Conlang Random Thread

[Imralu]

Whatever happened to the translation challenges? Are we not doing these anymore?

They've been much less frequent here for a while, even before the move. I head over to the CBB for translation challenges. They have a dedicated forum just for TCs, where both con- and natlangs are welcome.

There used to be the rather annoying habit of stacking them all into some kind of compendium of the phrase, with related languages next to each other, and the same ungrammatical German translation someone who doesn't know German for unknown reasons decided to provide gets copied for 15 posts before someone corrects it ... and then the next person coincidentally posts at about the same time and copies the post before the correction and the thread continues with the bad translation, and meanwhile no one has unbolded the Turkish translation that was added six posts ago, even though bold is only for new translations.

Thankfully, that's no longer the done thing. You just include the language(s) you're translating and people want to see glosses and pronunciation. Also, feel free to necro old translation threads!

[mèþru]

I would post TCs but my conlangs need more work before I can participate.

[Raphael]

*Loooong sigh*


This is a really weird thing to say on a conlang-centered board whose members basically have all been conlanging for a while, but, well, I think I might start to do some conlanging myself. (In case you're wondering, I'm here mainly as a kind of legacy from the original Virtual Verduria Message Board.)

Thing is, I have some conhistory lying around, but no one in it has any names, because I don't have any languages for it from which the names could be derived yet. So I'll need at least naming conlangs. And the conhistory in question involves people and places from several fictional cultures and linguistic communities, so I have to come up with several conlangs, preferably some related and some fairly distinct ones. I want the names and the rules they're derived from to be plausible, original, and preferably not too stereotypical or clichéd - no English/German/Latin imitation no. 789,346,340,436, no pseudo-Elvish, no "evil guttural" language, nothing completely familiar, but preferably nothing ultra-"exotic"-for-the-sake-of-being-ultra-"exotic" either. I don't want to mess up. And it's all very overwhelming.

Anyway, for a start, I bought myself the LCK. Guess I'll see how it goes.

Ok, as a first step, I've made a list of everything I'll need to fill my currently written conhistory with personal and place names. If I count even those languages in which I'll only need one or two names, I'll need 30 naming languages and up to 440 name elements, although I'll be able to reuse some of the personal names, so I'll be able to work with fewer than 440. Thankfully, for naming languages, the main task is phonology and some basic morphology. I guess I'll start by playing around with phono and gen.

Now I just have to actually do it.

[Salmoneus]

Do you really need that many languages? Even a history of the real world probably wouldn't feature names from more than a dozen or so languages, I should think.

[also, don't forget that in a realistic settting, a lot of the languages in your history will be descendents of earlier languages in your history - almost all of Europe is now covered by the descendents of only three languages, spoken less than two thousand years ago, all of which are themselves related.]

[Raphael]

Do you really need that many languages? Even a history of the real world probably wouldn't feature names from more than a dozen or so languages, I should think.

[also, don't forget that in a realistic settting, a lot of the languages in your history will be descendents of earlier languages in your history - almost all of Europe is now covered by the descendents of only three languages, spoken less than two thousand years ago, all of which are themselves related.]

Good question. The would-be conlangs from which the would-be personal and place names would be derived can be divided into three main groups:

1) The languages spoken in my main conculture's core area at the time the part of my conhistory that I've worked out so far starts. For now, I plan to have eight languages in that group. That might sound like a lot, but at the start of the parts of my conhistory that I've worked out so far, this core area contained a lot of recently liberated former slaves, who or whose ancestors had been enslaved in many different places. While for most of them, their ancestors had been enslaved for long enough to start speaking only their slaveholders' language, a significant minority had been enslaved recently enough to preserve their ancestors' languages, and, at least among themselves, use names from those languages.

2) The languages spoken in areas later conquered by my main conculture. For now, I plan to have nine languages in that group, most of which would be related, since they're geographically adjacent.

3) Languages of cultures that either had some kind of major political, cultural, or economic interactions with my main conculture, or were the ancestral homes of a significant number of immigrants in my main conculture, some of whose descendants rose to prominence. For now, I plan to have thirteen languages in that group, though now that you mention it, that might be overkill.

30 languages might sound like a lot, but I'm not sure if I could reduce the number of languages in each individual one of those three groups by too much while still staying plausible. I could, however, introduce some overlap between the first and the second group and the first and the third group.

[linguistcat]

Trying to figure out how to note tone changes in my language Nyango* since it's not exactly like Japanese pitch accent but I'm not sure it would be a normal tone language either. (Maybe I'm wrong on that latter point.) There can be up to two "accented" syllables in a word, 1 fall and 1 rise though they don't have to come in that order. I've been denoting a rise with <'> and a fall with <,> after the affected syllables, but I don't know if that's intuitive to just me or if there's a better way. Dunno what I'm going to do to denote contour tones when I have them.

Verbs are pretty simple accent wise. They are either unaccented within the root (and have a pitch fall after the root dependent on affixes), or they have a rise and fall dependent on the word's number of syllables. Verb+verb compounds take the accent of the last element. I haven't decided how noun+verb or noun+noun compounds will go, in part because nouns can vary much more in the accented syllables. But "verb-like" adjectives and "noun-like" adjectives can just follow the same rules as verbs and nouns respectively, so that makes things easy there. Adverbs that derive from adjectives might be a little funky, but that just makes things more interesting.

*working name

[Raholeun]

There can be up to two "accented" syllables in a word, 1 fall and 1 rise though they don't have to come in that order. I've been denoting a rise with <'> and a fall with <,> after the affected syllables, but I don't know if that's intuitive to just me or if there's a better way.

Why not mark the vowel with an accent grave or accent aigu?

[Vijay]

This is a really weird thing to say on a conlang-centered board whose members basically have all been conlanging for a while, but, well, I think I might start to do some conlanging myself.

Well, if it helps you feel any better, I've never made a conlang in the usual sense of the word, just Mountain Lion.

[Raphael]

This is a really weird thing to say on a conlang-centered board whose members basically have all been conlanging for a while, but, well, I think I might start to do some conlanging myself.

Well, if it helps you feel any better, I've never made a conlang in the usual sense of the word, just Mountain Lion.

Thank you, good to know! Now I'm curious: how is Mountain Lion different from a "regular" conlang?

[Salmoneus]

1) The languages spoken in my main conculture's core area at the time the part of my conhistory that I've worked out so far starts. For now, I plan to have eight languages in that group. That might sound like a lot, but at the start of the parts of my conhistory that I've worked out so far, this core area contained a lot of recently liberated former slaves, who or whose ancestors had been enslaved in many different places. While for most of them, their ancestors had been enslaved for long enough to start speaking only their slaveholders' language, a significant minority had been enslaved recently enough to preserve their ancestors' languages, and, at least among themselves, use names from those languages.

While I'm not saying this is impossible... has this ever happened in human history? The whole system of slavery is generally designed to assimilate cultures. And while I can imagine one slave culture surviving, once you mix up slaves from many different cultures, I'm not sure how those individual cultures could meaningfully survive. I suppose as one example, I think Efik has managed to survive in some areas of southern Cuba, due to mass importation of slaves from a single area?

2) The languages spoken in areas later conquered by my main conculture. For now, I plan to have nine languages in that group, most of which would be related, since they're geographically adjacent.

3) Languages of cultures that either had some kind of major political, cultural, or economic interactions with my main conculture, or were the ancestral homes of a significant number of immigrants in my main conculture, some of whose descendants rose to prominence. For now, I plan to have thirteen languages in that group, though now that you mention it, that might be overkill.

My broader point here is that most historical figures are mostly known through their names in major languages - either because what they did was famous in one language, or because historians wrote in one language. We talk about John Cabot and Christopher Columbus - not about Zuan Chabotto and Krishtoffa Kurungbu. Indeed, we don't even know what those men's "real" names might have been, as they left no record of them - in Cabot's case particularly, we don't even know what country he was from (Zuan was his Venetian name, but he's rumoured to have been Genoese originally). So a European history can be written without knowing anything about mediaeval Ligurian or Venetian dialects.

[Vijay]

This is a really weird thing to say on a conlang-centered board whose members basically have all been conlanging for a while, but, well, I think I might start to do some conlanging myself.

Well, if it helps you feel any better, I've never made a conlang in the usual sense of the word, just Mountain Lion.

Thank you, good to know! Now I'm curious: how is Mountain Lion different from a "regular" conlang?

Mountain Lion is more of a jokelang. It doesn't even have many words in it.

[zompist]

Do you really need that many languages? Even a history of the real world probably wouldn't feature names from more than a dozen or so languages, I should think.

As a counterpoint— I have 17 languages worked out for Almea, plus at least a dozen naming languages. I have names and a few words for about 200 languages. Yet I have entire continents that are basically faked up!

And sure, I'm crazy that way. But for Earth, an actually worldwide history would certainly refer to names or terms drawn from Latin, French, Portuguese, Spanish, Italian, English, German, Norse, Russian, Greek, Persian, Sanskrit, Hindi, Indonesian, Japanese, Chinese, Aztec, Quechua, something Algonquian, Arabic, Hebrew, Assyrian, Aramaic, Turkish, Mongol. That's 25. Plus you're certainly going to refer to a few place and people names from easily a dozen other countries.

A lot of the terms come to English indirectly, of course. E.g. we use Greek names for lots of places in the Middle East. But the Greeks had to get those names from somewhere!

I hasten to add, of course you're right that Raphael doesn't need 30 languages. You have to start somewhere, so just start with one language. Plus, you can always fake it! Add some exotic names to the map and worry about where they come from later.

I do advise keeping a wordlist for each potential language, even if it starts with just two or three words. You can try and keep the words consistent, and eventually get a mini-phonology and maybe some common derivations ('land', 'person'...).

(Also, all this is based on Raphael's statement that he wants languages for a conhistory. That's a great medium for showing off multiple languages. It'd be different if he was writing, say, a novel.)

[Raphael]

Ok, I've now reduced the number of would-be surviving slave languages from seven to four. I think four slave languages can be justified - there was a significant number of first- second- and third- generation slaves at the time of liberation.

I've also introduced some would-be linguistic family structure to my would-be languages. I now plan to have four language families, for each of which I will have to come up with a basic phonology and morphology for their proto-language. Additionally, there are two languages that are either language isolates or the only member of their respective family that "matters" in the conhistory I've written so far - I'll have to come up with a basic phonology and morphology for them, too. Finally, there are seven would-be "languages" that each contribute only one personal- or place name to the conhistory that I've written so far, so I'll probably be free to just make up random words in their case.

[Raphael]

A plausibility question: Would it be plausible to have "all-except-sky" as the etymological derivation for my conplanet's name?

[Znex]

A plausibility question: Would it be plausible to have "all-except-sky" as the etymological derivation for my conplanet's name?

Why not "land(-and-sea)"? Or why not "creation"? It seems a bit odd to have an excluding name if you already have conworld contrasts to the sky.

[akam chinjir]

Fwiw, Chinese 天下 world is (etymologically and in classical Chinese maybe actually) under the sky. As far as I know it's never been used as a word for a planet. (I'm most used to it referring specifically to the human world.)

[Raphael]

A plausibility question: Would it be plausible to have "all-except-sky" as the etymological derivation for my conplanet's name?

Why not "land(-and-sea)"?

I don't really want to use "land-and-sea", because IIRC that's already the etymological derivation for "Almea", and I don't want to look like I'm just copying the main conlang of this forum's owner.

Hm, what about "all-places"? Or "all-lands"?

Anyway, thank you for your replies, Znex and akam chinjr!

Another plausibility question: Would it be plausible to have different words for "child" as in "first generation descendant" and "child" as in "young person"?

[akam chinjir]

Hm, what about "all-places"? Or "all-lands"?

That makes sense to me, so long as your quantifier is a constituent of the noun phrase (which isn't the case in all languages). A nearby alternative would be to have a particular high number that's conventionally used to mean all. (Again I'm thinking of classical Chinese, this time 萬 ten thousand, myriad, all.)

Another plausibility question: Would it be plausible to have different words for "child" as in "first generation descendant" and "child" as in "young person"?

Yes. (My impression is that this is fairly common, but I can't remember where I got that idea. Anyway at some point something convinced me it was a mistake to assume that the same word must be used for both of these senses.)

[linguistcat]

There can be up to two "accented" syllables in a word, 1 fall and 1 rise though they don't have to come in that order. I've been denoting a rise with <'> and a fall with <,> after the affected syllables, but I don't know if that's intuitive to just me or if there's a better way.

Why not mark the vowel with an accent grave or accent aigu?

Mainly for ease of typing and search/replacing. But maybe when I make an official romanization scheme, I'll use them

[Raphael]

Newbie question: When coming up with a romanization scheme for a conlang, is there an easy and not-too-confusing way of indicating that a vowel is both accented and long?