Relationship between language similarity and ease of foreign language acquisition?

Question

Inspired by this infographic, which lists a bunch of languages as either easy to learn, hard to learn, or medium. I noticed that all the languages in the easy category were either Romance or Germanic languages, about half of the medium category was Indo-European, and none of the four hard languages were Indo-European.

Clearly for English, there's some correlation between genetic similarity and ease of foreign language acquisition. I'm wondering how much this pattern holds for other language families.

For example, if the same chart was made from the point of view of a Mandarin speaker, which languages would be in the easy category and which in the hard category? Would Tibetan, a language related to Chinese, be easier for a Mandarin speaker than Vietnamese, which is not? How much easier is it for a Mandarin speaker to learn Japanese (as compared to an English speaker)?

I've looked for equivalent infographics / data for languages other than English, but so far I haven't found any. Either scientific or anecdotal evidence is fine.

Answer 1

This response is personal opinion, since I think it would be difficult to find scientific evidence that addresses the full spectrum of what this question entails, and anecdotal evidence is even worse, since no one knows enough languages to give a consistent answer. So all you're gonna get is opinions of languages people have learnt, and likely-misinformed beliefs about languages they think they know enough to comment on.

Consider different ways of and motivations for second language learning. Some learn in a classroom setting. Some learn naturalistically in environments where the target language is spoken. A few among those cut off contact with their native languages altogether. People learn a second language because it's expected of them (e.g. school); for business/work; to communicate with specific people (e.g. missionaries); out of sheer necessity (e.g. immigrants); or just as a hobby. Then people have differing standards on what counts as functional competence. All these variables also differ from one target language to the next.

With that extended disclaimer out of the way, here's my actual response to the question.

I don't think genetic closeness is really an important consideration in and of itself. I would consider similar sound inventories, shared vocabulary, social/cultural predispositions, and typological likeness to be far greater factors.

Broadly speaking, I think the average Mandarin speaker would have more trouble learning Tibetan than Vietnamese on those grounds, despite genetics. Tibetan phonemic makeup is maybe slightly closer than Vietnamese to Mandarin, but Tibetan phonotactics (e.g. complex onsets and cross-morphosyllabic phonological effects) would probably pose a greater challenge than Vietnamese's much friendlier syllable structure. Most Tibetan cognates of Chinese words are probably not immediately recognizable as such, outside of the obvious numerals and kinship terms, which Vietnamese partly shares with Chinese anyway from extensive borrowing. The deciding factor is probably Tibetan's inflectional morphology, with case endings and verb conjugations, in stark contrast to the isolating nature of Chinese/Vietnamese. Vietnamese, on the other hand, shares significant typological features with Chinese, including SVO word order (cf. Tibetan SOV) and noun classifiers/measure words (cf. Tibetan classifiers, which are never used for counting).

As for social/cultural predispositions, China has historically had closer ties with Vietnam than with Tibet, so there is perhaps a tendency for the average Chinese speaker to view Vietnamese more favourably (there'll naturally be great regional/individual variation).

Anecdotally, I've heard that Arabic, German, French, and Russian are considered very difficult for Chinese speakers. Many also say that they find Chinese the most difficult, which might refer to the writing system, the continuum between literary and vernacular Chinese, and/or the internal diversity of Sinitic 'dialects'.

Answer 2

The language similarity (typically stemming from common origin) is a huge factor.

As you pointed out, the first tier of languages is exclusively Romance and Germanic. English is originally a Germanic language which at certain points acquired a huge number of vocabulary of ultimately Latin origin. Thus it is similar in structure or in vocabulary to the languages in the first tier. Furthermore the languages are spoken in Western Europe (at least originally - Afrikaans comes from Dutch), so not only are they similar intra-lingually, but also they tend to capture the similar extra-lingual realities, which due to the long-term contact between the nations, they tend to do in a similar way.

Second tier contains some Slavic languages, which have very unrestrictive phonotactics of consonants and complex inflection of both verbs and nouns. So they are more difficult for English speaker to learn. My native language is Czech, so it is comparably easier to me to learn Russian or Serbian than to an English person (but it is still quite simple to learn English). There are even some comparably complex languages that a Czech person does not have to learn to understand, like Slovak (which is mutually understandable) and Polish (here you need to strive a little but you do not "learn" the language as much as you are "getting used to it").

Third tier is languages with different origin and wildly different grammar, also with little shared contact and many different realities. So they would be presumably equally difficult to learn for me as a Czech speaker as they would be for American English speaker.

Also it is necessary to define a little what it means to learn a language. IMHO for the utterly pragmatic purposes, the most important thing is understanding as it is the most demanding part of communication (if you express yourself incorrectly, the native speaker is still likely to understand a lot by inferring). And the more similar the language, the easier the understanding - getting back to Czech and Slovak: most Czech people would be unable to speak Slovak, but practically all of them understand it without effort or learning on their part. I imagine the similar goes for e.g. Portuguese and Spanish - due to the similarities it is easy for you to understand and then when you speak, however bad, the natives will most likely get the meaning somehow.