I have lots of documents in ISO-8859-15. When I use UTF-8 the high-bit characters are displayed as unknown (reverse-illumination question marks). Is there a way to get UTF-8 to display them correctly?

ISO-8859-15 and UTF-8 are different character encodingss. Any given file will only be one or the other, not both. You can convert between them using iconv.

At some point the Linux kernel started making the default character encoding UTF-8 instead of ISO-8859. I have hundreds of thousands of documents with high-bit-set ISO-8859 characters. Naive users of the VT default encoding display them as unknown characters. I added 'vt.default_utf8=0' as an argument to the kernel to turn this off.

It appears that UTF-8 displays these characters as unknown. I would like to import ISO-8859 for the characters 80-FF. I hoped to find a character set in /usr/share/kbd/consolefonts but didn't. If I composed my own (I would have to figure this out.) could I make this happen?

I think there comes a time when you have to bite the bullet and convert files from obsolete encodings. If they're plain text, then you can batch-process them with iconv. When I adopted Linux, I had hundreds of files that had to be manually converted, as my encoding wasn't recognised!