I vaguely remember that scientific terms exist for two ways to count
loan words in a language, for example the words of French and Latin
origin in English, versus original words from the Germanic heritage.
One method is simply to count the words from a list in which every
words appears once.
The other is to take realistic texts (corpuses) and count what occurs
in them, counting multiple occurences multiply.
What were those terms?
--
Ruud Harmsen,
http://rudhar.com/new