This doesn't count words in a file. It almost counts unique lines in a file. What it actually does is list each unique line in a file and the number of times it occurs. This is useful in some situations, and I'm sure it's quicker to do in Perl than in Java. It's hardly a case-insensitive word count.

Is this exactly the code you submitted to solve their problem, or did you retype this from memory?