According to the article you say you read:
First you need to find the main dictionary file used by your version of
Office, which will have the suffix .lex.
In the case of Office 97 it will be called Mssp2_*.lex, whereas for Office
2000, the file is called Mssp3*.lex, where * represents your language. For
example, for US, UK and Canadian English, the file is called Mssp2_en.lex
(Office 97) or Mssp3en.lex (Office 2000).
Where this file is stored varies, depending not only on the version of
Office, but also on whether you upgraded from a previous version, did a
clean install, or have multiple versions installed. So the easiest way to
find it is to go into Windows Explorer, press Ctrl+F, and search for
Mssp2_*.lex (Word 97) or Mssp3*.lex (Word 2000).
If in doubt, you can get the full path and file name of the main dictionary
file from the Registry (although even the Registry path to look in varies,
so it's impossible to do programmatically!). This method is covered at the
end of this article.
Thanks for the link for instructions re: excluding apparently approved
spellings.
But, we did miss out on the part of the question that asked what the name of
the data file was; I saw the part about the relevant file also being
referred to as a lexicon, and then found 36 different .LEX files (I think I
can safely exclude the ones in the Adobe folder though). Is one of these
the data file for the Word speller? Or is it a matter of it just being ONE?