Randomata: You've possibly already heard about the immense complexity of the Japanese language both spoken and written. I have not even begun to understand the nuances of the spoken language. All I know is that vernacular usage rarely seems to follow the dictionary's prescriptions, and can vary greatly depending on the situation and who is talking to who.
For example, I asked somebody how to say "I want to sleep" and she explained that you could say "netai," "nemutai," "nemurutai," or "nemuru shitaiimasu" (or something) depending on how 'correct' you wanted to be - and other factors that she couldn't explain. I have confused several native speakers by asking what the correct particle is for a given situation (de, no, ni, o, etc - at, on, of, in, 'n stuff); sometimes they can't decide, and sometimes they disagree with each other.
Also, like all people, the Japanese play with their language, shortening words and blending common phrases in a dynamic way that is extremely hard to capture in a dictionary.
Ok, and then the written language.
There are three systems: Kanji, hiragana, and katakana. (By the way, I am writing this just to convey my ignorance and confused learning, not to display well researched knowledge). Kanji was borrowed from China some 600 or 3000 years ago. Every swish and symbol in kanji can have phonetic/syllabic content, loosely iconographic content, and symbollic content - none of which words mean much to you or me, but each of which I use anyways. Further, a given form can have many many many symbollic meanings ascribed to it. So, one symbol could (theoretically, i mean) mean heart and depth and liquid pump and lungfish, as well as a 'sc' sound. There are thousands of individual kanji symbols.
Hiragana is a blend of indigenous Japanese ingenuity and laziness, originally derived from kanji forms. Katakana was simply a class or style of hiragana hundreds of years ago, but eventually developed into a delineated character set. The kana are nearly true phonetic syllabaries, so there are about 50 symbols in each system, with diacritics for voicing (ie to make 't' into 'd') and specific ways to adjoin the symbols to make sounds like "jo" (jo = shi voiced, truncated by adjoined yo).
Today, hiragana is used for indigenous japanese words, and katakana is used exclusively for those identified as 'loan words,' which have been borrowed into japanese from another language within some undefined span of historical memory. Some of the katakana closely mimics the hiragana. Some of the hiragana seems reminiscient of some of the kanji. All of the kanji is finely detailed and oversignified.
On top of this, romaji or roman characters pop up all over the place. Roman numerals are pretty widely used. Plus, of course, they use different fonts all over the place, which seem to completely change the look of the symbols to a novice like me.
What makes it so tricky is that a given sentence can mix kanji and kanas with no spaces between the writing systems or words, and there is nothing to mark off words with a specific grammatical purpose (like the predicates).
I haven't done a very good job of explaining the whole thing: it's much more impressive and complex than it sounds.