“Words that tear and strange rhymes”

In his youth, Paul Simon thought of himself as a poet:

On a tour of one-night stands
My suitcase and guitar in hand
And every stop is neatly planned
For a poet and a one-man band

He thought about his craft:

My thoughts are scattered and they’re cloudy
They have no borders, no boundaries
They echo and they swell
From Tolstoy to Tinkerbell
Down from Berkeley to Carmel
Got some poems in my pocket and a lot of time to kill

or, rather negatively given that poetry should communicate,

I have my books
And my poetry to protect me
I am shielded in my armor

And surprisingly often he describes problems with the process:

And the song I was writing is left undone
I don’t know why I spend my time
Writing songs I can’t believe
With words that tear and strange rhymes

or

The poet writes his crooked rhyme
Holy holy is his sacrament
Thirty dollars pays your rent
On Bleecker Street

or

My life seems unreal, my crime an illusion
A scene badly written in which I must play

or again

My words like silent raindrops fell
And echoed in the wells of silence

These songs included much of his best work at the time. But then, after a while, he was able to write

Funny how my memory slips
While looking over manuscripts
Of unpublished rhyme
Drinking my vodka and lime

And after that, this theme becomes much less prominent in his work.

For me, things were somewhat similar. Like many people, I wrote poetry in my youth. Julian Jaynes says something like “Poems are rafts grasped at by men drowning in inadequate minds”, but I think I knew from early on that one of the main reasons was to practise my writing, so that when I had something to say I could say it clearly. When Bob Dylan renounced the over-elaborate imagery of Blonde on Blonde for the clean simplicity of John Wesley Harding, I took that as a role model.

Could Simon’s experience happen in mathematics? It is possible to imagine that an important mathematical truth is expressed in “words that tear and strange rhymes”. More worryingly, an argument written in the most elegant style could be wrong, and we may be less likely to see the mistake because the writing is so good.

Recently I have been struggling with a proof in which the trivial case fought back and I had to devise an absurdly complicated argument to cope with it, but all the rest came out quite easily, with the inequalities falling into place. So there can be a mismatch between subject and style, which is not entirely our fault.