John Cowan <cowan@ccil.org> wrote to me Thu, 08 Nov 2007 about the ratio of L1-L2 speakers in English, well worth quoting.
Needless to say, John is completely right !

. .However, I must take issue with your claim that the position of

English today is comparable uniquely with that of Latin.  If there was

a period when Latin had more L2 than L1 speakers, it was surely during

Imperial times, when Latin was by no means dead but had not yet fully

differentiated into the Romance languages.  There may well have been more

L2 speakers in Gaul, Britain, Africa, and the other western provinces

than native speakers throughout the empire.

 

But this was commonplace in pre-modern empires of every sort.  It was

very likely true of Alexander's empire as well, and provably true of the

Persian Empire, which after abandoning dead Akkadian went on to adopt

Aramaic as the standard language of record-keeping and communication

throughout the empire, for if educated men were rare, actual Arameans

were even rarer.

 

In modern times, there are quite a few languages with more L2 than

L1 speakers: a googling of Ethnologue comes up with Afrikaans, Bulu

(Cameroon), Indonesian, Sango (Central African Republic), Sranan

(Surinam), Swahili, and Thai; there are undoubtedly more.  Swahili is

particularly notable, with an L2/L1 ratio of almost forty to one.

Creoles that have become national languages also have this property:

almost everyone in Vanuatu speaks Bislama, but less than one in twenty

are native speakers.  Admittedly none of these has the worldwide reach

of English, being confined to particular nations, as indeed were the

imperial languages, including Latin, listed above.

 

So I don't think any claim for Latin and English "cross[ing] an

extraordinary threshold" can be sustained.

Thank you John for this note - I stand corrected!