Language Planning for a New Order 487
be told that the multiplication of word forms by flexions is foremost
among obstacles to learning a language In Chapteis III, V, X, XI,
we have seen that the difficulties are of two sorts:
(i) Some flexions (e g gender, number accord between noun and
adjective) have no semantic value at all and their existence is
an arbitrary imposition on the memory ,
(u) Even when meaningful, flexions which do the same type of work
may show widely different forms
Thus language-planners meet on common ground in recognizing
that a satisfactory auxiliary must have, (a) no useless flexions, (&) regu~
lanty of what flexions it retains About what constitutes regularity
advocates of a constructed language do not diifer. To say that flexion
must be regular means that if we retain a plural, we must form the
plural of all nouns in the same way; if we retain a past tense every
verb must take the same past tense affix In short* a single pattern
of conjugation—a single pattern of declension. To the extent that this
measure of agreement exists, any constructed language offers fewer
grammatical obstacles to a beginner than do such languages as French,
Russian, or German,
Unanimity with reference to what flexions are useful has come about
slowly; and is not yet complete. At the tune when Volapuk and Espe-
ranto took shape, and long after, planners were enthusiastic amateurs
blinded by peculiarities of European languages they knew best Nine-
teenth-century linguists made the same assumptions as nineteenth-
century biologists. They took for granted that what exists necessarily
has a use Awareness of the universal drift from flexional luxuriance
towards analytical simplicity in the history of Aryan languages was not
yet part of their intellectual equipment. None of them recognized
the many similarities between English, which has travelled furthest
on the road, and Chinese, which consists wholly of unchangeable in-
dependently mobile root words Professional philologists, who could
have enlightened them, were not interested in constructive linguistics.
In this setting it was a bold step to sacrifice gender or mood, and the
accepted grammatical goal seemed to be a language of the aggluti-
native type illustrated (Chapter V) by Turkish, Hungarian, or
Japanese.
Intellectual impediments to a more iconoclastic attitude were con-
siderable, and we need not be surprised by the tenacity with which
earlier pioneers clung to grammatical devices discarded by their