I cannot agree that much of what is, these days, called poetry, is in fact poetry.
I base that assessment on a simple rule:
If it isn't memorable, it isn't poetry, q.e.d.
If I read something, and an hour later can remember only that it is something to do with, say, a roof, then it is not poetry.
The traditional poetic device of memorably melodic metaphor marching through the mind reinforces recall. Recall of rhyme recalls the time sublime of childhood play with blocks and bricks of wooden words.
Such is the power of poetry.
The Moving Finger writes; and, having writ,The Rubaiyat
Moves on: nor all your Piety nor Wit
Shall lure it back to cancel half a Line,
Nor all your Tears wash out a Word of it.
By Omar Khayyam
This wonderful verse is so memorable that it is very widely known. As a metaphor for the arrow of time and the hasty action regretted, it cannot by any means be faulted. Unfortunately, as a metaphor for writing, it fails to pass the test of scientific plausibility.
The arrow of time as a 'moving finger' metaphor fails in its application to writing because it is only applicable to everyday speech. Only in ordinary speech are we barred from canceling half a line.
Words once spoken in the ordinary course of human affairs cannot be unsaid.
By writing our words we gain an advantage. We can add, delete, substitute, re-phrase, re-format and re-style to our heart's content. It is the creative shine on an invisible surface by use of invisible polish that makes the difference between the apparent fluency of carefully crafted words and the self-evident disfluency of so much of our daily speech.
This simple fact fools the compilers of prescriptive grammars. For any language, for all its dialects, styles of writing and registers of speech there can only be one grammar of the orthography, syntax, idiom and suchlike factors of that language. Where there is a need for two distinct grammars there will be found two distinct languages. There is no grammar of writing. There is no grammar of speech. There is one evolved and evolving natural grammar.
The greatest influence on the natural diurnal evolution of a human language is its community of speakers and their intercommunication by means of both speech and writing in all of their various forms.
The least significant influence on a language in the long term is the list of rules based on a finite corpus of written materials compiled by a prescriptive grammarian. To base a theory on a subset of data from the physics of ink, paper and parchment whilst ignoring the much greater amount and diversity of acoustic data available is, to my mind, most unscientific. By application of similar methods, by ignoring, say, aerodynamics, one could deduce that only balloons are real aircraft.
Books about writing style are to be encouraged, the more so if they are based on observations about how language really works. But to call such books 'grammars' is an abuse of a precise scientific term. A grammar is a scientific description of the sufficient and necessary conditions under which any average person may communicate ideas by means of natural language. As of now, only partial natural grammars have been inferred or deduced, and none is backed up by repeatable, observer-independent observations.
Whilst I commend the writers of 'grammar' books for their contributions to the teaching of clarity and style, I would remind them of what a great man once said about education:
Education is not the filling of a pail, but the lighting of a fire.William Butler Yeats