Does language improve over time, or does it just change?

Question

Obviously none of the languages that existed 5000 years ago are still in use today. Rather, we use their descendants. Over time, the lexicon and grammar change and transform, subtly, so that the language slowly evolves into something very different.

But I'm wondering if any studies have ever been done regarding the nature of linguistic evolution.

Does language change at random, perpetually, never settling, or is it refining itself? Are modern languages in any non-arbitrary way superior to ancient languages, e.g. easier to write, more expressive, more understandable, more consistent?

Is linguistic evolution purposeful?

Answer 1

Superiority/inferiority is by definition arbitrary, at least in the realm of linguistics. Any language is ultimately able to convey any thought, and that's all that really matters. I can't really imagine being able to objectively judge ease-of-communication on individual points - there are so many subjective decisions that would have to go into setting up a standard to judge by that I doubt it's possible.

For your individual points:

If 'easier to write' means 'closer to a 1:1 grapheme:phoneme correspondence ratio', then typically languages become harder to write over time - if spellings are standardised, they may not change for centuries (e.g. Tibetan, which was standardised a millenium ago); while the language will continue to change as normal. Typically some form of standardisation is the default (unofficial though it may be - since writing is associated with prestige, and prestigious things tend to be conservative), so in order for a language to become 'easier to write', its speakers have to make a conscious decision to change the way they spell things. It's not a naturally-occurring process at all.

I don't know what 'more expressive' means. If you're a native speaker, you'll be able to express pretty much anything with extreme ease - and while there may be a construction here and there that's not that easy, there's almost certainly a good number of languages out there that struggle with something your language finds easy. If you're referring to the range of topics that can be discussed, that's more of a vocabulary question; and vocabulary can be changed and extended so easily that it really doesn't work as a standard of 'ease of communication' - you can just add new words then and there and then perfectly easily talk about whatever it was you couldn't have before.

I also don't really know what 'more understandable' means. If you're a native speaker, you're going to understand it! :P If you're not, how easy it is to learn depends on what your native language is. For example, Portuguese is pretty easy for Spanish speakers to learn, harder for English speakers, and very difficult for Mandarin speakers - but Cantonese is relatively easy for Mandarin speakers, and much harder for English and Spanish speakers. Something like Burushaski or Xhosa or Nuxálk would be quite difficult for all of them - but then, Xhosa would be relatively easy for Zulu speakers. It's a question of similarity - if a language is similar to your native language in terms of grammatical structure (and vocabulary helps too), it'll be easy to learn. If it's structured very differently, it'll be harder to learn.

Clearly, languages are not becoming more consistent (by which I assume you mean 'grammatically regular') - see Modern English, for example! Not that earlier stages were any -more- regular, but they weren't really -less- regular, either. Languages in general tend to maintain the same general level of regularity over time (with some exceptions, and different languages may differ greatly in regularity). The reason for this is that the two kinds of changes that languages undergo - sound change and grammatical change - have opposing effects on regularity. Sound changes create irregularity (as words get worn down, individual morphemes become harder to identify and harder to separate out), while grammatical changes typically create regularity (as individual words' inflection patterns are modified to better match more general patterns). These opposing forces tend to create a kind of equilibrium - sound changes make things more irregular, but once they get too irregular then grammatical change kicks in and makes them more regular again. And if you move on from grammatical regularity to other kinds of regularity (semantic, phonological, etc), you'll come across other issues - phonological patterns are extremely regular, though they might be horrendously complicated; while semantics, being more a 'space' of meanings that are arbitrarily divided, doesn't have very much in the way of 'patterns' at all outside of a few domains.

So no, languages are not improving, because the term 'improving' is meaningless. Even if you agree that things like grammatical consistency and phonemic spelling are valuable, languages aren't improving. (And you have no real reason to value those things! Things like grammatical inconsistencies and irregular spellings are often windows into the history of a language - we can learn about how things used to be from the weird half-patterns they've left behind. Yes, they make a language harder to learn, but there's more to value in a language than how easy it is to learn.)

Answer 2

English spelling did deteriorate, due to almost abruptive historical changes. It will never become phonetic (nation versus national), and we won't see much reform there besides through/thru. So evolution is not an automatically proceding process.

We live with a much richer language in qualitative terms. Philosophics showed the historical introduction of a Greek notion for (non-religious) soul (Latin animus). So nowadays we have a bag of notions like creative, tolerance, open-minded. Nevertheless philosophy, or psychology did contribute nothing to our linguistic mental toolset describing mental processes themselves. We did not become better argumentators. And are still as gullible to wrong arguments (financially supporting child raising, despite world over-population; energy saving lamps, despite very toxic waste). Despite better expressiveness. So the meta-level of language did not evolve.

It is interesting in this respect to look at the artificial language Esperanto. (I am Esperantist and anglophile.) The language can be seen to correct many of the grown short-commings of evolution: instead of the many suffixes to make a word adjective (-al, -ic, -ive) or noun there is just one -a or -o. The spelling is again phonetic. Rules are few and without exception. A linguistic evolution in general has no such opportunity to make a clean break. Modern Hebrew was such a forced effort and might be worth readding upon.

In general language evolution is a process of growth, friction and fitting. Missing are aspects of the biological evolution that made so good survivors: hunt, competition for the same resources, colaboration, genetic reproduction. Hence language evolution is nothing to be overly optimistic about. Especially as there is a large qualitative equality between languages. Historical or otherwise.

Answer 3

What's interesting in this context is that the original historical linguists were driven by exactly the opposite motivation. They were convinced that languages were deteriorating over time and their mission was to record the perfect state (and if possible to halt the decline). Labov describes it in his Principles of Linguistic Change

Answer 4

"Are modern languages in any non-arbitrary way superior to ancient languages, e.g. easier to write, more expressive, more understandable, more consistent?"

I can't address that last two parts of your question, for two reasons. First, I know of no instance in the speakers of language A have more trouble understanding each other than the speakers of language B owing to some grammatical inadequacies of language A. Second, the inconsistencies of a language are not a problem for the language's native speakers. Most people can use even rare and complex language constructions in their native tongues by the time they are five years old. See this article: https://courses.cit.cornell.edu/psych215/PDF-files/language-acq-II.pdf

As for ease of writing, there doesn't seem to be a lot of evidence to support the notion that one type of orthography (say, logographic vs. alphabetic) is more easily written than another type. See this post on the Linguist List: http://linguistlist.org/issues/2/2-322.html. Ease of learning may be another matter, but I haven't found information about this.

As for whether ancient languages are inferior to modern languages in any non-trivial way, we have no way of knowing what the most ancient languages were like.

Our species is about 195,000 years old. See the Wikipedia article on homo sapiens: https://en.wikipedia.org/wiki/Homo_sapiens. Writing systems were devised only within the last 6,000 years. See the Wikipedia article on the History of Writing: https://en.wikipedia.org/wiki/History_of_writing#Inventions_of_writing). So, on the evolutionary time scale, so-called "ancient" languages like the first language to be written, Sumerian, might just as well be modern.

We don't even know for sure whether language first developed in our species or among our pre-homo sapiens ancestors! Some anthropologists believe that the origin of modern language happened about 50,000 years ago, when humans first left traces of "behavioral modernity" (manufacture of keen-edged stone tools, use of pigment for art, burial of dead, and other behaviors). See the Wikipedia article on behavioral modernity: https://en.wikipedia.org/wiki/Behavioral_modernity However, other experts believe that "behavioral modernity" began long before 50,000 years ago, and that the evolution of language may have occurred before the human line evolved into anatomically modern humans. See this paper by Francesco D'Errico and a number of others: http://www.academia.edu/290586/Archaeological_Evidence_for_the_Emergence_of_Language_Symbolism_and_Music_an_Alternative_Multidisciplinary_Perspective

(Disclaimer: I haven't slogged all the way through this last paper: I concentrated on the abstract and what I gleaned by using the finder to locate the word "language" in various places in the text.)

So while it makes sense to assume that the most ancient natural languages were less sophisticated than the languages of modern humans, we have no way of knowing what those first, rudimentary languages were like, or how they became as complex and expressive as every language that linguists have studied.

Answer 5

Language can "improve" over time as well as "just changing".

This is not to say that each change is an improvement or that improvements never come with negative side effects or that improvements must necessarily grow to outnumber negative properties. This also does not negate the fact than any concept can be expressed in any language.

Languages actually can adapt innovations either invented by a speaker or observed in use in another language.

The obvious is example is the acquisition of a writing system and literacy. Once a language can be written it is possible to communicate over greater distances are timespans. Few people would argue that any language was not improved when it gained a way to be written.

Most of the world's languages either have no writing system, only gained a writing system in the 20th century, or have small literate populations despite governments or missionaries devising an orthography. Yet which seldom-written languages are prominent world or national languages?

Another example would be borrowing, calquing, or devising words for new concepts. Whether a language has an academy or not, speaker populations will begin referring to new concepts such as horses, guns, cars, satellites, sushi, computers, e-mail, etc.

Now could it possible for languages to acquire other improvements such as grammar and syntax? I seem to recall reading somewhere once that features such as relative clauses or some forms of recursion were not always universal but were passed from language to language much like other good ideas. I find this possibility very tantalizing and would like to read any references about it. Please add a comment (-:

Answer 6

If by improve you mean, easier for adult learners, then a large adult learner population can cause such changes. According to "Language Structure Is Partly Determined by Social Structure" by Gary Lupyan and Rick Dale

This also suggests that whatever the reason for the change, implies some criteria according to which it "improved".

examples (note how some of these are "against" each other):

• easier for adult learners (morphological simplification ie becoming isolating, regularity)

• easier for infant learners (morphological complexity causing redundancy to be communicated, but regularity eg children say/expect "gived" not "gave")

• faster/shorter speech and less likely to mix up (irregular conjugation, change of word endings, phonological change)

• foster national/tribal identity by changing differently than sybling languages (some think of the linguistic variety of Europe as "improvement" over proto-indo-european or whatever, similarly all romance languages are seen by some as better than 1 standard latin)

The first 2 are taken from "Language Structure Is Partly Determined by Social Structure" by Gary Lupyan and Rick Dale, the rest I have probably read somewhere, but I do not remember.