Completely unbiased auxlang

A forum for all topics related to constructed languages
Squall
greek
greek
Posts: 578
Joined: 28 Nov 2013 14:47

Re: Completely unbiased auxlang

Post by Squall » 13 Nov 2014 23:53

All right, there are no phoneme inventories that are perfect for all languages.

A new approach would have to be done. If we exclude languages that have less than 10 million speakers and we allow at most one unfamiliar phoneme for each language, we may be able to create an easy auxlang.
Fanael wrote:
Squall wrote:If the auxlang had /ɺ/ without /l r/, the learners would use /r/ or /l/ without conflicting phonemes.
I always found this argument rather weak. I can imagine that being a problem for learners whose L1 distinguishes /r/ and /l/, who'd be looking for a distinction that isn't there.
If the L1 lacks /r/ or /l/, the speaker will use the one that is present in their language.
If the L1 has /r/ and /l/, the speaker will learn that the word can be pronounced either /r/ or /l/.

The same idea can be used to choose plosive phonemes.
Mandarin: p pʰ
English: b pʰ
Italian: b p
A single phoneme /p~pʰ/ would be the solution. The English speaker will learn that some foreigners can pronounce this phoneme as /b/.
Last edited by Squall on 14 Nov 2014 01:17, edited 1 time in total.
English is not my native language. Sorry for any mistakes or lack of knowledge when I discuss this language.
:bra: :mrgreen: | :uk: [:D] | :esp: [:)] | :epo: [:|] | :lat: [:S] | :jpn: [:'(]

User avatar
Creyeditor
mongolian
mongolian
Posts: 4517
Joined: 14 Aug 2012 19:32

Re: Completely unbiased auxlang

Post by Creyeditor » 14 Nov 2014 00:35

Indonesian wouldn't be excluded that way, it has ~23 million speakers and allows no unfamiliar phoneme. Also arabic, so no p/f-distinction.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]

Squall
greek
greek
Posts: 578
Joined: 28 Nov 2013 14:47

Re: Completely unbiased auxlang

Post by Squall » 14 Nov 2014 01:25

I mean that each language would have to learn at most one unfamiliar phoneme.

Arabic has /b/, it would replace /p/.

Does /f~ɸ/ sound like /p/ for Indonesian and Tagalog speakers?
If /f/ were long (/ff/), the air blow could be noticed, but the speakers of the auxlang would make /f/ short.
English is not my native language. Sorry for any mistakes or lack of knowledge when I discuss this language.
:bra: :mrgreen: | :uk: [:D] | :esp: [:)] | :epo: [:|] | :lat: [:S] | :jpn: [:'(]

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 14 Nov 2014 02:54

Squall wrote:All right, there are no phoneme inventories that are perfect for all languages.

A new approach would have to be done. If we exclude languages that have less than 10 million speakers and we allow at most one unfamiliar phoneme for each language, we may be able to create an easy auxlang.
If this is going to be written in the Roman alphabet (oh my God, the bias! the exclusion!), why not just use that. Kill off <c>, <q>, <x>, <y>, and don't bring them back in to represent something else. If you really can't let the l/r thing go, bye-bye <l>. <j> is /j/. No digraphs. Sans <l>, that'd leave:

a, b, d, e, f, g, h, i, j, k, m, n, o, p, r, s, t, u, v, w, z

A Japanese-y, Spanishy middle-of-the-road sized inventory that you could bring home to meet your mom (Nix one more, and you're down to a nice round 20, which you could hype in your sales pitch). Allow for plenty of allophony (which is what natlangs do anyway, it's called speaking Language X with an accent). Put the IPA away; many of the people you're supposedly trying to reach won't be familiar with it. And note that making choices involves exclusion. If one thinks that rises to the level of disenfranchisement, well then, good luck with this.
道可道,非常道
名可名,非常名

Fanael
greek
greek
Posts: 475
Joined: 19 Jul 2012 21:26

Re: Completely unbiased auxlang

Post by Fanael » 14 Nov 2014 03:07

Squall wrote:If the L1 has /r/ and /l/, the speaker will learn that the word can be pronounced either /r/ or /l/.
As I said, I have a feeling the speaker would be looking for a distinction that isn't there and would assume that when somebody whose L1 has [r] but not [l] and somebody else whose L1 has [l] but no [r] say the same word, they're actually different words. Of course recognizing [l] and [r] are in free variation isn't something that can't be learned, but it does require non-zero learning effort, and given that many English natives seemingly can't fathom the idea that a language prefers monophthongs to diphthongs, I can imagine /l~r/ causing a problem too.

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 14 Nov 2014 04:11

Squall wrote:The English speaker will learn that some foreigners can pronounce this phoneme as /b/.
Fanael wrote:... (G)iven that many English natives seemingly can't fathom the idea that a language prefers monophthongs to diphthongs, I can imagine /l~r/ causing a problem too.
If this is meant to be an exercise in unbiased-ness, why are theoretical English speakers framing the discussion? :wat:
道可道,非常道
名可名,非常名

Fanael
greek
greek
Posts: 475
Joined: 19 Jul 2012 21:26

Re: Completely unbiased auxlang

Post by Fanael » 14 Nov 2014 05:23

Lao Kou wrote:why are theoretical English speakers framing the discussion?
Because English is a natlang, so a completely unbiased auxlang should be as easy/difficult for natives of English as it is for natives of Iraqw, Ogami or Basque. Really, it was just an example.

cntrational
roman
roman
Posts: 953
Joined: 05 Nov 2012 03:59

Re: Completely unbiased auxlang

Post by cntrational » 14 Nov 2014 05:50

Fanael wrote:
Lao Kou wrote:why are theoretical English speakers framing the discussion?
Because English is a natlang, so a completely unbiased auxlang should be as easy/difficult for natives of English as it is for natives of Iraqw, Ogami or Basque. Really, it was just an example.
IMO, this is a wrong track to take -- Ogami speakers can understand Japanese, and Basque speakers can understand Spanish. Maybe the Iraqw wouldn't understand anything else, but a good number probably speak Swahili.

You shouldn't base your auxlang off languages, you should base it off people -- i.e., a good auxlang will be neutral in respect to commonly spoken languages, not every single damn language in the world. You can simplify things considerably by going that route.

I would use the following languages primarily: English, Spanish, Mandarin, Swahili, Arabic, and Russian.
Use the following languages secondarily: Hindi-Urdu, Bengali, Indonesian, Japanese, German, Portuguese, and French.

You could add more languages, but this is based on the idea that the "secondary" languages will have users who either speak one of the primary languages, speak something similar, or have notable speaker populations that are nonetheless not as widespread.

Indians who speak Hindi-Urdu or Bengali will almost certainly know English if they wish to bother with an auxlang (sure, rural guys won't be speaking as much English, but are those really the type of people who need an auxlang the most?). Portuguese is similar to Spanish, as is French to a lesser degree. Japanese is not widespread, and is heavily influenced by English. German and Indonesian are also fairly local, but have enough speakers to warrent notice.

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 14 Nov 2014 06:11

Fanael wrote:Because English is a natlang, so a completely unbiased auxlang should be as easy/difficult for natives of English as it is for natives of Iraqw, Ogami or Basque. Really, it was just an example.
Yes, and you should probably take most of what I say as tongue in cheek. Still, English was dealt a pretty good hand as phonologies go, so make the <r> a /ɾ/, there's your one "foreign" sound, and call it a day. Any accommodation for languages with smaller inventories will most likely involve taking away sounds English already has. Unfair, perhaps, but that's just the way it is. Turn up the volume on difficulty in some other area of language.
道可道,非常道
名可名,非常名

Fanael
greek
greek
Posts: 475
Joined: 19 Jul 2012 21:26

Re: Completely unbiased auxlang

Post by Fanael » 14 Nov 2014 18:46

cntrational wrote:a good auxlang will be neutral in respect to commonly spoken languages, not every single damn language in the world
It seems you miss the point of this thread: it's not to create a practical auxlang (besides, we already have English), it's to create a completely unbiased auxlang. And the original point is that a completely unbiased auxlang can't exist.
cntrational wrote:sure, rural guys won't be speaking as much English, but are those really the type of people who need an auxlang the most?
Of course not. If they needed an auxlang, they'd learn English. But these people still exist, so you have to keep them in mind while creating a completely unbiased auxlang.

In all honesty, if I were to create a completely unbiased auxlang, I'd probably go the opposite way: instead of trying to make it equally easy for everyone, I'd make it equally impenetrable for everyone.

Serena
sinic
sinic
Posts: 273
Joined: 26 Sep 2013 15:58

Re: Completely unbiased auxlang

Post by Serena » 14 Nov 2014 21:35

I felt like I should pass by and say something about this. In my honest and subjective opinion, this thread is not only misleading, but also kind of ridiculous. I know the original poster meant it to be a joke, (so, please, don't bash me), but since people took it seriously and actually tried to develop a completely unbiased auxlang, I will give my opinion.

You (almost all) completely mixed up biasedness and variety, which are two clearly distinguishable concepts.

1) Is it unfairly biased against Chinese people to have a sound that Chinese doesn't have? Absolutely not. Every single phonology is 100% unbiased if you chose it just because you wanted to, and 100% biased if you chose it because you, as an individual, are comfortable with it.

2) You can not look at a phonology and say: "Ow, that's really biased!", because biasedness is not a property of the product itself, it is a property of the production process.

3) Why do people keep worrying about phonologies at all? Grammar is where the real biasedness lies.

User avatar
Thrice Xandvii
mongolian
mongolian
Posts: 3722
Joined: 25 Nov 2012 10:13
Location: Carnassus

Re: Completely unbiased auxlang

Post by Thrice Xandvii » 14 Nov 2014 23:37

Can't we all just agree that it is 100% impossible to create anything, much less an entire language, free from bias? That being said, that doesn't mean give up. But the goal of creating anything free of bias is like trying to convince an idiot that 1 and .99999999... are equal! Or that an equation with a limit "if you go far enough" will reach the limit!

It can't be done since everyone everywhere exists in a unique environment that instills certain inherent biases.

So, the goal should be to reduce bias, not eliminate them. Also, Serena makes a good point though, that grammar is where the real demons lie, phonology can usually be surmounted well enough to allow folks to understand one another, it'll just have an accent to it, or sound a bit off. That's really a very minor thing if you can get around the far thornier part of getting on the same page with word order and cases (etc., ad nauseum)!
Image

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 15 Nov 2014 04:28

Serena wrote:1) Is it unfairly biased against Chinese people to have a sound that Chinese doesn't have? Absolutely not. Every single phonology is 100% unbiased if you chose it just because you wanted to, and 100% biased if you chose it because you, as an individual, are comfortable with it.
How do you convince people that one isn't the other? I think people are trying to avoid sentiments like, "Isn't it just like the Man to make me have to say a /θ/!" If a majority of the world's speakers supposedly have difficulty producing, say, [θ], including it is taken by many auxlangers as a gigundous "F*ck you!" to would-be learners.
2) You can not look at a phonology and say: "Ow, that's really biased!", because biasedness is not a property of the product itself, it is a property of the production process.
I think you're playing with the pragmatics here. If someone looks at a phonology and says, "Ow, that's really biased!", what they mean is the production process was biased. No, there is nothing inherently biased about sounds an sich, but if an auxlang has Czech <ř>, Georgian-style clusters, and contour tones, "Ow, that's really biased!" means "Wow, how did you get here?!"

As you've noted, you could assemble a bunch of sounds for an auxlang (use a lotto machine to generate them, for all I care) and say, "C'mon, sh*t or get off the pot. Let's learn this!". Even if you get a sizeable group to say, "Fine, I'll learn your damn auxlang!", that's not enough. Many auxlangers don't just want you speak their auxlang, they want you to want to speak their auxlang. You can't spell "damn" with the letters of "kumbaya".
3) Why do people keep worrying about phonologies at all? Grammar is where the real biasedness lies.
I would guess it's because phonology is a more tangible aspect of language. You can ask a Japanese to say "lalapalooza" and get immediate results; you can bean-count how many languages of record have /θ/; you can measure vocal chord pulsations; you yourself can try your hand at [ʘ͡qʼ] and see in the mirror if it looks like you're having a stroke; and working with phonology, you can feel like you're actually doing something. Far easier to point at [ɸʷ] as the downfall of an auxlang than to discuss how to get learners beyond the chronic-advanced-beginner stage, how to balance the appeal to good will and altruism with on-the-ground self-interest, how to convince a Michelle Obama to learn it and make it "sexy", how to get conlangers to stop tinkering, and how to make an auxlang, if it hails from the First World (and who can afford to armchair auxlang?), seem less like cultural imperialism with a twist (and I suspect that's more meta than /θ/, SVO, or case endings) and within the First World, less like tilting at windmills.

I'm still rather enamoured of:

[xP]
道可道,非常道
名可名,非常名

User avatar
Micamo
MVP
MVP
Posts: 6996
Joined: 05 Sep 2010 19:48
Contact:

Re: Completely unbiased auxlang

Post by Micamo » 15 Nov 2014 04:42

XXXVII wrote:Can't we all just agree that it is 100% impossible to create anything, much less an entire language, free from bias? That being said, that doesn't mean give up. But the goal of creating anything free of bias is like trying to convince an idiot that 1 and .99999999... are equal! Or that an equation with a limit "if you go far enough" will reach the limit!

It can't be done since everyone everywhere exists in a unique environment that instills certain inherent biases.

So, the goal should be to reduce bias, not eliminate them. Also, Serena makes a good point though, that grammar is where the real demons lie, phonology can usually be surmounted well enough to allow folks to understand one another, it'll just have an accent to it, or sound a bit off. That's really a very minor thing if you can get around the far thornier part of getting on the same page with word order and cases (etc., ad nauseum)!
It's because supposed cultural neutrality is the only advantage an artificial auxlang can claim to have over something like Basic English. If auxlangers gave that up, then they'd have nothing left to design towards.
My pronouns are <xe> [ziː] / <xym> [zɪm] / <xys> [zɪz]

My shitty twitter

Tanni
greek
greek
Posts: 808
Joined: 12 Aug 2010 02:05

Re: Completely unbiased auxlang

Post by Tanni » 15 Nov 2014 13:13

Serena wrote:2) You can not look at a phonology and say: "Ow, that's really biased!", because biasedness is not a property of the product itself, it is a property of the production process.
It results from the cultural background of the speakers who use it, so it results from the context in which it is used.
Serena wrote:3) Why do people keep worrying about phonologies at all? Grammar is where the real biasedness lies.
Moreso the lexicon.

I just wanted to propose that seperating auxlanging form making a culturally unbiased language. But after that
Micamo wrote:It's because supposed cultural neutrality is the only advantage an artificial auxlang can claim to have over something like Basic English. If auxlangers gave that up, then they'd have nothing left to design towards.
it might not be a good idea. But regardless of that, there is no language without culture.
Last edited by Tanni on 15 Nov 2014 13:22, edited 1 time in total.
My neurochemistry has fucked my impulse control, now I'm diagnosed OOD = oppositional opinion disorder, one of the most deadly diseases in totalitarian states, but can be cured in the free world.

Squall
greek
greek
Posts: 578
Joined: 28 Nov 2013 14:47

Re: Completely unbiased auxlang

Post by Squall » 15 Nov 2014 15:16

Lao Kou wrote:If this is going to be written in the Roman alphabet (oh my God, the bias! the exclusion!), why not just use that.
What about Cyrillic or Greek? Those are easy alphabets.
Roman is the most common alphabet, therefore it is the less biased one. Cyrillic would have Slavic bias.

A new alphabet could be created for the auxlang.
Fanael wrote:It seems you miss the point of this thread: it's not to create a practical auxlang (besides, we already have English), it's to create a completely unbiased auxlang.
The use of natural language is unfair, because the native speakers do not have the effort of learning a foreign language. The worst problem is that the native speakers control the evolution of the language. It is not good for an international language.
Serena wrote:1) Is it unfairly biased against Chinese people to have a sound that Chinese doesn't have? Absolutely not. Every single phonology is 100% unbiased if you chose it just because you wanted to, and 100% biased if you chose it because you, as an individual, are comfortable with it.
The phonology is unbiased when the author does not check any natlang. But it may be biased because it is likely to be based on the languages that the author knows.
2) You can not look at a phonology and say: "Ow, that's really biased!", because biasedness is not a property of the product itself, it is a property of the production process.
Yes, but the question is: how difficult is to learn the language? An auxlang must be easy for as much people as possible. Therefore, its features must be carefully chosen.
3) Why do people keep worrying about phonologies at all? Grammar is where the real biasedness lies.
Both things are important, but many basic sentences are easy without grammar, for instance, "want.INF 1SG water" can only be interpreted as "I want some water".
XXXVII wrote:So, the goal should be to reduce bias, not eliminate them.
Yes
English is not my native language. Sorry for any mistakes or lack of knowledge when I discuss this language.
:bra: :mrgreen: | :uk: [:D] | :esp: [:)] | :epo: [:|] | :lat: [:S] | :jpn: [:'(]

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 15 Nov 2014 15:53

Squall wrote:Roman is the most common alphabet, therefore it is the less biased one.
By this reasoning, English is the most common lingua franca, therefore it is the less biased one, and I doubt that's your position. What makes a common alphabet less biased but a common natlang more biased?
A new alphabet could be created for the auxlang.
Well sure, but how does this fit in with:
An auxlang must be easy for as much people as possible.
Therefore, its features must be carefully chosen.
By whom? And how are things "carefully chosen" if:
The phonology is unbiased when the author does not check any natlang.
The use of natural language is unfair, because the native speakers do not have the effort of learning a foreign language. The worst problem is that the native speakers control the evolution of the language. It is not good for an international language.
Who will "control the evolution" of the auxlang? Who will control the fairness (can't imagine English words creeping into the auxlang will be deemed "fair")?
道可道,非常道
名可名,非常名

Squall
greek
greek
Posts: 578
Joined: 28 Nov 2013 14:47

Re: Completely unbiased auxlang

Post by Squall » 15 Nov 2014 17:18

Lao Kou wrote:
Squall wrote:Roman is the most common alphabet, therefore it is the less biased one.
By this reasoning, English is the most common lingua franca, therefore it is the less biased one, and I doubt that's your position. What makes a common alphabet less biased but a common natlang more biased?
English is not a component of other natlangs, but the Roman alphabet is. Speakers of languages written with the Roman alphabet will not say that the alphabet is foreign.
Lao Kou wrote:
A new alphabet could be created for the auxlang.
Well sure, but how does this fit in with:
An auxlang must be easy for as much people as possible.
An auxlang with 10 consonants and 4 vowels would have only 14 symbols to learn. It is easy to learn symbols that are few strokes, such as Roman letters.
Therefore, its features must be carefully chosen.
By whom? And how are things "carefully chosen" if:
That is when a linguistic research is needed to find out how to make it easy for as much people as possible.
Who will "control the evolution" of the auxlang? Who will control the fairness (can't imagine English words creeping into the auxlang will be deemed "fair")?
The evolution of English is controlled by the Americans. The evolution of a neutral auxlang is not controlled by a specific nation.
English is not my native language. Sorry for any mistakes or lack of knowledge when I discuss this language.
:bra: :mrgreen: | :uk: [:D] | :esp: [:)] | :epo: [:|] | :lat: [:S] | :jpn: [:'(]

User avatar
Lao Kou
korean
korean
Posts: 5665
Joined: 25 Nov 2012 10:39
Location: 蘇州/苏州

Re: Completely unbiased auxlang

Post by Lao Kou » 15 Nov 2014 17:45

Squall wrote:Speakers of languages written with the Roman alphabet will not say that the alphabet is foreign.
It is easy to learn symbols that are few strokes, such as Roman letters.
The evolution of English is controlled by the Americans. The evolution of a neutral auxlang is not controlled by a specific nation.
Wow.
道可道,非常道
名可名,非常名

User avatar
Xing
MVP
MVP
Posts: 5292
Joined: 22 Aug 2010 18:46

Re: Completely unbiased auxlang

Post by Xing » 16 Nov 2014 09:10

Tanni wrote: I just wanted to propose that seperating auxlanging form making a culturally unbiased language. But after that
Micamo wrote:It's because supposed cultural neutrality is the only advantage an artificial auxlang can claim to have over something like Basic English. If auxlangers gave that up, then they'd have nothing left to design towards.
it might not be a good idea. But regardless of that, there is no language without culture.
[+1]

And we must also take into account the fact that people change language over time. Just as English has been adopted by lots of people in lots of countries outside its region of origin, we can't prevent groups of people from starting using a future auxlang as their L1/native language. This would create a new kind of bias.

Post Reply