Just because you can figure it out by context in songs (rarely upon the first listen, mind you), that doesn’t mean the added cognitive load isn’t excessively burdensome in everyday speech.
Realizing no one's going to change a language with 900 million speakers, do you think it's because there's a lot of ambiguity, or is it because it's a cognitive load people aren't used to? Mandarin is a newer language than Cantonese, and it has fewer tones. Languages tend towards laziness, so I wonder if it settled on the right number, of if it's an ongoing trend.
Edit: About languages losing features, English used to be declined like German or Latin. Only pronouns are declined in modern English, and we don't usually teach it as "pronouns are declined."
Both languages descended from a common ancestor, so you can't necessarily say that one is newer than the other. However, it is the case that Cantonese preserves several features that Mandarin has lost, in particular the complete inventory of final consonants and all of the tone categories of Middle Chinese, which makes it seem better suited for reciting 1000+ year old Tang dynasty poetry where rhyming and tones were especially important.
On the other hand, Cantonese has lost other features that Mandarin has preserved (such as medial vowels and the three-way distinction of initial sibilant consonants), but these features aren't as critical with respect to reciting Tang poetry. For this reason, Cantonese may seem "older" than Mandarin, even though in reality, it's simply that they each have preserved different features and the features that Cantonese preserved happened to make it better for reciting old poetry.
> Languages tend towards laziness, so I wonder if it settled on the right number, of if it's an ongoing trend.
All languages change and will continue to change over time, and while laziness may drive changes in some features of a language, often times other parts of the language become more complex to compensate. This process is called grammaticalization, and is thought to occur in cycles: http://websites.umich.edu/~jlawler/TheGrammaticalizationCycl...
Just like human are newer than, say, some monkey because it comes later, even we evolve from same ancestor, is a matter of fact. The recitation of tang poetry and more complicated speak a lot of this. Mandarin is later.
I suspect what's going on here is that in music, it doesn't matter if you understand it right the first time.
How many songs do you misunderstand the lyrics for on the first few listens, in your native language? For me, in English, I either can't tell exactly what they're saying for some proportion of lyrics, or just totally mishear them _quite_ often (especially depending on the genre).
Music doesn't require every word to be perfectly understandable. Communication does, ideally.
Languages often lose features but they also gain features. Complexity of language is hard to compare, but we can still find many examples.
Modern English has less complex verbal morphology and noun declension (as you mentioned, only in pronouns). But the set of vowels in Modern English is more complex than that of Old English. Also the vocabulary of Modern English has two main sources: Germanic words (native) and French/Latin/Greek words where a single idea can be expressed in either vocabulary source with different nuances. Old English was mostly comprised of Germanic words with some words borrowed from Latin.
Another interesting thing to note is that languages without tones can gain tones (tonogenesis) and in Old Chinese tones played a much smaller role than the modern descendants. This is often the result of syllables/sound systems becoming less complex and losing contrast so the tone of the word becomes contrastive to maintain a distinction between words.
From my understanding Mandarin has a lot of two-syllable words and in many of the words the second syllable doesn't add much, if any, additional meaning.
Contrast that with Cantonese, which I believe still uses a single syllable for most words. (Someone please correct me if I'm wrong)
So it makes sense with less tones, because you have more syllables to disambiguate.
> which I believe still uses a single syllable for most words
Not sure about "most" (depends on the sample distribution I suppose), but single syllable (i.e. character) words are used much more often relative to Mandarin.
So in general you're probably right. Not sure whether that is a cause of more strict adherence to tones in songs though. It could be alternatively argued that the more complex syllable (due to more tones among other things) in Cantonese allowed it to retain single syllable words without having to add extra syllables to clarify any ambiguities.
> It could be alternatively argued that the more complex syllable (due to more tones among other things) in Cantonese allowed it to retain single syllable words without having to add extra syllables to clarify any ambiguities.
