How do languages other than English form lists of words? [closed]

Question

(Apologies if this is off-topic for Linguistics.) I'm trying to properly internationalize a web site. I have a sentence like, "You've earned badges A and B." Because the number of badges can vary, it can appear as:

You've earned badge A.
You've earned badges A and B.
You've earned badges A, B, and C.
You've earned badges A, B, C, and D.

I'm trying to deal with the full generality, so I'm wondering about other languages, and how they form lists like these. (For the purposes of this question, let's overlook the Oxford comma issue, and the pluralization of "badges", which I'm familiar with.)

This question shows that Chinese uses a different comma here than for other purposes, for example. What other possibilities do I have to take into account?

Answer 1

Haspelmath (2007) surveys coordination structures in the world's languages. He lists the following as logical possibilities for binary coordination. The examples are also taken from Haspelmath 2007. co here indicates the coordinator/conjunction.

• A B (widespread, e.g. Sarcee (Athapaskan; Canada), Maricopa (Yuman; Arizona), Kayardild (Tangkic; northern Australia))
• A co-B (English, Lango (Nilotic; Uganda))
• A-co B (Classical Tibetan)
• A B-co (Latin)
• co-A B unattested
• co-A co-B (Yoruba (Kwa; Nigeria))
• A-co B-co (Martuthunira (Pama-Nyungan; western Australia))
• A-co co-B (Homeric Greek)
• co-A B-co (Latin)

For the purposes of NLP, the distinction between "A-co B" and "A co-B" might be important for reasons of punctuation. For example in Arabic, the conjunction و wa is prefixed to the following word, without a space.

According to Haspelmath 2007, who cites Stassen 2001 as the authority, of the last four strategies, only "A-co B-co" seems to ever be used as a default, but it is widely-attested. "co-A co-B" is used only to contrast with "A co-B". The last two strategies are "extremely rare" (Haspelmath 2007) and the languages displaying them almost certainly have other strategies for coordination. For example, Latin has "A et B".

Moving to multiple coordination, Haspelmath 2007 provides the following useful chart of major patterns of correspondence between binary and multiple coordination:

Surprisingly, the pattern of coordinator omission "A-co B C D", though rare, is attested in Classical Tibetan and Amharic. It is also possible to omit all coordinators, leaving "A B C D".

For the purposes of NLP, then, to really handle all the possibilities, you would need to be able to specify what separators occur in the following positions, where A...Z are the nouns.

▓▓▓ A ▓▓▓ B ... M ███ N ... Y ███ Z ███

(A couple of notes: I've made the first two blocks lighter ▓▓▓ to show that supporting these is likely to yield marginal returns, as most (all?) languages would support alternate strategies for coordination. Also, this would not handle the Oxford comma, which treats lists of length 2 and >2 differently.)

Sources:

Haspelmath, Martin (2007). Coordination. In Shopen, Timothy (ed.) Language typology and linguistic description, 2nd ed. Cambridge: CUP.

Stassen, Leon (2001). Some universal characteristics of noun phrase conjunction. In Plank, Frans (ed.) Noun phrase structure in the languages of Europe. Berlin: Mouton de Gruyter.

Answer 2

For french:

Vous avez obtenu le badge A.
Vous avez obtenu les badges A et B. 
Vous avez obtenu les badges A, B et C. 
Vous avez obtenu les badges A, B, C et D. 

Answer 3

In Russian it will be like in English but without Oxford comma

Вы заработали медаль А.
Вы заработали медали А и B.
Вы заработали медали А, B и С.
Вы заработали медали А, B, C и D.

Answer 4

For Turkish:

A rozetini kazandın.
A ve B rozetlerini kazandın.
A, B ve C rozetlerini kazandın.
A, B, C ve D rozetlerini kazandın.