I feel like we're getting somewhere now. I can see something I can use, but it's column one of a database. How do we get the words out of there and into a garden variety perl array without creating a sea of mojibake?

To read freqrnc2011.csv into a Perl data structure, one could use Text::CSV_XS, which is smart enough to auto-decode UTF-8 bytes into Perl wide characters by default:

use Text::CSV_XS 'csv';
my @words = map { $_->{Lemma} } @{
 csv in => "freqrnc2011.csv", headers => "auto", sep_char => "\t"
};
[download]

Make sure to set an :encoding(...) PerlIO layer on your STDOUT when you work with (and print) wide characters.

I'll have to admit that wasn't able so far to read the .var files (which seem to contain the actual words mixed with binary data when read as CP-866) from the latter source without the use of Starling for DOS from the same website. We may have to contact the original author's son about the file format if you are interested in dictionaries from there.

perlunitut

$ perldoc path::tinY |ack -in slurp
      $guts = $file->slurp;
      $guts = $file->slurp_utf8;
[download]