Peter Bleackley scripsit:

> English has a phonology at least as complex as both
> its parent languages (it retains the dental fricatives, which German has
> lost and French never had), and its vocabulary was greatly increased.

True enough, but its morphology was greatly reduced and regularized.
How much this is a result of internal factors, and how much from the
influences of Danish and French, is a question.

At any rate, when we compare English with German we instantly see how
little morphology is left and how comparatively rare the irregulars are.
Excepting Latin and Greek borrowings, there are only about 30 irregular
nouns and only about 200 irregular verbs left, and only a single
declension and a single conjugation survive from OE days.  In German, many
more verbs and almost all nouns are irregular.  Dutch, not surprisingly,
is somewhere between, which perhaps is attributable to its degree of
influence from French: much less than English, but more than German.

As for the dental fricatives, it's interesting that they are preserved
precisely in the languages spoken in Britain (in Greek and Spanish they
have arisen secondarily; Icelandic is a special case); JRRT was of opinion
that this coexistence helped to preserve the sounds in both languages.

All this does not make English a creole, to be sure.

> Irregularity seems to arise partly when several different roots assimilate
> into a single word. This is why "to be" seems to attract irregularity.

It's not so much that it attracts irregularity: it's that the word is
so common in languages that have it, that any subsisting irregularity
will tend to be preserved.  Irregulars are preserved proportionately
to their frequency of use, in general: rare words, like headless and
rootless words, tend to be quite regular.

> From this I'd predict the opposite of your theory - the more invasions,
> the more irregularity.

There's plenty of evidence the other way.  Manchu is far more regular than
the other Tungusic languages, almost certainly due to the hammering it
received from Chinese over the centuries.  Cham, which has been adopted
by speakers of other languages repeatedly over the last two millennia,
is so regular and semantically transparent it looks almost like a conlang;
to a lesser degree, the same is true of Bengali.

John Cowan                                   [log in to unmask]
        "You need a change: try Canada"  "You need a change: try China"
                --fortune cookies opened by a couple that I know