Daughter Languages or Dialects?

[Davush]

Given this set of cognates (numbers) between three languages derived from a parent language, would you imagine the following to be dialects of one language, or separate languages? Of course this answer isn't always clear-cut, but I'm aiming for enough diversity that they are clearly separate languages.

Also, does the amount of divergence look realistic? I'd like the languages to be identifiable as belonging to one group, but have a different 'feel'̇.

LANG-A
jaʔ
tiri
ʃaŋ
ʃu
mʃaʔ
rina
ʧia
pɔa
njɔ
aʃar

LANG-B
jek
teari
ʃaŋ
ʃuo
saʔ
jena
ʧiwa
pawa
ʤaw
aʃax

LANG-C
jah
tɛʒɨ
ʃaŋ
ʃu
mɨsa
lɨna
ʧɨva
pava
nɨʤaw
ʃar

[WeepingElf]

Hard to say from such a sample, but would say that these are different languages, clearly related to each other, but probably not easily mutually intelligible.

[Micamo]

The thing is, people are remarkably proficient at figuring out regular sound correspondences unconsciously and using them to "decode" what they hear, so long as they recognize the words in the first place. The factor that contributes most to mutual unintelligibility is semantic shift within the lexicon. So in trying to guess whether two sets are dialects or separate languages, what I'd actually ask is how much of each lexicon has shifted away from its original meaning. If over 50% of what you hear are words you don't recognize, or words you do recognize being used in a different way, intelligibility is going to be very low even if both lects are very conservative with regards to phonological change.

[Salmoneus]

The thing is, people are remarkably proficient at figuring out regular sound correspondences unconsciously and using them to "decode" what they hear, so long as they recognize the words in the first place. The factor that contributes most to mutual unintelligibility is semantic shift within the lexicon. So in trying to guess whether two sets are dialects or separate languages, what I'd actually ask is how much of each lexicon has shifted away from its original meaning. If over 50% of what you hear are words you don't recognize, or words you do recognize being used in a different way, intelligibility is going to be very low even if both lects are very conservative with regards to phonological change.

I agree mostly, but I think the biggest thing is actually the degree of contact.

If you and I live in neighbouring towns, we can see each other on TV every week and we meet each other when we both go to the capital, then I can come to "decode" really quite sophisticated differences, including semantic drift - at least to catch the drift, if not to get all the nuance. But if i've never heard your accent before, I'll struggle to make sense of it even if changes are minimal.

To give a concrete example: I can understand almost every UK dialect without the slightest difficulty. Not to get all the nuances, but when I see people speaking Geordie or Scouse, I know what they mean almost all the time, without consciously having to 'decode' them. That's not boasting, that's just how things are over here. Because we know these people personally, and we see them regularly on TV (Match of the Day, the world's best guide to weird English accents...). So it goes without saying that our dialects are mutually intelligible.
And yet I know that quite a lot of Americans seriously struggle to understand what Liverpudlians, for example, are saying. Because they haven't had that weekly training. Conversely, Americans have a much easier time understand AAVE speakers that English listeners would struggle much more with.

So intelligibility isn't just a function of the languages themselves, but also of their cultural context.

[cntrational]

Yeah, it's just the normal human ability to pattern-match, which depends on regular input.

[HoskhMatriarch]

Well, in order to say if they're different languages or dialects, along with the other things people said, you have to look at grammatical differences, and even political factors (like Low German dialects, which are considered German in Germany and Dutch in the Netherlands, even if it's the same Low German dialect). Basically, you did not give enough information. You could compile a word list like that for basically every Germanic language and make it look like they're the same language, or you could take different dialects of, say, English and make them look like different languages depending on what your criteria are.

[Curlyjimsam]

I agree mostly, but I think the biggest thing is actually the degree of contact.

Yeah. I find modern American accents easier to understand than (say) 1950s American accents - though quite possibly the latter are in fact closer to my own (British) English. This is presumably because there's more exposure to modern American accents in the media, I come into personal contact with Americans who were born more recently more often, and so forth.

[Tanni]

Given this set of cognates (numbers) between three languages derived from a parent language, would you imagine the following to be dialects of one language, or separate languages? Of course this answer isn't always clear-cut, but I'm aiming for enough diversity that they are clearly separate languages.

Also, does the amount of divergence look realistic? I'd like the languages to be identifiable as belonging to one group, but have a different 'feel'̇.

LANG-A
jaʔ
tiri
ʃaŋ
ʃu
mʃaʔ
rina
ʧia
pɔa
njɔ
aʃar

LANG-B
jek
teari
ʃaŋ
ʃuo
saʔ
jena
ʧiwa
pawa
ʤaw
aʃax

LANG-C
jah
tɛʒɨ
ʃaŋ
ʃu
mɨsa
lɨna
ʧɨva
pava
nɨʤaw
ʃar

Sometimes it may help to look at actual language families:

... would you imagine the following to be dialects of one language, or separate languages?

Turkic languages, see section Vocabulary comparison.

[mira]

All three versions here use very similar sounds. They could be 3 closely related languages if the rest of each differ more significantly. However, if they all use the same grammar then they would be very convincing accents of the same language. For instance received pronunciation, London English, and Scottish English.