Kanji are fun to learn, because they are constructive to some degree, and actually pictorial to some degree.
If you can imagine a language where 2k+ emoji are used as parts of words, with all the combination rules which emoji have, that would give you some idea.
But it does tax your memory (nothing compared to Chinese, though!), and takes time when writing by hand. Typing is significantly easier because a reasonable IME gives you variants to choose from when you type the pronunciation.
A good chunk of the roughly ~2100 Kanji have two versions of expression. Some of them have upto 8-10 ways of usage (although very few). As a Japanese resident,
(1) It isn't fun. Not everyone of us knows the whole set and have to keep a digital dictionary in smartphone
(2) We don't assume the names of people based on their Kanji, because (surpise!) people do get offended by minor changes in pronunciation e.g Yamasaki vs. Yamazaki, with similar Kanjis.
That's why I always prefer to inscribe my Japanese notes with Kana in a superscript wherever needed. It is very tight in grammar but the language is not easy by any means. In fact, same goes for Chinese & Koreans.
Some make direct sense like the tree/forest, others you have to deep dive into their history for them to make sense (if at all).
For example, the character for people (民) comes from the image of a person being pierced through the eye, which was done to mark slaves in ancient China. Eventually the character and meaning evolved to the way it's used now.
So you can't say "rest" without mentioning a tree? Finnish is funny in similar ways due to its ubiquitous forests. To "hunt" is "metsästää", "metsä" being "forest" the place you went to look for food.
You can say you metsästää for mushrooms (and other edibles) in the forest too, though, can't you? So maybe it's actually more of a direct correspondence to "forage". And, is that perhaps derived from / related to "forest"? (Or fodder?)
It's not that bad, most sequences of kanji have just a single (common) way to pronounce them.
Although some sequences are completely new, so you need to figure out which word ends where.
And the most commonly used kanji also have the highest number of different pronunciations, sometimes in several ways that are impossible to tell apart grammatically (or even semantically, obviously this is almost never annotated, because adding the pronunciation is for words the author thinks you don't know, even when the pronunciation is entirely unambiguous*)
*: No I'm not bitter I had way too much trouble figuring out how to annotate Japanese text with the pronunciation to make it vaguely readable, why do you ask?
Most Kanji have at least two ways to pronounce them, the On reading and the Kun reading. For example, 水 (meaning water) can be pronounced as both “Mizu” and “Sui.”
Some have a lot more. 下 has two On readings (Ka and Ge) and several Kun readings (Shita, Shimo, Moto, Sa, Kuda, and O).
While combinations typically do have one reading, some can have multiple readings, especially people’s names. Still, it is hard for learners to know which reading for individual Kanji’s are the right ones a lot of times.
EDIT: Fixed a couple of typos and a premature submit.
While those readings might be technically possible, usually only a few of them form a known word. For instance 水 in isolation is (almost?) never pronounced 'sui'. To say it has 2 different pronunciations is a bit like pointing out 'ou' has about 4~5 different pronunciations in English.
Thanks for your reply. Maybe I can help clear this up a bit.
> "While those readings might be technically possible, usually only a few of them form a known word."
Are you suggesting that these various readings are academic and not commonly used? If so, that is incorrect. All the readings are used in common words.
Here are examples of both the On and Kun readings for the kanji 水 being used:
水道: Pronounced "suido" and means "water supply" or "water service."
水着: Pronounced "mizugi" and means "swimsuit."
Here are examples of both the On readings for the kanji 下 being used:
下降: Pronounced "kakou" and means "descent."
下水: Pronounced "gesui" and means "sewer."
Here are examples of all the Kun readings for the kanji 下 being used:
下着: Pronounced "shitagi" and means "underwear."
下々: Pronounced "shimojimo" and means "commoners" or "common people".
法の下 (can also be written as 法の元): Pronounced "hounomoto" and means "under the law."
下さい: Pronounced "kudasai" and means "please."
下りる: Pronounced "oriru" and means "to get off."
Your comparison to the various pronunciations of "ou" in English seems off as well. For example, if you mispronounce "cough" using the "ou" pronunciation of "rough," you'll sound weird but will most likely be understood. However, if you misread "oriru" (下りる) as "kariru," using the On reading "KA" instead of the correct Kun reading, you'd be verbalizing a completely different word. Instead of telling others you're getting off of something, you'd be saying you're borrowing something.
I'd characterize your example of "ou" as a mispronunciation in English, whereas your example in Japanese would be a misreading.
Thanks for the detailed response. I think I do understand your point.
In my (limited) experience so far however I find myself remembering the words themselves, rather than work out their pronunciation from their constituent kanji. To me working from the pronunciations of the kanji themselves is like trying to pronounce 'cough' by fitting together pronunciations for 'c', 'ou', and 'gh' (all of which have several options interestingly enough).
If you can imagine a language where 2k+ emoji are used as parts of words, with all the combination rules which emoji have, that would give you some idea.
But it does tax your memory (nothing compared to Chinese, though!), and takes time when writing by hand. Typing is significantly easier because a reasonable IME gives you variants to choose from when you type the pronunciation.