Classification by subjects would be an exceedingly
useful method if it were practicable, but
experience shows it to be a logical absurdity. It
is a very difficult matter to classify the sciences,
so complicated are the relations between them.
But with books the complication is vastly
greater, since the same book may treat of different
sciences, or it may discuss a problem involving
many branches of knowledge.
As it was pointed out by Eric De Grolier (1974, 58;
1979), one of the most important scholars of library
taxonomy in the XX century, bibliographic classification
systems are more influenced by the society
which has elaborated them and by political orientations
and ideological views, than by the philosophical
classification of sciences. In De Groher's opinion,
indeed, bibliographical classifications are an "artefact
culturel" and should be analysed also from a
sociological point of view.