Now it's too late to change but Japan could have done it, just as Korea did. About homophones:
The closest you can come is 'フオク', which would be pronounced 'fuoku'.
But, you can modify the vowel in katakana with a 'little letter', giving you the common way to say 'fork': フォーク, which sounds very similar to the English pronunciation.
It's not a true alphabet, because you still can't whip together unpronounceable crap like we can do in English, but it's better than having no way to represent out-of-language sounds.
But the reason for this was that a lot of Japanese vocabulary originates from Chinese, which is tonal, whereas Japanese is not. Kanji are used because they are necessary for Sino-Japanese vocabulary. Native Japanese vocabulary has no need for them.
This is a common misconception (which I shared until I studied Japanese).
Writing a sentence may be more of a hassle than with an alphabet based system - and even that would remain to be seen - but reading it is much more efficient due to the more condensed form factor: a page of kanji would take much more space written in romaji.
The fact that Japan and China stuck to an ideogram-based writing system did not stem from a lack of alternative. Japanese actually has had not only one but two syllabic alphabets for more than a millenia (1). Yet kanji were kept, notably because of the numerous homophones in the Japanese language.
(1) http://en.wikipedia.org/wiki/Hiragana#History http://en.wikipedia.org/wiki/Katakana#History