Zamenhof admitted in Esperanto unlimited direct derivation without defining the derivatives arrived at, and further admitted the use of arbitrarily selected suffixes as independent roots which gave Esperanto the character of artificiality. Couturat, for Ido, reversed the process by introducing logic which eliminated direct derivation except from verbal roots. Both de Wahl for Occidental, and Jespersen for Novial admitted direct derivation for a limited number of cases.
The two autonomistic systems followed in indirect derivation the agglutinative pattern using monosignificant affixes. But some words, so derived, lost their international aspect [deleg-it-ar/o = delegation]. Other roots were hidden in the dictionary by accepting as complete new words those which could not be regularly derived [sesion/o = session, but sid/ar = to sit]. Thus the autonomists retained the precision of all derivatives, while de Wahl was able to derive according to the rules of flexion many international words.
The naturalists departed from the full international word rather than the root, obtaining words less precise, less clearly defined, and based on the analogical forms in other European languages. In fact, the naturalists were forced to use a number of different suffixes to denote the same relation [/ario, /ero, /ist]. So far I have not referred by name to Peano's Latino sine flexione because of his sense of humour: He says that the easiest grammar is no grammar at all, and though he does give certain rules based on Latin grammar, he relies in the use of his affixes on complete analogy to Latin or to the Romanic group of languages, refusing to give any definition to the affixes in his language. The quickest way then to learn Latino is to learn Latin and several of the Romanic languages, hardly in accord with the professed object of a constructed auxiliary language. But broadly Latino stands on the extreme wing of the naturalistic school and much that has been said about the naturalistic systems applies to Latino. Jespersen bases his proposals on an entirely novel conception of vowel terminations, a conception which may still play a part in interlinguistics.
The difference between ethnic and constructed languages in the use of affixes
should be clearly borne in mind. In ethnic languages even well-defined
suffixes may be used in conjunction with certain roots, yet not with others
[-ology = the science or theory of... : techn-ology, method-ology, but not in
astr-ology], nor can the suffix be applied for chemical science where we say
blind-ness, great-ness, loneli-ness, dark-ness, but cannot say obscure-ness.
In a logically constructed system a suffix once selected and defined may be
used wherever and whenever it expresses a logical relationship. A detailed
study of ethnic suffixes would be extraordinarily interesting a throw
further light on the problems of derivation. But Skeat remarks in his
Etymological Dictionary that
the number of suffixes in modern English is so great, and the forms of
several so variable, that an attempt to exhibit them all would tend to
confusion.
For the sake of clarity and the practical use to be made from further
researches for the benefit if the final form of constructed language, I should
suggest dividing suffixes into two groups, (1) semantic suffixes which still
impart their whole meaning or part of their meaning to the root, and (2)
those which have, for all practical purposes, ceased to live independently
[as -c-le quoted by Skeat for parti-c-le] or those which fulfil a
purely grammatical function better described as indicative endings [-s
= third person singular]. A method could easily be devised to examine the
suffixes in different European languages, and to determine, through such
study, the principles most likely to give us the easiest and most precise
means of derivation in a constructed language. Again such research would
overlap with the study of our ethnic languages and may well advance the
teaching of language. The Interim Report on Vocabulary Selection of
1936, lists among subjects for research the understanding of derivatives
made by means of prefixes and suffixes. Again, Helen S. Eaton mentions in
her Semantic Frequency List, published in 1940 by the Committee on
Modern Languages of the American Council on Education, that she is now
engaged in compiling an affix frequency study based on words in her
semantic frequency list.
In summarizing the principles of derivation in our selected systems we
have seen that they are modelled more or less closely on the practices of
ethnic languages. It follows that, should we decide to make a language
for specific purposes, we could adapt it easily to its task by choosing its
elements from the material known to us to-day, and by selecting its
structural features according to the purpose we have in mind. We could,
in fact, create a logical, regular, precise, and yet simple means of
communication. And we could reach the state of linguistic progress which
Jespersen foresaw for ethnic languages more quickly and more easily. More
quickly because we understand its direction and mechanism, more easily
because we are freed from the tradition, usage and conventions of the
ethnic languages. Again I quote Jespersen who shows us the way to
linguistic clarity and logical precision:
Human minds in the early times disported themselves in long and intricate
words as in the wildest and most wanton play. Nothing could be more
beside the mark than to suppose that grammatical and logical categories
were in primitive languages generally in harmony . . . primitive speech
cannot have been distinguished for logical consistency; nor, so far as we
can judge, was it simple and facile; it is much more likely to have been
extremely clumsy and unwieldy. Language.
We have seen that the problems that beset the making of a language are the problems presented by our ethnic tongues and that, in fact, the development of a constructed language overlaps with the development of language in general as I have tried to show. Linguistics and interlinguistics no longer develop side by side and apart from one another but tend to become an integrated whole. By neglecting one we risk neglecting the other. The inquiry into every aspect of linguistic behaviour and the problems thrown up by the constant advance of science make it desirable that we employ the whole mechanism which linguistics and interlinguistics together have developed.