Roth explains the myth in a fashion of his own. {70a}

Here, then, as Kuhn says, 'we have three essentially different modes of
interpreting the myth,' {70b} all three founded on philological analysis
of the names in the story.  No better example could be given to
illustrate the weakness of the philological method.  In the first place,
that method relies on names as the primitive relics and germs of the
tale, although the tale may occur where the names have never been heard,
and though the names are, presumably, late additions to a story in which
the characters were originally anonymous.  Again, the most illustrious
etymologists differ absolutely about the true sense of the names.  Kuhn
sees fire everywhere, and fire-myths; Mr. Muller sees dawn and
dawn-myths; Schwartz sees storm and storm-myths, and so on.  As the
orthodox teachers are thus at variance, so that there is no safety in
orthodoxy, we may attempt to use our heterodox method.

None of the three scholars whose views we have glanced at--neither Roth,
Kuhn, nor Mr. Muller--lays stress on the saying of Urvasi, 'never let me
see you without your royal garments, _for this is the custom of women_.'
{71}  To our mind, these words contain the gist of the myth.  There must
have been, at some time, a custom which forbade women to see their
husbands without their garments, or the words have no meaning.  If any
custom of this kind existed, a story might well be evolved to give a
sanction to the law.  'You must never see your husband naked: think what
happened to Urvasi--she vanished clean away!'  This is the kind of
warning which might be given.  If the customary prohibition had grown
obsolete, the punishment might well be assigned to a being of another, a
spiritual, race, in which old human ideas lingered, as the neolithic
dread of iron lingers in the Welsh fairies.

Our method will be, to prove the existence of singular rules of
etiquette, corresponding to the etiquette accidentally infringed by
Pururavas.  We shall then investigate stories of the same character as
that of Urvasi and Pururavas, in which the infringement of the etiquette
is chastised.  It will be seen that, in most cases, the bride is of a
peculiar and perhaps supernatural race.  Finally, the tale of Urvasi will
be taken up again, will be shown to conform in character to the other
stories examined, and will be explained as a myth told to illustrate, or
sanction, a nuptial etiquette.

The lives of savages are bound by the most closely-woven fetters of
custom.  The simplest acts are 'tabooed,' a strict code regulates all
intercourse.  Married life, especially, moves in the strangest fetters.
There will be nothing remarkable in the wide distribution of a myth
turning on nuptial etiquette, if this law of nuptial etiquette proves to
be also widely distributed.  That it is widely distributed we now propose
to demonstrate by examples.

The custom of the African people of the kingdom of Futa is, or was, even
stricter than the Vedic _custom of women_--'wives never permit their
husbands to see them unveiled for three years after their marriage.' {72}

In his 'Travels to Timbuctoo' (i. 94), Caillie says that the bridegroom
'is not allowed to see his intended during the day.'  He has a tabooed
hut apart, and 'if he is obliged to come out he covers his face.'  He
'remains with his wife only till daybreak'--like Cupid--and flees, like
Cupid, before the light.  Among the Australians the chief deity, if deity
such a being can be called, Pundjel, 'has a wife whose face he has never
seen,' probably in compliance with some primaeval etiquette or taboo.
{73a}

Among the Yorubas 'conventional modesty forbids a woman to speak to her
husband, or even to see him, if it can be avoided.' {73b}  Of the
Iroquois Lafitau says: 'Ils n'osent aller dans les cabanes particulieres
ou habitent leurs epouses que durant l'obscurite de la nuit.' {73c}  The
Circassian women live on distant terms with their lords till they become
mothers. {73d}  Similar examples of reserve are reported to be customary
among the Fijians.

In backward parts of Europe a strange custom forbids the bride to speak
to her lord, as if in memory of a time when husband and wife were always
of alien tribes, and, as among the Caribs, spoke different languages.

In the Bulgarian 'Volkslied,' the Sun marries Grozdanka, a mortal girl.
Her mother addresses her thus:--

   Grozdanka, mother's treasure mine,
   For nine long years I nourished thee,
   For nine months see thou do not speak
   To thy first love that marries thee.

M. Dozon, who has collected the Bulgarian songs, says that this custom of
prolonged silence on the part of the bride is very common in Bulgaria,
though it is beginning to yield to a sense of the ludicrous. {74a}  In
Sparta and in Crete, as is well known, the bridegroom was long the victim
of a somewhat similar taboo, and was only permitted to seek the company
of his wife secretly, and in the dark, like the Iroquois described by
Lafitau.

Herodotus tells us (i. 146) that some of the old Ionian colonists
'brought no women with them, but took wives of the women of the Carians,
whose fathers they had slain.  Therefore the women made a law for
themselves, and handed it down to their daughters, that they should never
sit at meat with their husbands, and _that none should ever call her
husband by his name_.'  In precisely the same way, in Zululand the wife
may not mention her husband's name, just as in the Welsh fairy tale the
husband may not even know the name of his fairy bride, on pain of losing
her for ever.  These ideas about names, and freakish ways of avoiding the
use of names, mark the childhood of languages, according to Mr. Max
Muller, {74b} and, therefore, the childhood of Society.  The Kaffirs call