The Rosenfelder Challenge

[ashmoonfruit]

Standard Arabic. The same point would apply to Hindi as well, and probably Mandarin. You have to start somewhere.

I had thought about the 'cipher' problem. That's the main reason for offering multiple etymologies for the same basic word, and mentioning the need to add plenty of culture-specific words. And of course the program won't produce any grammar. I figure it'll still save a lot of time. If you've ever sat down with a list of words and tried to come up with conlang words for them, then you'll know how tedious it is. If the 'grunt work' were done automatically, then the conlanger will have more time to focus on more interesting tasks like coming up with idioms etc.

Also, languages are partly ciphers of each other. Please don't think that I'm overestimating the extent of this, but English 'dog' French 'chien' are pretty much ciphers of each other. When it comes to how compounds are formed, sometimes languages do it the same way, and sometimes not: 'Grandmother' and 'grand-mére' are the same compound, 'black box' and 'béte noir' (as in 'pet hate') are not. So, I think such a program can cope with this problem, especially once those compound words are edited for sound changes where the two source words meet.

I'm sure a lot of people wouldn't want to use such a short-cut, but for someone who wants to knock up a quick conlang to use in an RPG or fantasy/sci-fi composition, it could be really handy.

[Yng]

Standard Arabic. The same point would apply to Hindi as well, and probably Mandarin. You have to start somewhere.

If you plan on going into a place like your Peruvian village with MSA, you'll probably be disappointed with MSA. Also I would like to see the results of that, very, very much.

I had thought about the 'cipher' problem

no you haven't

And of course the program won't produce any grammar. I figure it'll still save a lot of time. If you've ever sat down with a list of words and tried to come up with conlang words for them, then you'll know how tedious it is. If the 'grunt work' were done automatically, then the conlanger will have more time to focus on more interesting tasks like coming up with idioms etc.

1) if you think words are the most important thing then why are you trying to take that away from people
2) automatic word generators already exist
3) grammar is an important part of word formation

Please don't think that I'm overestimating the extent of this

You definitely are

When it comes to how compounds are formed, sometimes languages do it the same way, and sometimes not: 'Grandmother' and 'grand-mére' are the same compound, 'black box' and 'béte noir' (as in 'pet hate') are not.

You can't do French, stop it

I'm sure a lot of people wouldn't want to use such a short-cut, but for someone who wants to knock up a quick conlang to use in an RPG or fantasy/sci-fi composition, it could be really handy.

See above. Random word generators already exist.

[Vuvuzela]

Grand is an Old French loanword. The English compound was probably calqued on the French. Granted, German uses a similar compound, Großmutter, and I'm not sure if that ties in with this etymologically.

[Terra]

ク as "ku or k", for instance (actually it more often transcribes "k" rather than "ku" from English)

The worst is how /o/ is always added onto syllables that end with /t/ or /d/.

And sometimes they just got the pronunciation wrong, such as チョコレート /tʃokoreːto/ for chocolate /ˈtʃɒklət/ (although sometimes when they seem to have got the vowel wrong it's actually because it's from another European language, eg "botan" for button from portuguese or "pompu" for pump from dutch).

For "chocolate", one has to explain how stres matters in English, and how unstressed vowels are often turned into schwa. Whether they'll actually remember and heed it, I don't know.

that float around and occasionally make their way back into actual English (1UP, GET, Pokemon, anime, etc), they can be annoying when you teach English and your students use them thinking that they're actual English.

Japanese likes to shorten a word to 2-3 morae and combine it with other shortened words, but English usually either doesn't shorten the words, or uses an acronym.

Also, I think "GET" is only ever used humorously, intentionally mocking the poor English translations of video games (especially when they weren't as popular and couldn't afford large budgets for accurate translations).

because they tend to just assume that anything written in katakana is from English.

Interesting. FWIW, sometimes it works out alright, like "biru" (or is it "biiru"?). It was borrowed from Dutch, long before English became fashionable.

But anyway, overall it's a lot easier than you make out.

But could Ashmoon work through all this if he had no forward knowledge of it? Then again, did you?

I still don't know the real word for "change" in Japanese, for instance.

kaeru : change (transitive verb)
kawaru : change (intransitive verb)
kawari : change (noun)

~*~ The more you know ~*~

[ashmoonfruit]

Hello Yng. [blows kiss].

...

Yep, that's all I have to say.

[finlay]

ク as "ku or k", for instance (actually it more often transcribes "k" rather than "ku" from English)

The worst is how /o/ is always added onto syllables that end with /t/ or /d/.

And sometimes they just got the pronunciation wrong, such as チョコレート /tʃokoreːto/ for chocolate /ˈtʃɒklət/ (although sometimes when they seem to have got the vowel wrong it's actually because it's from another European language, eg "botan" for button from portuguese or "pompu" for pump from dutch).

For "chocolate", one has to explain how stres matters in English, and how unstressed vowels are often turned into schwa. Whether they'll actually remember and heed it, I don't know.

that float around and occasionally make their way back into actual English (1UP, GET, Pokemon, anime, etc), they can be annoying when you teach English and your students use them thinking that they're actual English.

Japanese likes to shorten a word to 2-3 morae and combine it with other shortened words, but English usually either doesn't shorten the words, or uses an acronym.

Also, I think "GET" is only ever used humorously, intentionally mocking the poor English translations of video games (especially when they weren't as popular and couldn't afford large budgets for accurate translations).

In English, maybe. But you see people writing things like "300th post GET". In Japanese it's very much the common usage of the word. They use it in Pokemon, for instance.

[Legion]

Hello Yng. [blows kiss].

...

Yep, that's all I have to say.

You're a liar and a troll lol.

[Drydic]

Hello Yng. [blows kiss].

...

Yep, that's all I have to say.

brilliant response

10/10 would read again

must see! vivacious young actors!

[ashmoonfruit]

Here's part of a conversation I had recently with a Cocama (Amazonian) friend who I met in Perú. As you can see, I have learned a little more Spanish since I met her (she comments on it: "tu castellano es mejor").

castellano.png (98.57 KiB) Viewed 3773 times

Here's a video I made about the animal sanctuary where I worked at time when I was learning Spanish using my system: http://www.youtube.com/watch?v=enuEGSZ_EAE&feature=plcp
Here's a video I posted shortly after leaving the village, in which I use my simplified Spanish to admire the mountains that I had just climbed:
http://www.youtube.com/watch?v=Wc0jxz7VtSo&feature=plcp
Here is a description of my conlang, Latha, which is based on a core lexicon of about 100 single syllable roots - this formed the basis of the core lexicon I first learned in Spanish:
http://www.mediafire.com/view/?c0fzss15ms54ytz
Here is the program I wrote recently to help others learn Latha using my system:
http://www.mediafire.com/download.php?d7u6c17b676sz2s
...and here is the same program adapted for Spanish (I'm in the process of developing versions in more languages, and making the program better as I go along):
http://www.mediafire.com/download.php?a6sb3c2ybulcsfl
I could go to the trouble of putting anyone who stills doubts my story in touch with my Perúvian friends who witnessed me learning Spanish using this system (if the conversation on FB there isn't enough for you), but it would be better if you just tried out my system so you can see it working for yourself. Given a little time, I will even make a version of my program in the language (or conlang) of your choice, if you provide me with my list of core words translated.

I know that the idea that a scruffy backpacker can demonstrate that some of the assumptions that some linguists make about how much of a language you need to know to be able to communicate are a little pessimistic may be a little difficult to believe. Having seen the criticisms of Basic English, and being told by several people who know much more about languages than me that it couldn't possibly work (especially using less than half of Basic's 850 word lexicon), the extent of my success with this surprised me too. But it's true. Go figure. I guess some linguists need to get out more - they might even meet some sexy latinas.

[Ser]

Responding to Yng's points in an intelligent way would be a better idea, especially the cipher/cognate points.

[Torco]

well, sure, just learning 200 words can get you from english to spanish, I mean, they're similar enough... but I'm quite certain it will not work for weirder langs like japanese, mandarin, urdu, zulu, mapuzungun, guarani, you know, the langs that have different grammars

however, people are very smart, and if you produce grammatical nonsense its easy for them to decode it, if you pace yourself and make simple constructions. like

know I spanish some. learn it rosenfelder challenge by method did I. bad, fact in, not good, is it. understand other can people still, good so that's. bathroom is where. i thanks you give. i can haz cheezburgr ?

people will generally know what you mean. I mean its pretty awesome, gaining a functional level of Spanish. Sounds like it would be a nice immersion
system, but realize its mostly due to the inherent similarities between both 'langs. try doing that with, say, Czech and, if you pull it out, *then* you'll impress us.

But yeah, you're still right, immersion plus enthusiasm and a bit of native goodwill works across similar languages, and even should work with very different ones: a friend of mine went to Europe knowing shit of English, like not being able to read simple text in the 'lang, and now she's making friends with russians using it, in less than a month. now, this doesn't challenge the assumptions of linguists; every linguist knows that basic communication is possible even without *any* common knowledge of language, like I could find myself a monolingual speaker of Navajo, me knowing no navajo at all, and if we had to, we'd end up understanding one another at a basic level in a matter of hours, and after a couple of days we'd be able to discuss a few complicated ideas. This isn't new. It's a coolio experience, and I'd like to know more about the process of acquiring that sort of extremely basic, barebones functional spanish, but yeah, its not like it defies the established linguistics paradigm or something. hell, I remember talking for hours on a train with a dude who spoke Czech and horrible german, me speaking english and some french, and spanish, of course, and we got along just fine, talking about how hot slavic women are and kutna hora and the 'zie geist uf zie czech folken', as he put it.

[Rodlox]

Also, languages are partly ciphers of each other. Please don't think that I'm overestimating the extent of this, but English 'dog' French 'chien' are pretty much ciphers of each other.

er...no.

(unless you don't know what a cipher is. if you don't know, just say so, its okay if you're honest)

[brandrinn]

Some words are extremely similar. Maybe it's true that no two words are ever 100% the same, but some certainly pass 99%. The usage may differ, but it's a fine line between usage and... you know, actual grammar. The area where hidden differences will often lie is slang, metaphor, connotations, and register. But most of that doesn't make it into the dictionary definition of a word.

[ashmoonfruit]

Serafin:
I can't see Yng's posts any more, which is a good thing. I'm not too keen on Rodlox' tone either.

Sure, technically, replacing one word for another would be a code, not a cipher. Not important. A conlang that just switches English words for made-up words would be 'just a code', yes - but I don't that as a reason why someone shouldn't do that, if that's what they want to do - or even that doing so is in some way inferior conlanging. The difference between doing that and an extremely complex and comprehensive technical description of a conglang is only a question of how complex is the algorithm that translates one language to the other. All languages are 'just codes' of each other.

Torco:
Yes, this is exactly what I'm talking about.

Some words are extremely similar. Maybe it's true that no two words are ever 100% the same, but some certainly pass 99%. The usage may differ, but it's a fine line between usage and... you know, actual grammar. The area where hidden differences will often lie is slang, metaphor, connotations, and register. But most of that doesn't make it into the dictionary definition of a word.

...and no two speakers of the same language follow identical rules of word usage and grammar, either. This is of central importance to this discussion. The most complete description of English ever written is not a description of the way that anyone actually speaks. I'm not knocking it; it's a brave and worthwhile thing to attempt, and preferable to prescriptivism (although, I am a fan of 'fonetik' spelling, in theory). In a way, language descriptions would be more accurate if they were more vague.

My main interest is in actually speaking (and writing) foreign and constructed languages. I observe that the thing we really can't do without is a certain number of words. We can mangle your spelling and pronunciation a bit and get away with it. But with a lot of grammar, we can be really quite chaotic about it and still order a beer, discuss climate change with the locals and crack enough jokes to get off with one of them. To say this is not an attack on linguistics. It just a useful reminder for those who are interested in using natural or constructed languages to communicate.

Kereb:
I was led a believe that ZZB is a conlang forum first, and a linguistics forum in addition. The fact that a few hundred words and a reckless disregard for grammar work fine as an initial basis for communication is relevant to both. The reason I posted the picture and the links was as evidence for those who accuse me of lying about taking up the 'Rosenfelder Challenge' in Perú.

[Legion]

Nobody answer him.

[Yng]

la la la la la if I just refuse to engage with any arguments against my tenuous position there's no way I can ever be wrong la la la la

I suppose if he blocks enough people he can convince himself the ZBB is agreeing by omission, or something

[Pthagnar]

ashmoonfruit: could you please conduct all your future attempts to persuade us that you are talking sense and being revelatory and useful in broken english/spanish? you should not lose anything of value by doing so

[Miekko]

Sure, technically, replacing one word for another would be a code, not a cipher. Not important. A conlang that just switches English words for made-up words would be 'just a code', yes - but I don't that as a reason why someone shouldn't do that, if that's what they want to do - or even that doing so is in some way inferior conlanging. The difference between doing that and an extremely complex and comprehensive technical description of a conglang is only a question of how complex is the algorithm that translates one language to the other.

It's not just a question of how complex the algorithm is, since the information coded for in different languages is different. There is actual information loss going both ways! The algorithms used must guess information, and if we keep passing the message back and forth from language A to B, the final form will contain a little bit of the original information and the rest will have been replaced by guesses. That is not how a cipher works.

All languages are 'just codes' of each other.

This is quite genuine bullshit.

...and no two speakers of the same language follow identical rules of word usage and grammar, either. This is of central importance to this discussion. The most complete description of English ever written is not a description of the way that anyone actually speaks. I'm not knocking it; it's a brave and worthwhile thing to attempt, and preferable to prescriptivism (although, I am a fan of 'fonetik' spelling, in theory). In a way, language descriptions would be more accurate if they were more vague.

Now, of course two speakers of the same language do differ when it comes to lexicon, connotations in the lexicon, grammar, etc etc. So the grammar we try to describe is a weighted average, possibly with notes of common variations. Many language descriptions are accurate exactly because they point out vagueness. As I said, a weighted average, with distributions, etc.

My main interest is in actually speaking (and writing) foreign and constructed languages.

If your main interest is speaking, why an interest for constructed languages (except maybe Volapük and Esperanto, Ido, that Slavlang and possibly uh, Latino sine flexione or whatever its name was) - certainly there's too few to speak them to for it to be worthwhile?

But with a lot of grammar, we can be really quite chaotic about it and still order a beer, discuss climate change with the locals and crack enough jokes to get off with one of them. To say this is not an attack on linguistics. It just a useful reminder for those who are interested in using natural or constructed languages to communicate.

I don't see it as an attack on linguistics, and sure, these things are possible. However, very much of the communication done by natives will conform fairly well to the crest of the bell-curve of the distribution of the likelihoods of how sentences will be formed, and conforming to that helps people understanding with a minimum of effort as well as a minimal risk of misunderstanding. And you seem to desire to be in denial of this. Finally, see my post in the other thread regarding "using ... constructed languages to communicate".

[clawgrip]

I'm not even sure what's being argued here anymore. I guess ashmoonfruit you're basically trying to say that cultural knowledge (rules of social interaction, history, taboos, famous people, humour, etc.) that can be most easily acquired through human interaction is important to language use and is something that is unlikely to be found in reference grammars?

Or, are you saying that extremely basic communication requires words but not necessarily grammar (for instance, if I went to Turkey, and looked in my Turkish dictionary for the words "toilet" and "where" I could probably make myself understood)?

Or both?

[Miekko]

I'm not even sure what's being argued here anymore. I guess ashmoonfruit you're basically trying to say that cultural knowledge (rules of social interaction, history, taboos, famous people, humour, etc.) that can be most easily acquired through human interaction is important to language use and is something that is unlikely to be found in reference grammars?

No, he's not saying any of that. Trying to save his face by reading that into what he's saying is very charitable, but it's far from what he's actually been saying.

[ashmoonfruit]

Miekko, I think you make good sense a lot of the time. I came here expecting a degree of pointless argumentation and hair-splitting, and I'm glad of the 'foe' function, so that I can focus instead on the more informed and constructive criticism that comes from you, and others.

I think that what you say about languages as codes and information loss is reasonable, and I also recognise that the best analyses of natural languages account for the variety of usages. Your point about communication being more effective when it conforms to the peak of that bell-curve is insightful and well put. ...and correct, of course. The question that I have tried to tackle is how a learner can most easily climb that curve. I noticed that learning a description of a language from a book is a difficult route, and that immersion is much faster but extremely difficult at first, since you need to be able to say something before you can access any conversation at all. In taking up the 'Challenge', I think I have demonstrated that carefully constructing a sort of 'pidgin' is a way to hop far enough up the curve to make immersion much easier.

I was never in denial of the fact that grammar is essential. Since joining the ZBB and taking in what critics have said, I plan to take a less cavalier attitude toward grammar. With the help of friends who are native speakers, I'm now applying my method to German and Arabic (!), and I'll be incorporating multiple forms of essential verbs (because 5 forms of 'be' are more useful than 5 animals or whatever) - although I'll be including them as vocab items (because that'll make them easy to learn).

Clawgrip, while it's true that those aren't exactly the points I was trying to make, I do agree with both, and I think that trying to read through others' words to understand them better is worthwhile.

Even more so, I'm not sure what's being argued here either. My first intention was to tell the story of the 'Challenge', explain the conclusions that I drew from it, give an indication of where I intend to take the idea from here - so that others might benefit from the experiment, and also so that the ensuing discussion might help to guide my next steps as I apply the method to a bunch of other languages that are more different to English. All of these things have happened, so I'm happy.

[Legion]

HOW LOW CAN YOU GO

HOW LOW CAN YOU GO

[ashmoonfruit]

...and, why the interest in conlangs, given that speaking is my priority?

Speaking is my priority when trying to learn a natural language. My interest in constructed languages is more to do with writing fantasy fiction, and just for the fun of it - but speaking is one of my measures of success. I have an interest in auxiliary languages too, because learning about them is useful for both other interests, particularly as examples of simplification.

[Torco]

No, what I'm saying is not what you're saying, ashfruitmoon, I'm saying that your claim that learning langs based on like two grammatical rules, a few hundred words worth of lexicon, and a buckton of good will from native speakers is possible is true, but that's its true for reasons different from those you think.

[ashmoonfruit]

I don't know why it's true, Orcrot. Why do you think it's true?

I'm just glad that it is true (quid pro quo), and I'm busy exploiting that fact.