6

As some of you may know, auxlangers tend towards isolating languages. At the very least, the direct object is determined by word order rather than with a case ending (mostly because most West Europeans struggle to understand the accusative suffix in Esperanto). But reading through the Language Construction Kit 2, the author seems to believe the opposite, that fusional languages are easier (he argues they require less rules), and that children pick up agglutinating (though not inflectional) case endings sooner than they do word order (in that part he even explicately states: are you listening auxlangers?) He also has a segment the briefly covers Mandarin grammar, just to demonstrate that configurational languages are highly complex. He claims he shows people a book by the name of 'Mandarin Chinese: a Functional Reference Grammar', which is 700 pages long, to anyone who claims that isolating languages are easy.

But I've been wondering, are they actually any harder than languages with say case endings? I can see some things, like English tending to use words as if they could be any part of speech, and often times a word's meaning changes depending on whether its used as a noun, verb, or w/e.

But I fail to see how having the accusative be specifically marked really helps. If your language is consistently SVO, then any noun that follows the verb must be a direct object (of course things can get muddied if you have ditransitive verbs). If their parts of speech are marked, then why would an accusative marker be necessary? Other than permitting free word order. I could understand how someone who's used to speaking a language with subject and object markers could struggle with this, but really, people who are used to making do without such things struggle to make use or case endings or ad-positions that explicitly mark roles in a sentence.

Besides English (obviously), the only conjigurational language I've ever tried to learn is an Auxlang named 'Kah'. The grammar is obviously inspired by Chinese, though it seems highly simplified. At least it looks that way looking through the grammar. I've never learned it well enough to actually read texts.

Anyway, is Rosenfelder correct in saying that fusional languages are actually easier than isolating languages? He clearly seems to think so. I think at one point in his book he even states that a grammar of even a highly inflecting language full of irregularities has fewer rules than English or Chinese. He doesn't appear to know any inflecting language to any large extent. The only language (besides English), he really gives any indication of knowing is Quecha, which is an agglutinating language with case endings and polypersonal agreement on verbs. It would appear to me that it's also SOV, but pretty much all I know of the language comes from the sample sentences he gives in his own books (he mentions it quite often).

Personally, I think this would be hard to test. I mean, is there really a language out there that's neither isolating nor fusional? I don't see how such a thing could exist. Therefore, you'll find one or the other easier to learn depending on whether your own language is isolating or fusional.

Improve this question
7
  • 3
    My experience of Koine Greek is that fusional morphology can be a nightmare to learn! The same forms have multiple uses and sometimes even multiple parts of speech. The grammar of the more agglutinative Biblical Hebrew is much easier.
    – curiousdannii ♦
    Aug 4 '17 at 5:45
  • 1
    He claimed in his book that case endings were only easier to learn if they were 'agglutinating and regular'. Inflecting languages are clearly harder. In a way, I guess case endings would be easier since the role the noun phrase plays is, essentially, outright stated, rather than implied through word order. Also, in configurational languages it can be hard to tell what part of speech each word in a sentence is supposed to be, because there's so many words that can be of different parts of speech, and often times have very different meanings in each. English is a prime example of this.
    – user19661
    Aug 4 '17 at 7:27
  • 2
    This is very hard to quantify, but I think if you want examples of languages that are easy to learn, look into creoles. Creole grammar tends to be more or less similar cross-linguistically, so many think they're "basic" or "natural" in some sense. For verbs, for example, creoles tend to mark tense-aspect-mood with simple, distinct, multiple, non-bound auxiliaries. Inflections tend to be hard, I think, because they split words into unpredictable word-classes, resulting in the memorization of different forms for the same meaning/grammatical role ("the ablative of X is…")
    – melissa_boiko
    Aug 4 '17 at 9:46
  • 7
    I don't see how having a 700-page reference grammar about a language says anything about whether that language is ‘hard’ or ‘easy’. Even languages that have quite a simple and uncomplicated core set of rules that will enable you to communicate fairly well have heaps and mounds of unexpected irregularities and nuance-distinguishing patterns that you can spend decades trying to decode. There isn't a language on the planet that wouldn't easily support a 700-page reference grammar.
    – Janus Bahs Jacquet
    Aug 4 '17 at 12:50
  • 2
    There are probably studies available what languages are easy to learn and hard to learn for native speakers of English. I wouldn't be too surprised when isolating languages are among the easy ones, but this may be just because they match English in type.
    – jk - Reinstate Monica
    Aug 4 '17 at 13:03
6

L1 difficulty

It is not at all obvious that it even means anything to say that one language is harder to learn (as L1) than another. If some language were really so hard that children simply didn't learn it, then they wouldn't have learned it and it would have died out eons ago. You might imagine that a test of difficulty of learning could be quantified as the mean age at which children acquire Complete Knowledge of the language. An obvious problem with that is that lexical items like "epicene, empirical, tergiversation" which are part of English may not be learned at all (fess up, how many know all 3?), and many people never learn "I would have gone" rather than "I would of gone", or "If I were confused" rather than "If I was confused". I suspect that some (modern, urban) children learning Javanese never really acquire the full language encompassing krama inggil, krama, madya and ngoko.

It's easy to point to specific things that tend to be learned later: CV syllables are manifested first, CCCVCC syllables are later, and that doesn't make CCCVCC syllables "hard to learn". Patterns for which there is scant evidence will not be mastered as quickly as blatantly-obvious patterns. The point is, there is no meaningful metric of "difficulty" for a language (much less a language of some ill-defined "type"). It's unreasonable to expect complete fluency by age 3 or 4.

L2 difficulty

The question of L2 difficulty might be meaningfully addressed, experimentally, provided one controls the relevant variables, by attempting to teach random languages to random subjects and see whether they can acquire fluency (defined somehow) after a given period of time – if a language is intrinsically hard to learn, fewer subjects will acquire fluency, and subjects will acquire a lower mean degree of fluency, as somehow computed.

Some variables to control: native language of subject, similarity of L1 and L2. Amount of exposure that subject has to L2; individual's attitude towards L2-speaking society compared to their own society; pedagogical expertise of L2 instructors (example: on average, French teachers are more effective as teachers than Somali teachers, because French teachers enjoy the benefits of vastly more collective teaching experience than do Somali speakers, who most often bring to the table the fact that they speak Somali). Not to mention the problematic concept of objectively quantifying degree of fluency in a culture- and language-independent way. In other words, although you might sort of imagine a huge experiment to answer the question empirically, I don't see any plausible way to actually conduct the test.

There is reasonable non-quantificational evidence that Spanish speakers tend to learn Portuguese much quicker than they learn Russian, Norwegian speakers learn Swedish much quicker than they learn Saami, and Swahili speakers learn Luganda quicker than English speakers can. The one generalization about L2 acquisition that might have some plausibility is that a person attempting to learn an L2 that is sufficiently similar to L1 will have an easier time than a person whose L1 is substantially different.

Improve this answer
4
  • 1
    You do have a great point. But there are some things I think would make a language harder, even if only negligibly. Let's say we had one language where all verbs were regular. Now let's say we had another language that marked the same things on the verb, and its affixes all had one-to-one correlations with the first language, but it had ONE verb that was irregular. That one irregular verb would on its own create an extra rule that must be learned. Thus the second language would technically be harder than the first, assuming they were otherwise identical.
    – user19661
    Aug 6 '17 at 2:41
  • 1
    So it can be argued that irregularity does increase difficulty, or at least the time it takes to master a language. Oh, and I heard of a comment a Chinese linguist made, regarding the Euro-centric bias of most auxlangs, such as Esperanto. Particular annoyances for Chinese speakers is the grammatical number and tense markings. He opined that 'it was easier to speak without the distinctions your own language makes, than it is to learn new ones that you're not accustomed to'. And personally, I think that is true. I.E. How many people struggle with languages that lack articles?
    – user19661
    Aug 6 '17 at 2:42
  • Though on second thought, I've always been annoyed with the lack of grammatical number in Japanese. I don't mind the lack of articles, but every time I hear a Japanese noun, I automatically assume its singular. And since nouns are more likely to be singular than plural, this can be a very easy habit to fall into to. Though it is interesting that lacking definiteness causes me no problem, but the lack of grammatical number does. I wonder why? Is that normal or is it just me?
    – user19661
    Aug 6 '17 at 2:52
  • Really, you're talking about your own habits in language learning and use. That's cool, and introspection is very useful. However, others' habits are different, and they have their own history and introspective structures. And others have often found different habits, with different degrees of difficulty for different reasons. What I'm saying is you can design your own lang to be as hard or easy as you want, but nothing's guaranteed for other people. I'm a native speaker of English, and I find I prefer speaking Spanish (which I don't speak nearly as well) because I hate English pronouns.
    – jlawler
    Aug 6 '17 at 13:53
0

I'm not sure what Mark actually said, since no direct quotation or link was given.
And if he did say something like that, I likewise have no idea where he got the information.

Questions about "Which (kind of) language(s) is/are easier?" are perennial, especially in English,
since American English speakers are notoriously leery about learning other languages.

But they're really several different questions, all difficult to answer in general terms.

  1. Which (kind of) language(s) is/are easier for a child to learn as the first language?
    (Or is there any difference at all?)

  2. Which (kind of) language(s) is/are easier to learn as a second language?
    (Does it make a difference whether it's before or after puberty?)

  3. Which (kind of) language(s) is/are easier to learn for speakers of <Insert Language>?
    (Almost always, "English" is understood, which is part of the problem right there.)

So far as is known, the answer to all of these is

  • It varies, from person to person, from situation to situation.

The reason is that people, even people who're speaking the same language, are each doing it -- and each learned it in the first place -- in their own individual manners, for their own individual reasons, with their own individual strategies and histories and perspectives and beliefs, and all that may fit in nicely with the way a particular language is set up. Or it may not. There's no way to tell in advance.

That's the reason why some people seem to pick up a language right away while others never lose their native accents in their native language, even after decades away from it. Some people feel more comfortable speaking one language in a given situation rather than a different language; this seems about as normal as feeling more comfortable with certain foods or clothing in a given situation than with other choices, and there's no way of predicting that, either.

We can't even characterize language learning usefully, let alone describe it or predict it; everyone's language habits are uniquely their own; just like their microbiomes.

Improve this answer
3
  • 1
    Here's the paragraph where he mentions this. Note that its in a section discussing L1 aquisition:
    – user19661
    Aug 6 '17 at 1:10
  • "Do these stages vary by language? Absolutely! For instance, Turkish-speaking children start using case suffixes around 2;0; as they're agglutinative and regular, they're learned very easily. In contrast the more complex and inconsistent case endings of Serbo-Croatian are learned later. Agglutinative case actually seems easier to learn than word order (are you auxlangers listening?)."
    – user19661
    Aug 6 '17 at 1:13
  • And how many children, learning how many dialects, from how many socio-economic classes and educational histories were involved in these studies? I would be very hesitant to generalize from findings like this, especially if the studies on Turkish and Serbo-Croatian were not conducted with the same protocols.
    – jlawler
    Aug 6 '17 at 13:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy