t may perhaps illustrate the
origin as well as the decay of human speech. The only question that it
raises for us is the possibility of distinguishing our own homophones
by accentuation or by slight differentiation of vowels; and this may
prove to be in some cases the practical solution, but it is not now
the point in discussion, for no one will deny that such delicate
distinctions are both inconvenient and dangerous, and should only
be adopted if forced upon us. I shall assume that common sense and
universal experience exonerate me from wasting words on the proof
that homophones are mischievous, and I will give my one example in a
note[8]; but it is a fit place for some general remarks.

[Footnote 8: The homophones sun = son. There is a Greek epigram on
Homer, wherein, among other fine things, he is styled,

[Greek: Ellanon biotae deuteron aelion]

which Mackail translates 'a second sun on the life of Greece'. But
_second son_ in English means the second male child of its parents. It
is plain that the Greek is untranslatable into English because of the
homophone. _The thing cannot be said._

Donne would take this bull by the horns, pretending or thinking that
genuine feeling can be worthily carried in a pun. So that in his
impassioned 'hymn to God the Father', deploring his own sinfulness,
his climax is

But swear by thyself that at my death Thy Sonne
Shall shine as he shines now,

the only poetic force of which seems to lie in a covert plea of
pitiable imbecility.

Dr. Henry Bradley in 1913 informed the International Historical
Congress that the word _son_ had ceased to be vernacular in the
dialects of many parts of England. 'I would not venture to assert (he
adds) that the identity of sound with _sun_ is the only cause that has
led to the widespread disuse of _son_ in dialect speech, but I think
it has certainly contributed to the result.']

The objections to homophones are of two kinds, either scientific and
utilitarian, or aesthetic. The utilitarian objections are manifest, and
since confusion of words is not confined to homophones, the practical
inconvenience that is sometimes occasioned by slight similarities may
properly be alleged to illustrate and enforce the argument. I will
give only one example.

[Sidenote: Utilitarian objections not confined to homophones.]

The telephone, which seems to lower the value of differentiating
consonants, has revealed unsuspected likenesses. For instance the