The Universal Adoption of Latin Letters

L'Adoption Universelle des Caractères Latins

(Société des Nations. Librairie Stock. Paris 1934.)

[The following is a translation of an article which appeared in the journal Novialiste, no. 4, Nov. 1934, in the language Novial. The article is by Otto Jespersen, and was originally written in French. The translation into Novial was by N. Haislund. Novial to English translation by James Chandler.]

There is no doubt that intellectual cooperation throughout the whole civilized world could be facilitated enormously if the same system of writing were used everywhere; the great diversity of alphabets used is in fact one of the biggest impediments to the coming together of the nations and races.

The very numerous documents, originating from specially authoritative sources, which have been collected in this volume, throw light on all the various parts of this great problem, and show clearly the advantages as also the difficulties in a universal adoption of the Latin alphabet, modified or not. One is often struck to see the great forces which combat a reform in some of the countries which already possess another system. As well as the conservatism which has such deep roots in human nature, one may establish that here and there the force of nationalism opposes the adoption of an alphabet that has been borrowed from another nation; sometimes also religious arguments weigh in the balance; thus in Yugoslavia one sees the strange spectacle of a division of the population into two religious camps which use for the writing down of what is really the same language two different writing systems. One finds almost the same thing in India, where the Hindustani language is divided into a Mohammedan form (Urdu) and a Brahman form (Hindi) which use two completely different alphabets. Also the supporters of the just as strange renaissance of Hebrew consider every attempt to adopt Latin letters as "an act of profanity and of destruction against the spirit of the race." But however powerful these political and religious considerations might be for the peoples and for the practical solution of the problem, I must now abandon them as outside my remit, in order to deal exclusively with the linguistic aspects of the question.

If one were to weigh up the advantages and the inconveniences that would originate from the adoption of the Latin alphabet where it is not now used, it is evident that the interests of the indigenous population of the country in which the reform would be introduced should be considered first, while those of the world generally should be relegated to the second level. For the population concerned, the advantages might be of two kinds: internal and external. Among the former kind one must note especially the ease with which the schoolchild could learn the art of reading and writing of his mother-tongue. In many lands one would gain considerable time: Mr. Tanakadate is of the opinion that in the Orient more than half the time spent at school is dedicated exclusively to learning to read and write, while in European countries, even those which have an archaic and complicated orthography, these studies are infinitely shorter.

With the reform one could also fight illiteracy: in some lands the number of illiterates is really alarming, as one can see in several places in the documents contained in this volume: "one can count on the fingers the Cambodians who have read their literature"; regarding China Mr. Karlgren opines that the proportion of illiterates is 90 per cent.; many similar things are said about the Soviet republics. Mr. Karthaios says that "no Greek - except perhaps some Hellenists - can conscientiously affirm that he is able to write on any subject without a dictionary by his side." All that would be changed, if not totally, at least in large part, by the adoption of a simple alphabet based on the Latin letters as one sees in the beautiful example of Turkey.

Among those things which concern the "external" advantages I should mention that the Latin alphabet would facilitate considerably practical and scientific communications with the outside world, communications that are tending to become more and more urgent because of the development of modern technology. Let us here just mention the possibility of telegraphing everywhere and of using everywhere the same typographic machines and typewriters. Even now one observes that even before the abandoning of the traditional script they are starting in several countries to write for example the names of railway stations and streets with Latin letters alongside the native script; the same is the case for the titles of some periodicals and books destined for an international public, as also to a greater degree with chemical and mathematical formulas. Quotations of French and English phrases always make a bizarre impression in the middle of a text printed for example in Japanese letters, which would naturally be avoided if the whole text was printed in letters of one single alphabet.

It is evident that, in the question taken up by the Institute for intellectual cooperation, they have dealt exclusively with the possibilities for adopting the Latin letters and have not considered the possibility of choosing another alphabet and the creation of a new one for universal use. Nevertheless one should not forget the fact that the Latin alphabet does not fulfil all the conditions that one might wish. It was created originally to represent a single language without any consideration for other languages, and also for Latin one must say that it is not ideal. It has absolutely no system: if the letters b and p resemble each other and seem to indicate that the two sounds are produced by the same organic movements, one does not see from the form of these two letters that the same relation exists between d and t and between g and k. In short, the forms of the letters are totally arbitrary and have no natural connection to the sounds of the language. Nowadays another inconvenience appears in that the same letter is represented by several forms for which we can not see the reason or use: let one compare for example the Roman and cursive types a a, v v, and especially the capital and small letters A a, B b, D d, E e, F f, N n etc.; that which is for some letters - and which should be for all - only a difference in size, is for others a quite considerable difference in form, and I have not yet mentioned the differences in the habitual manner of writing these forms which one finds in different nations, such that instead of learning a single form for each letter, one must familiarize oneself with several; compare for example the way in which the French, the English and the Germans write a capital T.

We must also regret that two nations, Germany and Ireland, have recently begun to favour the use of a changed alphabet seen as national. That which in Germany they call "German script" is really just angular forms of the Latin letters, forms diffused during the Middle Ages and later in many lands (in Denmark, where they were often used till towards the end of the last century, they were called "Danish letters"). In Ireland they have revived a form of the Latin alphabet dating from the depths of the Middle Ages, which was at that time also diffused in England. In these two countries they have thus for nationalistic reasons cut themselves off from the civilized world, by making international communications more difficult in this way.

During the last centuries there has been in Europe just one single universal reform of the alphabet, namely the differentiation of i-j and of u-v - forms which were originally used without phonetic distinction and are now separated so that i and u are vowels and j and v (with w) are consonants. And even this such important reform has not been practised everywhere in a uniform manner: the letter j has four different values in French jour, English join, Spanish Juan, German ja, and the letter v is pronounced in German differently to in other languages.

The Latin alphabet equally possesses too few signs, the majority of spoken languages for example do not find the five vowels a, e, i, o, u sufficient and they also possess consonantal sounds that do not have symbols in the alphabet. In several languages they have introduced supplementary signs like ä, ö, ü in German and in other languages (in Magyar also special forms for the long sounds), å in Swedish, accented vowels in several languages, ç in French and several letters in Czech, in Polish and in Romanian. But unfortunately these modifications are neither identical nor comparable from one language to another, and finally in almost no language are they sufficient to express all the phonemes of the system. One also very often has recourse to digraphs, especially with an h: sh, ch, th or even a trigraph: sch.

Another inconvenience - and a very serious one - is that the phonetic value of the alphabetic signs is not everywhere the same; this is a consequence of the different development of the languages during periods of history. Thus the rounded velar vowel is denoted by u in German, by ou in French, by oo in English and by oe in Dutch; the letter c before i and e is pronounced in at least four different ways, etc. Consequently "the same name appears in the press in the form Cicerin in Italian, Tchitchérine in French, Chicherin in English, Tsjitsjerin or Tjitjerin in Danish etc. It is possible to calculate mathematically in how many ways the single name of the Russian author Tchekhov can be written, because the initial sound (or group of sounds) is represented according to the circumstances by Ch, Tch, C, Tsj, Tj, Cz, C, the middle consonants by ch, kh, k, x; and the final one by v, f or ff".

Nevertheless despite all these imperfections and defects the Latin alphabet is the only alphabet whose universal adoption could be recommended. It is in itself clearer and more suitable for writing and for printing than the majority of the other alphabets, but the argument that is without doubt the most decisive in favour of its universalization is the fact that the alphabet is firmly established in use in the whole western world, so in all countries most important for the whole world's civilization. But so that the adoption of this alphabet in the countries which do not use it could correspond in an effective manner to the aim that we have set out for it, a conditio sine qua non is that the letters must be used in a manner infinitely more consistent and systematic than in the traditional orthographies that exist for example in France or in England.

In the new application of the Latin alphabet to every language we must get rid of anomalies like that of English though and through, or of French seau and sot. Otherwise, if each language uses the signs in a uniform and easily memorizable way, it matters least that the same sign has in all languages the same value. A certain number of small divergences is absolutely unavoidable and can not be considered as very harmful, since one must always remember that an alphabet is destined principally for use by natives and in most cases for those of them that would never learn a foreign language. Concerning the others we must hope that they will be sufficiently intelligent to overcome this difficulty: they will find many others!

In the study that occupies us, it is convenient to establish a clear distinction between the languages that have not or have hardly been written until now, and those that already possess a literature written using a different system of representation to the Latin alphabet. The introduction of these last is of course very much easier in the first case than when it concerns peoples who already have a literary tradition: the older and more diffused it is in the population, the more difficult it is to bring about a radical reform.

The question is therefore relatively simple to solve for the languages which do not have a written literature, as most of the African languages. Concerning the latter there has recently been much progress in creating for a large number of these languages writing systems which merit the highest praise, as they are perfectly adapted to the individual character of the African languages to which they have been applied; see the masterly report of Mr. Westermann.

The languages which already possess a writing system can not all be considered from the same angle. If a large part of the population already know how to read and write, a reform is much less easy than if the majority are illiterate. It is this that has made comprehensible the great success of the highly audacious actions carried out in Turkey by Kemal Pasha, where they have in one step put the Latin alphabet in the place of the Arabic script: the reform has considerably reduced the number of illiterates, and thanks to the new facility afforded by the Latin script which is infinitely better adapted to the phonological system of Turkish than the Arabic, general education has progressed considerably during these last years.

Nevertheless we should not draw from this statement any immediate conclusions about Persia, where the Arabic alphabet has equally been applied to a language to which it was very badly suited. The conditions are here completely different, because Persia is a much more "literary" nation than Turkey and possesses a long poetic tradition of the highest value. By changing the alphabet one might here fear creating a state of things in which the traditional literature became dead letters or merely the monopoly of an intellectual elite who could afford themselves the luxury of learning two separate methods by which to read the mother-tongue. In spite of this we have perhaps no reason to suppose that the situation would become the same as around 1200 years ago, when the Pehlvi alphabet was replaced by the Arabic alphabet, the consequence of which was that all the old literature disappeared in a very short time because the script in which it was composed was not taught to children of the literate classes. Nowadays the danger would certainly not be so great, as thanks to typography one can fairly easily preserve the more precious part of the literature which has until now been written with Arabic letters. In all countries we observe further that the literature which specially merits to be constantly read, is constantly reprinted, while another category is reserved for a small number of erudites, who find it in the libraries. To have two alphabets used simultaneously, one for commercial and everyday use and the other for the "superior" literature would certainly be intolerable on a permanent basis. The script should be the same for everything and everybody.

There exist relatively few languages which use at the moment an alphabet other than the Latin one and for which we could content ourselves with a simple transliteration similar to that which is used in linguistics for Sanskrit and Pali, etc., since the majority of writing systems are not simple enough to permit such a letter-for-letter substitution. This appears with absolute clarity from the report on Modern Greek; here the values of the ancient letters have been so modified through the centuries that a simple transliteration would cause real chaos, while it seems possible, when one applies the principles recommended in the aforementioned report to establish an orthography corresponding to all that one could reasonably demand. For the language of Cambodia one could perhaps act in an analogous way and base an orthography on the current phonological system, reserving a transliteration for old texts and for inscriptions which have hardly any interest for those other than erudites. In many cases it would be ruinous to pay too much attention to the systems used until now by missionaries and authors of handbooks for use by Europeans; too often these innovators have not possessed a satisfactory linguistic education and have applied simply and arbitrarily the orthography of their own language, merely supplementing it with new signs according to their own imagination, sometimes without being able to distinguish all the really important nuances of the languages they wanted to write down. But we must admit than such negligence is becoming more and more rare.

Among the countries which possess a literature we must especially consider China and Japan. We know that the Chinese script is essentially ideographic or more correctly logographic, in that it treats each word as a unit without dissolving it, as we do, into its phonetic elements. In the excellent report of Mr. Karlgren appear very clearly the character of this writing system and the way in which it serves as a purely visual link between the different parts of this enormous country whose spoken language is not at all the same everywhere. One sees also how the means of expressing oneself in the spoken language is undergoing at the moment radical change and separates itself very clearly from the laconic character of classical Chinese. This recent development makes fundamental reform more and more inevitable; but on the other hand it encounters difficulties here perhaps greater than anywhere else, since even if we take as a basis the pronunciation that is beginning to be recognized as the national language, there is a feature specially characteristic to the structure of the language which makes representation with the Latin alphabet extremely difficult, namely the number of distinctive tones which one absolutely must indicate if one wishes to avoid innumerable ambiguities. It would be intolerable to indicate these tones, as is often done in teaching handbooks destined for Europeans, by codes or similar signs placed above or by the side of the letters. Several more or less complicated and more or less ingenious systems have been proposed. But none of them has won full support, and as long unanimity has not been arrived at regarding a practical solution of this thorny problem, the introduction of the Latin alphabet will unfortunately be impossible. And nevertheless the day will come, perhaps sooner than is imagined, when the whole of China and also the whole world will demand the reform which is practically inevitable.

It is true that the reform would mean a complete break with the old tradition and that the literature would become "incomprehensible to everybody who has not studied it as a foreign language." But, is that not already the case in China, and will it not continue to be the case even if the traditional system is rigourously preserved? Each Chinese child that starts to read nowadays is placed in front of a written language as fundamentally different from its natural spoken language that it is at least as difficult to learn as Spanish or Portuguese is for an Italian child; already in the present circumstances is the complete mastery of the literature in China reserved for the small privileged class which can sacrifice many years to the acquisition of this very complicated art.

In Japan the position is comparable but not exactly the same. The traditional script is composed of Chinese ideograms to which are added syllabic signs (Kana) which have become necessary on account of the linguistic structure of Japanese which is totally different from that of Chinese (synthetic as opposed to analytic), the signs indicate partly the flexional and derivative elements, partly the Japanese pronunciation of the Chinese sign (1). The introduction of the Latin alphabet would mean a considerable lightening of this whole complicated apparatus, and there it doesn't seem to bring difficulties comparable to those we encounter in China, because the phonological structure conforms much more to those of the Western world. Two rival systems have been proposed for the use of Latin letters in Japanese: the first elaborated first by Europeans specially for use by foreigners learning Japanese, but it is also said that certain Japanese prefer it (this system is generally called by the name of the lexicographer J. C. Hepburn) - and the second elaborated by some Japanese and generally called Nipponsiki (or Nihonsiki) Romazi (2).

It is perhaps foolhardy for a foreigner to pronounce upon the relative value of these two systems; if in spite of that I dare to express a preference for the second system, the reason is that it seems to me more in keeping with the phonological principles which are beginning to impose themselves on modern linguistics. But in order to explain this tendency I must add some general remarks on the question under consideration, which are so much more natural here because they touch upon a point of fundamental importance, not only for Japanese, but further still for those countries where one might think of introducing a new way of writing the language.

Recently several famous linguists (the Prague linguistic circle, Troubetzkoy, Sapir, Jones, Palmer and others) have established a distinction between phonetics and phonology. These two terms as we understand them now (3), can be defined as follows. Phonetics studies the sounds used in human language, their physiological production by the anatomical organs and the acoustic impression they make on the ear. The system of these sounds is universal, because the organs of speech are essentially the same everywhere, and they may be studied objectively even without understanding the languages in which we meet them. On the other hand phonology has as its object the peculiarities of a language or of a dialect considered separately. Each language has its own system. One in fact discovers that certain sounds which considered objectively appear definitely different, can in a language be used indifferently, either always or in certain circumstances, either before or after certain other sounds while in another language they cannot be interchanged without producing inevitable misunderstandings. That which has a proper value for meaning in one language can thus be regarded as negligible in another. We call phonemes the sounds which are able to play a distinctive role from the point of view of meaning in the special structure of a language in question. Analogically there exist differences of length, of stress and of intonation which play a decisive role in a language, but which the inhabitants of another country completely fail to notice. If we now come to establish an orthography for a language which could correspond to the legitimate demands of those who speak it, it is convenient to pay regard to the phonological individuality of this language and to find the most suitable way of marking that which is identical for native linguistic sentiments, even if, objectively and for a foreigner, it includes several different sounds (4).

Let us summarize the principal ideas of this introduction. To find the best way to apply the Latin alphabet to a language which has not used it before, we must penetrate into that which constitutes the individuality of the language in question and become fully acquainted with its phonological structure. For all the important nuances of the system under examination we must try to discover the most practical manner of notation, always taking into account the imperfections of the Latin alphabet and supplementing it as needed with digraphs or diacritical marks or finally by the special letters used by phoneticists (see the publications of the International Phonetic Association and the report of the Copenhagen Conference of 1925). But a man of a practical nature will always remember that the supplementary signs create difficulties when one wants to telegraph, print or use an ordinary typewriter; sometimes we must therefore content ourselves with a more or less satisfactory compromise. The Latin alphabet is a unity, but its applications are very varied. The universal introduction of Latin letters is full of difficulties and will not be able to be carried out in one step, but overall the difficulties are not impossible to overcome, and the advantages will be very great for the nations which know how to overcome them.

Otto Jespersen.

(Translated from the French by N. Haislund.)

(1) One can read about the development of this chaotic system in the first chapter of An Historical Grammar of Japanese by G. B. SANSOM, Oxford 1928.

(2) One may consult on this question as well as several pamphlets by A. Tanakadate, the pamphlet of the philological society of the Imperial University of Tokyo, Concerning the Romanization of Japanese, 1930, and especially Harold E. PALMER, The Principles of Romanization with Special Reference to the Romanization of Japanese, Tokyo 1930.

(3) Previously these terms have been used in several different significations.

(4) In Japanese the Nipponsiki system is correct phonologically in writing ta, ti, tu, where Hepburn is correct phonetically in writing ta, chi, tsu. Because the script does not indicate certain changes which occur completely automatically in natives, the Japanese system permits among other things, a very simple way of representing the conjugation of verbs, the relation between transitive and intransitive verbs, the alternations called nigori etc. (See Pocket Handbook of Colloquial Japanese, 2nd ed., Tokyo, 192, p. 11 onwards). Overall this system shows very clearly the morphological and semantic structure of the language.


Back to International Auxiliary Languages

This page hosted by Geocities.
James Chandler 18-Jan-98.