Improper Implementation of Culture.

[SBolshevik]

In EU4 it should technically already be there because you can't represent the rise of nation-states without them.

EDIT: We also shouldn't expect too much from Paradox on the language front because they only started using Unicode with Stellaris. You can't properly represent most languages east of Germany (not even Old Norse, actually) in CKII, not even in transliteration, unless you use ugly, workaroundy conventions.

 

[tywinzo]

I watched some lecture yesterday (some sociological stuff) about some monk in 12th century who wrote down names of pilgrims. The icelandic name was something like Kolbert and the german monk wrote something like Gulzenna. The transformation of graphemes etc was consistent, but there were different monks who write it differently. Anyway, that whole thing of merging names/languages is a bloody science in itself. I know from some previous reading that the same people are refered to using different names depending on where they were (e.g. Hugh [french] = Ugo [italian]).

 

[NecroGangster]

I agree that the culture system needs a huge overall. One mistake I see most people doing in these forums is culture = language. While I agree that language is a very simple and easy way to separate different cultures it isn't 100% accurate.

Culture is based on language, but also based on religion and traditions/way of living. People with the same traditions and religion but with a different language have a different culture, but the other way around is also true, people with the same language and religion but with different traditions have different cultures.

Since religion is already it's own thing separate from culture one way to fix this would be to also make language a factor separate from culture. It would be something like this, using french as an exemple:

Culture: French/Frankish > Gallic (Culture group)

Language: Old French > Gallo-Romance (Language Group) > Latin (Language Family)

Maybe cut the opinion penalty of having a different culture by half and giving a similar penalty to different language, that way having both different language and culture would give the same penalty as having a different culture does now.

 

[SBolshevik]

Maybe cut the opinion penalty of having a different culture by half and giving a similar penalty to different language, that way having both different language and culture would give the same penalty as having a different culture does now.

It would also bring together really large groups like Slavs and Romance-speakers, which is a plus.

 

[Sergeant Flutter]

Wow! So many great responses!

I realize that CK2 may not be able to handle this.
But the cultural system needs some love, I understand the names could suffer, but there are a few solutions, which I will post this evening EST.

 

[ahyangyi]

Some thoughts:

• Make this happen rare enough. So a player could still trigger it, but the AI is not very likely to do that.
• Use some simple procedures to transform names.
• Classify existing names to "Imported", "Local", "Religious" or something along that line.
  
  Religious names would merge well. Spanish "Jesús" is Bedouin "Isa", or English "John" is Greek "Ioannis", so such names could simply convert from one to one.
  Imported names have a certain origin, so if it happens to merge with its origin it could merge well. Otherwise we have to use the procedual ones.
  For procedual ones to work, we would rely on simple methods that could potentially produce some stupid names. But I personally won't find it too immerse-breaking if Norse name Ívar becomes Ivo in Italia, or a "von Schaumburg" becomes "Al-Sakhumburgh".

 

[Woifee]

I mentioned it in other culure mimimi threads already, but I still supprt the Idea.

Get rid of Culture at all.

What does it represent anyways? The common people? So if I culture change I totally go genocidal. And why do I use Birdmana for that instead admin?

So the ruling class then. So by changing I convince them to educate the nobles and Kids in my customs and grandmas cooking. But why do colonies in Africa keep their native culture?

It's obviously not about language, neither about customs nor food. So what is it?

Culture pretty much fullfill the use of nationalism. People identify with a political entity and rebell against other political entities.

But there is no nationalism in 1444 Europe. So maybe include growing nationalistic agenda with end Tech and add the malus in taxes and recruits into the new State system.

And there are other advantages:

There won't be any Threads about people whining about their culture not beeing represented or weirdly splitted up.

I know for some reason many players want it, but first PI should get clear about what they want it to be, what they want to achieve and if culture is the best way to do it.

And if everybody disagrees with me at least agree to change the name culture to "Tag identifying with" or "Nation identifying with".

Some goes for CK2. There is no room for nationalism. Replace it with "known languages" and add malus in diplomacy for misunderstanding. But characters should be able to learn new languages.

In EU4 it should technically already be there because you can't represent the rise of nation-states without them.

 

[Aries666]

Yeah, let's have Władysław and John in one namelist, everyone's gonna like that.
In any case, making such a dynamic system would be infinitely hard because it would actually have to have a system of evaluating the orthography of each and every culture and adapting it to another one, or else it would have to prepare variations for every culture, so you'd have to have Gion (Italian, and a stretch), Džon (Czech, Croatian, Serbian), Dżon (Polish, Pomeranian), etc. Not to mention that the latter two are technically impossible within the framework of Slavic languages before the 11th century, and then you'd have to make special variations for the Baltic languages in which nouns can't end in a null morpheme, so Džonas and Džons for Lithuanian and Latvian respectively (at least I assume so - someone may correct me). The argument could be made that these cultures already have this name, but then you lose the point of making a new namelist entirely and just have a culture identical to the previous one except in name and colour.

You mean something along the lines of Władysław III who ruled Poland from 1434-1444 and John I (nephew of Władysław III) who also ruled Poland from 1492-1501

 

[SovereignGrave]

You mean something along the lines of Władysław III who ruled Poland from 1434-1444 and John I (nephew of Władysław III) who also ruled Poland from 1492-1501

In CK2 wouldn't he be referred to as Jan I?

 

[Aries666]

In CK2 wouldn't he be referred to as Jan I?

Indeed I was mostly pointing out that a more abstract set of names could likely have been picked (I'm having a pedantic a**hole day)

 

[Sergeant Flutter]

Some thoughts:

• Make this happen rare enough. So a player could still trigger it, but the AI is not very likely to do that.
• Use some simple procedures to transform names.
• Classify existing names to "Imported", "Local", "Religious" or something along that line.
  
  Religious names would merge well. Spanish "Jesús" is Bedouin "Isa", or English "John" is Greek "Ioannis", so such names could simply convert from one to one.
  Imported names have a certain origin, so if it happens to merge with its origin it could merge well. Otherwise we have to use the procedual ones.
  For procedual ones to work, we would rely on simple methods that could potentially produce some stupid names. But I personally won't find it too immerse-breaking if Norse name Ívar becomes Ivo in Italia, or a "von Schaumburg" becomes "Al-Sakhumburgh".

Genius!

 

[Narvait]

Some thoughts:

• Make this happen rare enough. So a player could still trigger it, but the AI is not very likely to do that.
• Use some simple procedures to transform names.
• Classify existing names to "Imported", "Local", "Religious" or something along that line.
  
  Religious names would merge well. Spanish "Jesús" is Bedouin "Isa", or English "John" is Greek "Ioannis", so such names could simply convert from one to one.
  Imported names have a certain origin, so if it happens to merge with its origin it could merge well. Otherwise we have to use the procedual ones.
  For procedual ones to work, we would rely on simple methods that could potentially produce some stupid names. But I personally won't find it too immerse-breaking if Norse name Ívar becomes Ivo in Italia, or a "von Schaumburg" becomes "Al-Sakhumburgh".

The issue here is that for Bedouin Spanish mix it should be neither Jesus nor Isa, but something in between.

 

[krste]

Yet they were Russified over time and still exist today: Helgi, Helga, Ingvar and Hrøríkr became Oleg, Olga, Igor and Rjurik, with Valdemar being interpreted as Vladimir I imagine. The game cannot simulate such a system of nativisation.

With Vladimir it't the reverse.

 

[mudcrabmerchant]

It would be very difficult to properly merge name lists. You actually did have this happen historically - all western Romance languages have a huge proportion of Germanic given names.

But to merge them properly, you have to adapt the phonology of one language onto another, and then encode that in the dominant language's orthography. Making a tool to do this between all possible languages would be a royal pain in the ass to create, but without it, merged name lists would just look awkward and half-assed. I, for one, do not want to see a Swedo-Italian family with children named Sven, Giovanni, Alberto, and Styrbjorn.

And honestly, I don't think it's worth it. The number of times this happened in real life during the period and region are few enough that it's feasible to just create a few more melting pot events and cultures, and leave it there. We do not need a game mechanic to dynamically generate Welsh-Ethiopian melting pots.

Also remember that there are many examples of cultural shift in which the vanishing culture had very little impact on the assimilating culture, at least in terms of things that are represented in CKII. You can't really represent the assimilation of Arabic words into Spanish and Portuguese, or the elements of Greek influence on Anatolian Turkish culture.

 

[mudcrabmerchant]

The issue here is that for Bedouin Spanish mix it should be neither Jesus nor Isa, but something in between.

There are patterns to language shift. You don't just get something "in between". In this case, if we have Bedouins shifting to Spanish, and they want to name their kid after Jesus, they'll either call them Jesús (the pre-existing Spanish form), or adapt the Arabic form of the name to Spanish phonology... the end result of which would still be written "Isa".

What about the other way - if we have Spaniards shifting to Arabic, and for some reason they don't just use the Arabic versions of shared names, then they will adapt Jesús to Arabic phonology... which would be transliterated from Arabic script to Latin as Haysus, or I guess Jaysus, because in the Middle Ages Spanish j still sounded like a normal j.

Is Jaysus in between Jesús and Isa? I wouldn't say so. The problem of procedural culture melding is that the transfer of any word (including a name) from one language to another is dependent on the unique phonologies of both languages. This means that to do it right, you need to model the phonologies of all languages represented in CKII, and write a script that can adapt a word from the phonology of any one language into any other language.

Is it possible? Yes. But it would require a trained linguist (or at least someone who took a phonology class and remembers half of it), and it would take hours and hours and hours of trudgery, especially if you wanted to get the Medieval phonology right*. Which matters, as you can see in the Haysus/Jaysus thing.

And if your response is, "well we don't need something that complicated, something simple will do for me"... well, there is no simple way to do it right, and if we don't do it right, then you aren't arguing for a proper implementation of cultural melding, you're just arguing for one that is improper in a different way. To put it another way, only a few people will be able to object with the same level of detail as me, but it will feel wrong to more people than you might think.

*If Paradox would like to hire me to create a database of Medieval phonological systems, create IPA versions of every name in every culture, and then write scripts to transfer names between phonologies, and to transfer names from IPA to every language's orthographical system, feel free to contact me. It'd probably only take a week or two of solid work.

 

[Narvait]

You will need two weeks for Baltic languages alone )
And if you manage to fix whether Curonians were East or West Balts...
Whether Lettigalians indeed deserve their own culture or should be part of East Baltic common group in early start dates.
(Latvians should possibly be melting pot between East Baltic, West Baltic, Uralic under German superstrate).

Also good luck reconstructing fonetics of Uralic languages and relevant melting pots. If you do it in 2 weeks generations of linguists would thank you

In meantime you could also establish location of East Galindians and their fonetics (got extinct after 1000 AD, somewhere near modern Moscow) and ensure their namelist for all melting combinations.

 

[SBolshevik]

Some thoughts:

• Make this happen rare enough. So a player could still trigger it, but the AI is not very likely to do that.
• Use some simple procedures to transform names.
• Classify existing names to "Imported", "Local", "Religious" or something along that line.
  
  Religious names would merge well. Spanish "Jesús" is Bedouin "Isa", or English "John" is Greek "Ioannis", so such names could simply convert from one to one.
  Imported names have a certain origin, so if it happens to merge with its origin it could merge well. Otherwise we have to use the procedual ones.
  For procedual ones to work, we would rely on simple methods that could potentially produce some stupid names. But I personally won't find it too immerse-breaking if Norse name Ívar becomes Ivo in Italia, or a "von Schaumburg" becomes "Al-Sakhumburgh".

It would still require a lot of work to make even a simple system like this.

With Vladimir it't the reverse.

The More You Know, I guess.

You will need two weeks for Baltic languages alone )
And if you manage to fix whether Curonians were East or West Balts...
Whether Lettigalians indeed deserve their own culture or should be part of East Baltic common group in early start dates.
(Latvians should possibly be melting pot between East Baltic, West Baltic, Uralic under German superstrate).

Also good luck reconstructing fonetics of Uralic languages and relevant melting pots. If you do it in 2 weeks generations of linguists would thank you

In meantime you could also establish location of East Galindians and their fonetics (got extinct after 1000 AD, somewhere near modern Moscow) and ensure their namelist for all melting combinations.

Oh, you forgot to mention that Uralic cultures still have names like Altocumulus, i.e. their namelists are rubbish, like a good number of cultures in Eastern Europe.
Also, don't forget Norse is two dialects, so you'd have to make separate names for both if you want authenticity.
And also don't forget that Italian and German are such blanket cultures that they shouldn't even exist as is and have so much insane regional variation.

EDIT: There's no such thing as a "medieval phonology" by the way, considering many languages completely changed during this period, yet the game doesn't represent this whatsoever, and it would certainly require a separate culture for every culture every century or two, if we consider that the languages would go the way they did in history, which is questionable.
Let's take Slavs for an example - in 769 you start and you can see the name Svyatoslav in Russia. Yeah, Svyatoslav, a name that wouldn't get that form for centuries to come, and in the Common Slavic period, too. At this point it should still end with a big yer and have a nasal (i.e. Svętoslavŭ, impossible to represent ingame, tying into what I said about no Unicode). In fact, every Slavic word that ends in a consonant should end in a yer until the 11th century, yet we get the later forms in 769 and 867. All the cultures should still have a distinct yat and nasal vowels in this period too, yet, as seen with Svyatoslav, it's as if they never existed.

 

[mudcrabmerchant]

You will need two weeks for Baltic languages alone )
And if you manage to fix whether Curonians were East or West Balts...
Whether Lettigalians indeed deserve their own culture or should be part of East Baltic common group in early start dates.
(Latvians should possibly be melting pot between East Baltic, West Baltic, Uralic under German superstrate).

Also good luck reconstructing fonetics of Uralic languages and relevant melting pots. If you do it in 2 weeks generations of linguists would thank you

In meantime you could also establish location of East Galindians and their fonetics (got extinct after 1000 AD, somewhere near modern Moscow) and ensure their namelist for all melting combinations.

Well, I was figuring that a reasonable line to draw would be at the limits of the modern scholarly consensus. But then I guess that isn't being totally "right" in away similar to how Asbjorn di Canossa wouldn't be right... but it's as right as we can get within the bounds of modern knowledge.

So, at the risk of sounding hypocritical/arbitrary, we'd just make (as highly as possible) educated guesses as to the phonology of underdocumented/reconstructed languages, and/or assume for our purposes that they are identical to better documented relatives. It'll be right enough!

But once you have your inventories of phonological systems, it shouldn't be too hard. You then need a script that picks the best match-ups between phones in either language, and also applies changes to respect differences in phonotactics. The real bitch would be turning the IPA forms into each language's orthography, which means making a database of each orthography, and writing rules to turn IPA forms into that orthography. In less-phonemic orthographies, you'll just have to pick one grapheme or so for each phoneme.

But... after all that, I think you'd end up with a system that most linguists would consider good enough for a videogame. And that's the highest standard you can hope for, I think.

EDIT:

To demonstrate an example (half-assing the IPA):

Old Norse > Italian melting pot
Transfer 10 Old Norse names to Italian name pool
Asbjorn Arnbjörn Hroðgar Ingjald Kettilmund Olafr_Olaf Snorri Þorbjörn Öysteinn
/asbjorn/ /arnbjɔrn/ /hroðgar/ /ingjald/ /kettilmund/ /olavɾ/ /snori/ /θorbjorn/ /œystein/

Apply Italian phonotactics (and delete some? all? grammatical endings)
/asbjorno/ /arnebjɔrno/ /hroðegaro/ /ingjaldo/ /kettilmundo/ /olavo/ /snori/ /θorbjorno/ /œysteino/

Apply Italian phonology
/asbjorno/ /arnebjɔrno/ /rodegaro/ /ingjaldo/ /kettilmundo/ /olavo/ /snori/ /torbjorno/ /eisteino/

Apply Italian orthography
Asbiorno Arnebiorno Rodegaro Inghialdo Chettilmundo Olavo Snorri Torbiorno Eisteino

Aaaaand I notice there's a whole lot of other stuff you'd have to do. Preferred epenthetic vowels by language, how to represent grammatical endings like -r that are often lost in borrowing, how each language fits loans into their own grammatical system (I picked -o as the ending because most earlier Germanic loan names ended in it)...

OK, make it a month of work.

 

[Narvait]

Will think out loud.
What is melting pot is usually a base language heavily (or not so heavily) influenced by foreign language.
Theoretically that makes two possible outcomes for 2 involved languages
Arabic infuenced Norse
Or
Norse influenced Arabic

Then we have different relations.
Base language has a foreign substrate (culture expands into foreign province)
Base language has a foreign superstrate (culture has foreign elite that gets assimilated)
That makes four outcomes:
Norse with Arabic substrate
Norse with Arabic superstrate
Arabic with Norse substrate
Arabic with Norse superstrate

If foreign language is substrate, it would keep names of base language but with some fonetic change (I.e. Belarussian uses a instead of o because of Baltic substrate).
If foreign language is superstrate, then base language would assimilate foreign names, but also change some or many of its own fonetics.

Just putting it there to think about it more

 

[mudcrabmerchant]

So now we have to write a script for phonological interference across an entire name list, on top of the script for adapting name borrowings. Possible, yes. Cool, yes. But a bitch to do right, and irritating to some (well, at least me) if done wrong.

Also, superstrates generally don't affect phonetics, or really, anything but vocabulary, of the language they assimilate into, unless it's by adding new phonemes. I can't think of any significant Germanic influence on Western Romance phonology, and French had only the smallest effect on English phonology (the biggest one is introducing /v/ as a sound at the beginning of words, and not just as a variant of /f/ after vowels).