Geoffrey Pullum sifts through the contents of alleged “grammar” quizzes. He is scathing:
if quizzes on chemistry were as uninformed as those on grammar, they would ask silly questions on peripheral topics (“Who is the Bunsen burner named after?”), and would make no reference to the periodic table, or atoms or molecules. The web’s grammar quizzes deal in minor pieces of puristic flotsam with roots in dimly understood 18th-century grammatical analysis.
Why are these quizzes so bad? He rightly blames the incompetence of the setters and their obsession with “fiddly details, gotcha points, and timeworn myths”. But there’s a problem on the other side too: while a proper linguist like Pullum could write a proper test on the features of English grammar, most of the people who took it would struggle. He notes that “few educated people know anything about grammatical analysis”.
We are, most of us, lacking in the theory of grammar. We might vaguely remember a few schoolroom basics, like “a noun is the name of a person, place or thing”. But this is about as useful a definition as “a bird is an animal that flies”. Sure, it identifies the prototypical cases, and it can get a child into the right sort of territory to begin to understand, but the truth is much more complex.
Look at the underlined nouns in this example of Pullum’s:
There’s a great deal of unclarity about what the heck would happen in a crisis during your absence, and there’s every chance of that, for heaven’s sake!
No meaningful definition of “thing” can accommodate these. Instead, you have to define grammatical categories by their grammatical properties. But this knowledge is uncommon.
On the other hand, if you want to test people’s practical grasp of grammar – well, that’s easy. But pretty pointless.