SoylentNews is people
SoylentNews
English is currently one of the dominant languages on the planet due to the spread of the US and UK empires in the last century. With the rise of technology English may be made redundant with the advent of automatic language translation.
Just waiting for made up languages to become the norm (e.g. Esperanto), or hyper language learning.
Now ponder, as Douglas Hofstadter did, translating Lewis Carroll's Jabberwocky from English into French, German, and Russian (Cyrillic .GIF) or (ASCII transliteration).
This discussion has been archived.
No new comments can be posted.
English is King but Technology May Trump All
|
Log In/Create an Account
| Top
| 78 comments
| Search Discussion
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Life is to you a dashing and bold adventure.
(Score: 4, Interesting) by Arik on Monday January 28 2019, @04:21AM (6 children)
Yeah I know I don't write it right I don't have some of those keys on my keyboard.
I barely remember a word of it. It has a lot of problems but it's still the most widely spoken artificial language, right?
Or did Klingon pass it?
Anyhow, one idea with Esperanto was that it would be easy for to learn because it's (relatively) simple, it's (almost perfectly) regular, and the vocabulary is already widely known to speakers of other languages.
Well 1 is true, 2 is mostly true, 3 is sort of true. It sort-of-approximates a hypothetical 'standard European' and through that the vocabulary has inroads through for instance technical terminology worldwide, but most of it is still radically alien to speakers of many non European languages; the grammar even more so, simple as it is.
Still, on the grounds of generally easy to learn, it would probably be hard to be beat.
I think there are other grounds that should be considered, though. It's not the most expressive language, in fact IIRC I found it downright maddening how cumbersome it could be when I was really trying to work with it. It sounds a lot like Spanish or Italian but in those languages you can still pack a lot of punch into a single word; ¡dámelo! for example, powerful word. "(You) give me that!" Takes at least three words in English (unless you resort to baby-speak and say "gimme") and I don't think Esperanto can do any better. If I designed it, I think Esperanto would have a direct object pronoun that affixed to the verb.
In fact I probably wouldn't stop there. I like features that let you pack more data into shorter sentences, I find them aesthetically pleasing. But I'd try to hold myself back from going too wild with them, because it's important to preserve enough redundancy to communicate quickly over a poor quality link or over the top of loud noises now and then. There's a trade-off to be considered as long as that's a possible issue.
You can gain maximum room for meaningful distinctions by carefully utilizing your phonetic map, but there are limits. Heinlein's fictional artificial language (anyone remember what it was called?) would not actually work very well IRL because of that - mishearing a single sound inside a sentence in a natural language is likely to result in nonsense, and the subconscious is also likely to simply correct it without the conscious mind even being aware of it as a result. But in an artificial language that wrings just as much meaning out of each syllable as possible, a similar failure of hearing would be much more likely to make sense on the face of it, and thus much less likely to be detected or corrected or anyway. Heinlein's super language would simply be too frail for the real world - it would break constantly outside of a lab environment with proper headphones and microphones.
So in computing terms you do need some redundancy, checksums and the like. You can't squeeze it down infinitely. But you could squeeze it down quite a bit without causing problems if you did it carefully.
One way to do this is give all your phonemes plenty of space. The fewer phonemes you have, generally speaking, the less chance for them to be misheard. But that's not really correct. What matters is how much space you leave between them. If you have D do you really need T too? Well, maybe, but I'd pick one of them and go on through the series and only add the other later if I really don't have enough phonemes with the sparser set. The more distinctions you eliminate the more distinct the phonemes you have left actually become in practice.
There are epi-phonemic considerations too, and they play a role in many of the same calculations. For an example, t and th (the latter referring here to an aspirated t, not a fricative) are separate phonemes in many languages, but in English they are regular alterations of the same phoneme. We use the aspirated version at the beginning of a word, the other version anywhere else, barring particularly vehement speech in which case you might hear the aspirate in the middle of a word as well. The advantage of having them as different phonemes is a whole bunch more possible syllables, and thus potentially more words with fewer syllables. The advantage to doing what English does here is that it's one of those checksum-like-thingies. The vehemence with which we spit the initial consonants of words makes it easier for the listener to correctly split the sounds they hear back up into the words we intended to speak.
But as tempted as I might be to exploit a bunch of similar devices to cram more meaning into each syllable while avoiding excessive frailness, doing so might make the language much harder to learn. If we stick to just the aspiration example above - adopting the English usage would be easy for English speakers and speakers of other languages that do the same thing, but not really for everyone else. If I cram a dozen similar features in then the language might work exceptionally well - but no one would think it was particularly easy to learn anymore.
What's the current estimate of Basic Vocabulary? And how far off is it? That's what you would have to come to grips with to really map out the best balance of the phonemes - you first figure out how many distinct Basic words you need, it's going to be significantly more than Basic Vocabulary if you want to have sounds left over for fancy stuff like suffixes of course, but that's where you start. Then you figure out how many phonemes you need to fill your needs there in the minimum of syllables; 3? maybe you could squeeze it down to 2, but doing that means more phonemes you have to fit in the same space.
Anyone else actually care about this stuff?
Shalom
If laughter is the best medicine, who are the best doctors?
(Score: 0) by Anonymous Coward on Monday January 28 2019, @01:13PM
> It sounds a lot like Spanish or Italian but in those languages you can still pack a lot of punch into a single word; ¡dámelo! for example, powerful word. "(You) give me that!" Takes at least three words in English (unless you resort to baby-speak and say "gimme") and I don't think Esperanto can do any better.
Chinese can pack quite some things into a character. Do you want that? Probably not.
Anyway, “Donu!” should likely suffice given context, I think. At least that's how we use similar things in Russian (“Дай!”, “Дай сюда!”, “Отдай!” depending on what you mean) and it works just fine.
(Score: 1) by aixylinux on Monday January 28 2019, @03:34PM (1 child)
(Score: 3, Interesting) by Arik on Monday January 28 2019, @11:14PM
That doesn't sound right so I did some searching. Unfortunately I can't find the book, I may not have a copy here. From the web it seems likely I am thinking of "Gulf." He may well have used Loglan elsewhere.
At any rate the AL he described went to extremes to maximise phonemic inventory and use it all, so as to cram the largest vocabulary possible into the shortest utterences. So the phonemes approximated IPA, the language wouldn't just have a T, or a T and a D, but also aspirated AND glottal AND fricative AND affricative version of each, and so on. There's not just a click but every sort of click found in any natural language, or deducible from those, a glottal click and a dental click and a palatal click and so forth. It's hard to say just how many phonemes you would wind up with this method - the IPA claims 107 "letters" but it also has 56 additional characters that are used in combination, so the total number is vast.
By comparison Japanese has about 5 vowels and 14 consonants for a total of 29; English has about 15 vowels and 24 consonants for a total 39; in comparison to well over 100. With 5 vowels and 14 consonants, assuming all syllables are CV, you have 70 simple syllables to work with. With English phonemes, but still assuming that simple syllable structure, you'd get 15*24=360 syllables to use. But with, what, somewhere around 80 consonants and 40 vowels (a conservative guess) you'd wind up with 80*40=3200 syllables instead!
You could talk very fast if you learned a language like that, the big problem would be could you *listen* fast enough to understand it though. It's very easy to mistake each of these sounds for some of their close relatives. In most languages, the subconscious can narrow the possibilities down very quickly because only a few (or only one!) is actually possible. You thought you heard an aspirated 't' instead of the other one? Doesn't matter, they both mean the same here. You thought you heard a dental sibilant? Might be a speech defect, there aren't any words with an 's' there but if we put in a 't' it makes sense. You mind does this very quickly in the background so you rarely have to think of it. But if EVERY sound is used and EVERY possible syllable is used there's just no room for that sort of correction anymore. Oh, and if you misspeak - same thing. You can fumble over most of the sounds in a natural language and people can still easily understand you, because the sounds you're making aren't actual words, but they're close to something that is. With the phonemic inventory maxed out like this, that's no longer true - every time you misspoke the change would produce a valid word - just one with a different meaning from what you intended.
If laughter is the best medicine, who are the best doctors?
(Score: 2) by hendrikboom on Wednesday January 30 2019, @03:29PM (2 children)
The Esperanto I learned back in the 60's didn't have upside-down question and exclamation marks.
Is that another new development in the language? With new punctuation and a new particle and the new words that it takes on from other languages, that's a clear hint that the language is evolving, and perhaps becoming natural.
(Score: 2) by Arik on Wednesday January 30 2019, @11:22PM (1 child)
I'm nowhere near current on Esperanto but I did recent skim something on it, and there are several innovations that are adopted by some users, but I was just having fun with something I vaguely remembered from years ago.
If laughter is the best medicine, who are the best doctors?
(Score: 2) by hendrikboom on Wednesday February 06 2019, @01:56AM
Yes, I like the upside-down question marks too. Pity they're hard to type on some of my devices.