The Approaches to Artificial Language

By Harold E. Palmer, D.Litt.
Late Linguistic Adviser to the Japanese Department of Education and Director of the Institute for Research in English Teaching

IN ADDING THIS INTRODUCTORY note to the work of my colleague and friend H. Jacob on International Language, I am stressing the idea of Artificial Language. Let me state at the outset that I am using the term 'artificial' in its strict sense, viz., created by the art of man, and not in the less accurate (and usual pejorative) sense of 'imitation', 'substitute'. I am using it in the sense that we live in artificial houses (as contrasted with the caverns that were the natural houses of our remote ancestors), wear artificial clothes (as contrasted with the furs, feathers, and shells of our remoter ancestors), use artificial tools (as contrasted with the natural implements of the eolithic age), that we ride in vehicles, all of which, from the wheelbarrow upwards, are artificial, and write with alphabets that are every whit as artificial as the most artificial language designed.

By many the whole question of artificial language has been prejudged by the use of the term 'artificial'. To say that a language is artificial is to condemn it. To accompany this term by a sneer is to attempt to prove ipso facto that an artificial language is undesirable, and perhaps impossible.

If, as is said, all roads lead to Rome, there are many roads that lead to the idea of the growing necessity for artificial, as contrasted with the natural or ethnical, languages.

Let us now examine some of these roads and those who travel along them.


Many of those who work for the cause of harmony and brotherhood between nations have said: 'The chief obstacle to universal peace, international entente and cooperation is the lack of a common language; the peoples cannot talk to each other, they cannot communicate one to another their benevolent intentions and so remove distrust and fear. Most natural languages are national languages, which, from their very nature, can no more serve as an international language than can a National Anthem serve as an International Anthem; therefore the International Language must be an artificial one.'

Quoting Raymond Frank Piper in Language and World Unity: 'The most extensively used of all constructed languages, Esperanto (the one who hopes), was deliberately designed as a step towards a religion of universal brotherhood. Its author, Dr L. L. Zamenhof (1859-1917), lived as a boy in Russian Poland amid bitterly clashing races, languages and religions. Here Russians, Poles, Lithuanians, Germans and Jews misunderstood, feared and jostled one another. The people of Europe in fact speak no less than one hundred and twenty languages, all mutually unintelligible; thirty-eight of these are used by more than a millions persons. Then the brilliant idea occured to young Zamenhof that the hostilities and hatreds among peoples might be relieved through intercourse in a common neutral language - neutral in the sense of being free from racial, national, religious and other prejudices. Impelled by this stirring vision of peaceful concourse and eventual universal brotherhood, he proceed to create a great synthetic language. Such a vehicle is painfully needed especially in the Polands and Alsace-Lorraines of the world. It is an indispensable condition of international understanding and harmony.'


The work of international conferences is much impeded for want of a common language. Proceedings are held up at every moment by the intervention of interpreters who translate into one or more other languages what a speaker has said in his mother tongue, and committee work is hampered in still greater degree. And what of the delegate who happens to be ignorant of any of the 'major languages' specified by the organizers of the conference?

'Then let English be the International Auxiliary Language, the Universal Second Language', say some.

But the average delegate says: 'A very satisfactory solution of the problem for those whose language is English, for this means that these people will be forever dispensed from the task of learning foreign languages, that each has the privilege of expressing himself in his mother-tongue - to the disadvantage of the others. It is not equitable that any one people should be privileged to so great an extent. Let the handicap be fair, and the language-learning task be the same for all. And what could be fairer than to ask from each the comparatively light task of mastering an artificial language?'


Natural languages, from their very nature, are exceedingly difficult to learn - or expressed alternatively, take a long time to learn. The reason for the difficulty - or length of time - is the cumulative effect of (1) many unfamiliar speech-sounds, (2) irregular spellings, (3) vocabularies marked by redundancies, deficiencies and ambiguities, (4) irregular and illogical systems of derivation, (5) cumbrous grammatical and syntactical systems.

On the other hand modern artificial languages are marked in general by (1) a minimum of speech sounds, most of them common to all languages, (2) perfectly phonetic spellings, (3) adequate but simple vocabularies, (4) regular and logical systems of derivation, (5) the minimum of rules of grammar and syntax, all regular. For illustrations of these five facilities see the description of various artificial languages in the present volume.

It is, then, only to be expected that an artificial language can be mastered in from one quarter to one twentieth of the time needed for mastering any natural language - and such is indeed the case. The variation between this estimated 'one quarter' and 'one twentieth' depends on what is the learner's mother tongue and which of the artificial languages is the subject of study. The Chinese find any artificial language at least three quarters easier to learn than, say, English; while the English and Russians can master an artificial language in a twentieth of the time needed for learning Chinese. (In my opinion, backed by my experience in learning Esperanto and Ido, the latter is by far the easier language in that its 'five facilities' are greater than those of the former. However, the reader may judge for himself.)

The language-learner then says: 'If I am obliged to take the trouble to learn a foreign language, let it be a language that I can learn in the minimum of time - and that language is an artificial one.'


The saying 'what do they know of England who only England know?' may be aptly paralleled by 'what do they know of English who only English know?' Most educationists probably agree that the nature of the mother-tongue is best understood or appreciated when viewed from the outside. To view one's mother-tongue from the inside is comparable to propulsion in vacuo, to leverage without a fulcrum, or to estimating the value of an internal currency without comparing it with foreign currencies. This is one of the chief reasons for which the study of classical languages has been prescribed as a means of culture or of mental discipline. It is chiefly for this reason that Europeans study Latin or Classical Greek, that the pundits of India study Sanscrit and that the Japanese study Classical Chinese as part of their education. A knowledge of almost any foreign language reveals to us our own. Indeed the only English grammar generally known to the Englishman is that which has been gained as the reflection of the grammars of French or Latin.

Educationists hold it desirable that one should know something of the structure and nature of languages, just as it is considered desirable that the educated person should know something of the function and workings of mathematics or, in more modern time, the basic principles of sociology or citizenship.

Esperanto has been described as 'the layman's Latin', in that it contains, and so illustrates, the grammatical mechanisms of number, gender, case, agreement, etc.

And if it is true to say that those who know only English can know little of English, it is true to say that those who know only natural languages can know little of natural languages. Natural languages are partly rational and partly irrational. Much of natural language is an exemplification of logic and many of its thought categories conduce to orderly and clear thinking. The idiomatic element with which natural languages are shot through and through, clouds the unidiomatic element and mars the whole. Charming as idioms may be as marking historical or literary vagaries, they are as out of place in systems of linguistic symbolism as would be inexact and fanciful symbols in shorthand or mathematics. Quaintness and oddity may enhance the beauty of a work of literature or art, but not the utility of an instrument of precision. Machines are no longer built as they were 50 years ago according to aesthetic patterns, and our railway carriages and omnibuses no longer recall the shape of the old stagecoach.

Architecture would never have found its true development if the architect of a house or temple had been forced to design it as the extension of the inhabited cavern or the sacred grotto or tree of primitive priesthood. The conception of orderly and pure language freed from its inherited bonds and blemishes could never have come into existence without the conception of artificial language - and the Road of the Educationist leads to this ideal conception.


Scientists the world over have common interests and aims. To a certain extent scientific terminology is international. The scientist says: 'All mathematical signs are of artificial creation and are therefore understood universally. Our nomenclatures, including as they do the tens of thousands of zoological and botanical names, the names of mineral and chemical substances, etc., are made up of words arbitrarily derived from various and mixed sources - and are universal. The great nucleus of our international language exists. Regularization is still needed; so that, for instance, the name of the liquid now variously termed petrol, gasoline, essence, benzin, etc. (all created words) shall have one and the same artificial name just as the substance radium has one and only one name, which is also of artificial creation. Let us proceed further, and do for vocabularies in general what has already been done for vocabulary in particular - and the result will be a complete artificial language.'


Two business firms of two different countries, each ignorant of the other's language, wish to carry out a commercial transaction - to order goods, secure a credit, specify conditions. If the transaction is of a simple nature, neither party need use the language of the other, not any national or natural language; they need no translator nor interpreter, for they have recourse to international commmercial codes designed for this particular purpose. These codes consist of thousands of arbitrary groupings of letters, each of which corresponds to some specific word or sentence. These have been created by the compilers of codes, and are of purely artificial origin.

The business man says: 'Our artificial written language exists and works. Let now a code-deviser come along and compile a code of words that can be uttered as well as written - and the result will be a complete artificial language.'

The mariner says something very similar: 'The code of maritime signals was composed long ago; it is in daily use. An almost infinite number of combinations of flags will convey the meaning of almost any message we wish to send irrespective of national language. Quite recently British and Russian sailors were able to communicate with each other successfully in a situation where no other communication was possible. Let the groupings of this code of maritime signals be made pronounceable - surely an easy task for the expert - and the result will be a complete artificial language.'


Those who have made a study of Comparative Etymology say: 'In analyzing the contents of dictionaries we find that at least 7,000 words are more or less common in spelling and meaning to the six chief European Languages: English, French, Spanish, Italian, German, and Russian, not to mention several minor languages. Most of these words have come into these languages not by the process of linguistic evolution from earlier forms but by deliberate acts of creation, borrowing, and adaption. We find that in certain types of texts one-third of the vocabulary is made up of these international words. Let this process of deliberate borrowing be extended to the remainder of the vocabulary - and the result will be a complete artificial language.'


Jespersen has pointed out that a large proportion of our so-called natural languages is made up of an artificial element. And indeed we have only to glance at the pages of any fairly comprehensive dictionary to see how many words have come into existence, especially during the last two centuries, as a result of deliberation followed by decision.

A notable example is provided by modern Japanese. In 1858 the Japanese determined to adopt western civilization, science, and sociology. Their language provided very few of the words necessary to designate western institutions and activities. English-Japanese dictionaries of that period could give no Japanese words for, e.g., bank, telegraph, or gunboat, but merely definitions of such words. Just as European scholars coin new words for new needs by combining Greek or Latin roots, so the Japanese scholars coined the new words needed by combining old Chinese roots. If we, needing a word for it, created the Greek compound automatic telephone (pronouncing it in the English fashion), so the Japanese coined the Chinese compound self-move-lightning-speak-apparatus, which they spelt and pronounced in Japanese fashion ji-do-den-wa-ki. But this is hardly intelligible to the Chinese who in the Mandarin dialect would pronounce it tse-tung-t'ien-hwa-chi.

Let us stress here the fact that few coined words are 'invented words' such as kodak. Coinages of this sort are extremely rare in modern artificial languages, indeed the 45 correlative words of Esperanto are probably the only examples. What the word-coiner does, is to take some existing word or root and to derive the new word from it.

When scientists first observed the power of attraction possessed by amber and other substances when rubbed with silk, they needed a new word to designate it. If they had identified it with the phenomena of thunder and lightning they might have called it, as do the Chinese, 'lightning spirit', or they could have invented a word to designate it, as did the Belgian chemist Van Helmont, who in the 17th century invented the word gas. What they did, however, was to take the Greek word elektron, meaning amber, replace the on by icity, and stretch the meaning to cover the new concept. Indeed, in more recent years, those who wished to coin a word to designate the ultimate unit of the atom took the electr of electricity and reappended the Greek syllable on, using it in the sense of unit, thus restoring the original Greek word but with a completely new meaning.

This is the method used by most compilers of artificial languages - indeed Ido, Occidental, and Novial contain no invented words at all. Therefore those whose business it is to coin words say: 'Artificial languages differ from natural languages only in so far as they are wholly artificial instead of being partly so.'


The chief duties of those who compile dictionaries are to collect data on (1) Polysemia (how many meanings may be expressed by one and the same word); Synonymity (to what extent a given meaning may be expressed by two or more separate words); and (3) Translation (to what extent a given word in a given language may be translated by a given word in some other language).

Thus (1) according to the great Oxford Dictionary the word at has 36 separate meanings; (2) Roget's Thesaurus is a catalogue of concepts with all the ways in which they may be expressed; and (3) cross references in a bilingual dictionary show us that in comparing English and French:

one sense of take=conduire, and that
another sense of conduire=drive, and that
another sense of drive=chasser, and that
another sense of chasser=hunt, and that
another sense of hunt=look, and that
another sense of look=regarder, and that

Hence the need is felt for some organized system by which the units of signification (semantemes) may be identified and listed. The 1,000 numbered 'categories' of Roget's Thesaurus constitute but a step in this direction, for as this classification is based on English words, it is no instrument for a general semantic survey. Semantemes must be mapped out, as are the 'isoglosses' of dialectology, in the form of contiguous areas. (Indeed some years ago I was able to construct a few maps of this sort, and that for the preposition for took three months.)

The lexicographer (or, more properly, the 'semantician') says therefore: 'Finding no common measure of speech-sounds in the letters of national alphabets, the phonetician, disregarding all traditional orthographies, has drawn up a general comprehensive alphabet in which one sign represents one and only one speech-sound, and one speech-sound is represented by one and only one sign. Similarly, finding no common measure of semantemes in natural languages, we must draw up an artificial vocabulary in which one word represents one and only one semanteme and one semanteme is represented by one and only one word. For this some artificial language is needed, and experiments have shown that such an artificial language is possible.'


Pascal, Descartes, Leibniz, Wilkins, and other philosophers and logicians of the 17th century said much the same thing as the lexicographers of the present century, but with a different aim in view. Compilers of dictionaries have in mind students of languages - particularly foreign languages, whereas the philosophers are looking to a more perfect instrument of thought than are natural languages, with their redundancies, deficiencies, and antiquities. Wilkins, indeed, went further than his contemporaries in that he drew up the outline of a philosopher's language of this sort. It is as if the philosophers said: 'Let us find for each concept a word which will symbolize that concept and no other - and the result will be a complete artificial language.'

Thus we note the existence of different roads each with its travellers proceeding in the directions that interest them. Each of the groups of travellers may be unaware of the existence of the other roads or may be indifferent to the needs and aspirations of the other groups. Yet either the roads lead to the same end or their existence proves that the end, viz., Artificial Language, is a reality.

The author of this book describes the various forms in which this end has been conceived, and deals with the subject under the broader heading of Auxiliary International Language.

Back to International Auxiliary Languages

This page is hosted by Geocities.
James Chandler 2001.