Horace; and our so-called "standard" pronunciation is ever
imperceptibly but surely changing, and never continues in one stay.

In the very valuable _Lectures on the Science of Language_ by
Professor F. Max Müller, the second Lecture, which deserves careful
study, is chiefly occupied by some account of the processes which he
names respectively "phonetic decay" and "dialectic regeneration";
processes to which all languages have always been and ever will be
subject.

By "phonetic decay" is meant that insidious and gradual alteration in
the sounds of spoken words which, though it cannot be prevented, at
last so corrupts a word that it becomes almost or wholly unmeaning.
Such a word as _twenty_ does not suggest its origin. Many might
perhaps guess, from their observation of such numbers as _thirty,
forty_, etc., that the suffix _-ty_ may have something to do with
_ten_, of the original of which it is in fact an extremely reduced
form; but it is less obvious that _twen-_ is a shortened form of
_twain_. And perhaps none but scholars of Teutonic languages are aware
that _twain_ was once of the masculine gender only, while _two_ was so
restricted that it could only be applied to things that were feminine
or neuter. As a somewhat hackneyed example of phonetic decay, we may
take the case of the Latin _mea domina_, i.e. my mistress, which
became in French _ma dame_, and in English _madam_; and the last of
these has been further shortened to _mam_, and even to _'m_, as in the
phrase "Yes, 'm." This shows how nine letters may be reduced to one.
Similarly, our monosyllable _alms_ is all that is left of the Greek
_ele{-e}mosyn{-e}_. Ten letters have here been reduced to four.

This irresistible tendency to indistinctness and loss is not,