. Its even possible to have a word that depending on context can be its own antonym, sometimes called a Janus word. https://en.wikipedia.org/wiki/Auto-antonym We also have set phrases that are effectively words since they can have a set meaning separate from what the individual words would suggest if taken literally. – TimothyAWiseman Dec 16 '19 at 18:29
  • Just a layman, but I disagee with the axiom that all languages with native speakers are equally expressive. A very fine example sentence that can be expressed in English yet not in e.g. Hebrew is "all animals are equal, but some animals are more equal than others". The equivalent Hebrew would translate back as "all animals are worth the same, but some are worth more than others". Any translation which attempts to keep the "more equal" meaning feature as intended becomes an entire paragraph unto itself. We could also address puns, synonyms with varying connotations, and other features. – dotancohen Dec 16 '19 at 21:03
  • However, Through the Language Glass (Deutscher) and The Power of Babel (McWhorter) both assert that subordinate clauses are a feature of written languages, and don't appear (or are very rare) in non-written languages. – TRiG Dec 17 '19 at 16:09
  • 1
    @dotancohen Wordplay, including puns, is very challenging to translate. But that doesn't mean that similar concepts cannot be expressed in virtually any language. Even in English the quote from Animal Farm doesn't make sense if you take it literally. I don't know Hebrew, but I strongly suspect you could readily translate "We shall pretend that all animals are equal, but in truth some shall be given more authority" which is sentiment being conveyed by Orwell. – TimothyAWiseman Dec 17 '19 at 17:10
  • 1
    @TimothyAWiseman: Sure enough your quote can be translated into Hebrew. However, the explicit declaration of meaning is overt, whereas the original English quote with its covert subtleties conveys additional meanings, such as distrust. The subtleties are actually the relevant part of the quote, so losing them fundamentally changes the message. – dotancohen Dec 17 '19 at 18:19
  • 5
    @dotancohen Sure, but that doesn't make Hebrew less expressive than, it just means that wordplay doesn't translate easily. I don't know Hebrew but I think I'm safe in guessing that it has plenty of wordplay that doesn't translate easily into English. – TimothyAWiseman Dec 17 '19 at 19:24
  • 2
    @TRiG I haven't read those books, but if they make that claim it's obviously wrong. No Australian languages were written down until very recently, but all that have been studied make much use of subordinate clauses, so that's 300+ languages. The same could be said of the languages of PNG, so that's another 850+. – Gaston Ümlaut Dec 17 '19 at 21:32
  • @dotancohen in these cases it's useful to remember that English is thee languages in a trenchcoat pretending to be an adult. Very few languages can manage this level of ambiguity but that does not mean english is more expressive, just that its rules leave a lot of room for acrobatics between the meanings. – Borgh Dec 18 '19 at 08:30
  • I have read that there are languages in which the concept of "five" cannot be expressed. (It goes something like "one", "two", "many"). Now that's very specific, and I don't know how reliable that claim was. However, that claim isn't entirely unifiable with some made in this answer. – Jasper Dec 19 '19 at 09:09
  • Extremely interesting answer. Could you please provide some sources to support your claims? Such as "there's a sort of axiom in modern linguistics that all languages with native speakers are equally expressive". Thanks =D – Kyll Dec 19 '19 at 10:07
  • @Kyll Now that's a good question! The idea that all languages are equally expressive is pretty pervasive in linguistics, and I'm not sure when it first arose. But John McWhorter's research on creoles is probably the most thorough test of it—McWhorter argues that, while all languages are equally expressive, it's possible to detach the complexity of a language from its expressiveness, and complexity (all the little idiosyncrasies and irregularities and such) accumulates over time until it hits a limit that's inherent to the human brain; after that point, some complexities are… – Draconis Dec 19 '19 at 18:27
  • …smoothed out, while others arise, in a zero-sum game. His main evidence for this comes from creoles, which he argues have equal expressiveness but much lower complexity than other languages when they first arise. McWhorter's written prolifically about this, but from a quick search, the best citation would probably be The world's simplest grammars are creole grammars (2006). – Draconis Dec 19 '19 at 18:30
  • 1
    @Jasper The trick is, there's a difference between "language X has no word for Y" and "language X cannot express Y"! Ancient Greek has a single word for the number 10,000 (myriás), and English doesn't—but we get by just fine saying "ten thousand" instead. Quite a lot of our most basic words for numbers come from these sorts of workarounds: the reconstructed Proto-Indo-European ancestor of "eight" literally meant "two fours", for example. … – Draconis Dec 19 '19 at 18:35
  • … Now, one of the controversial claims about Pirahã is that it cannot express numbers above two, or possibly above one. In other words, the Pirahã language cannot support the concept of counting. But I personally don't find the evidence for that particularly convincing—it seems more likely to me that the Pirahã people just haven't needed words for larger numbers in their day-to-day life. In other words, the Pirahã don't use words for larger numbers just because they haven't needed them, not because they can't comprehend them. – Draconis Dec 19 '19 at 18:37
  • While it's nice to see interesting discussions on this website, I'd like to ask you to take it to Chat to avoid long comment threads, which by the way can raise automatic flags. Thank you. – Alenanno Dec 20 '19 at 13:47
  • Comments are not for extended discussion; this conversation has been moved to chat. – Alenanno Dec 20 '19 at 13:47
4

Disclaimer: I'm not a linguist, I'm an evolutionary biologist. But I think there are some parallels.

The languages we have now haven't been around that long. I mean, the English that I write right now is already noticeably different from that of a century ago and even more so as we go further back. The trajectory that English, or any language, has been going through is not one of refinement or improvement (is my English "better" than Shakespeare's? Surely not) but simply of change.

In evolutionary biology it is more or less taken as a given that life forms don't somehow get better and better (or more and more fit, or more and more complex) over time. They just change, as do their surroundings. This is referred to as the Red Queen hypothesis, with reference to the part in Through the Looking Glass where the red queen says:

Now, here, you see, it takes all the running you can do, to keep in the same place.

I suspect something similar is the case in the evolution of languages: they keep changing to keep up with the changes around them (and changes in the circumstances of the speakers) without getting more or less expressive, or complex, or sophisticated.

user3484024
  • 41
  • 1
  • 1
    I think you're position would actually support the claim that languages would get more complex (or maybe more simple), because of increased specialization, the need for technical language, etc. Obviously, if there's evidence that there isn't any change, then that hypothesis is wrong, but I think your perspective could be used to argue a different conclusion. – awe lotta Dec 17 '19 at 23:33
  • 5
    @awelotta Evolution doesn't work that way. It's a way to create complexity out of simple things, but not an undisturbed climb towards complexity. The thing you're glossing over is that dropping things that aren't all that useful is also part of evolution. For example, English used to have "formal" and "informal" you (just like most languages), but it isn't used anymore (the "Thou" in the Bible is the informal). Czech used to have many different tenses, but today is mostly left with three or so. English used to be very much like German (of the day), but dropped many of the complexities. – Luaan Dec 18 '19 at 09:11
  • As perhaps an interesting corollary, ask yourself if Danish is better than Icelandic, considering that Icelandic has preserved a lot more of the features of Old Norse: or if English is better than German, considering the information in the preceding comment. – tripleee Jan 04 '20 at 09:24
2

There are badly-written documents that have been preserved, but no indication that any attested language was "simple." One amusing early paper talking about bad speakers of complex languages is Bloomfield, Leonard (1927), "Literate and Illiterate Speech".

Some ancient fragments are just lists of things, but we presume that they had a complicated syntax. You may appreciate Deacon's The Symbolic Species, which presents some research on human and animal communication systems, along with theories about when/why complex languages came about in humans.

bruno
  • 105
  • 1
2

Languages use words to mark the distinctions that the speakers care about. Present-day English speakers usually don't care about the difference between maternal and paternal aunts while Latin used to have different words.

On the other hand if you look into the sciences English is able to distinguish credence, odds, probability, likelihood with different words. A term like statistically significant that has a clear meaning in the scientific discourse won't have an equivalent in older languages that aren't used to talk about such concepts.

Christian
  • 788
  • 8
  • 22