Dutch. A linguistic history of Holland and Belgium

[p. 36]

4 Spelling

Few European languages have such a logical, economic spelling system as Dutch. Even German, whose spelling is a godsend to the English-speaking student after his years of hassle with the spelling of his mother-tongue, has cause to envy Dutch spelling. The spelling of Dutch as we know it today has of course developed over a long period and controversy about certain aspects of it still has not ceased. Such issues will be dealt with later in this chapter.

The layman might feel inclined to say Dutch spelling is almost completely phonetic and to a great extent this comment is not far from the truth. However, if the linguist looks in detail at how the written symbols of the language relate to the spoken word, he soon discovers many non-phonetic spellings and recognises too that a fully phonetic spelling system is not necessarily a desirable thing. For example, Dutch has in common with German that final b and d are pronounced as p and t respectively i.e. they are devoiced in Auslaut: to write web and bed as wep and bet would not be desirable given that the plural of these words, both in writing and in speech, is webben and bedden. Here the eye, not the ear, seems to demand a consistency of spelling even though it is at odds with what is actually said.

A truly phonetic spelling system for any given language would be one that had a separate written symbol for every individual sound. In fact, what one loosely calls ‘phonetic’ spelling is correctly called ‘phonemic’ spelling i.e. one where each significant difference in pronunciation is reflected in the spelling, not simply every difference in sound. For example, the different pronunciation of the s in English ‘house’ and ‘houses’ is not a phonemic difference and thus a different spelling for the two separate sounds is not considered necessary. Similarly, phonetically speaking the k sound in ‘cat’ is different from the k sound in ‘kit’, as a result of the back and front vowels following the k in each word, but this difference is not significant to meaning; it is merely determined by the different phonological environment in the words concerned.

Yet another reason for avoiding a fully phonetic spelling is homonyms i.e. similar sounding words with different meanings e.g. calf (of the leg) - calf (young of a cow). For historical reasons (see p. 145-48) the Dutch can render the diphthong [εi] with two spellings, ei or ij. Retention of this distinction in spelling from a time in history when the two were pronounced differently, often helps the eye to distinguish homonyms e.g. wij (we) - wei (meadow), lijden (to suffer) - leiden (to lead). Similarly the sound [ɔu] can be written two ways in Dutch for historical reasons i.e. au or ou. This distinction can also occasionally help distinguish homonyms e.g. gauw (soon) - gouw (province).

We have thus seen that etymology can often be the reason for various spellings of the same sound. Having looked at some of the complications of a phonetic spelling

[p. 37]

and the reasons why even a phonemic spelling is not always possible or desirable, we shall now examine the basic rules of Dutch spelling and come to appreciate its economy and consistency.

The most striking feature of Dutch spelling is its economy of letters and nowhere is this better illustrated than in the way it reproduces the vowels of the language. It is immediately evident whether the vowels in any given word are long or short. Whether the long vowels are written double or not to indicate length depends on whether they stand in open or closed syllables (see glossary) e.g. boom (tree) - bomen (trees): in the singular the long vowel occurs in a closed syllable and thus the vowel is doubled, whereas in the plural it occurs in an open syllable and a second o would be a superfluous extra indication of length.^1. Equally the word could not be pronounced with a short o because a double m would then be used e.g. bom (bomb) - bommen (bombs). The same rule applies to a, e and u e.g. paal (pole) - palen (poles), peer (pear) - peren (pears), muur (wall) - muren (walls). Compare kat (cat) - katten (cats), bed (bed) - bedden (beds) and bus (bus) - bussen (buses). Only i employs a different system: short i is written as such e.g. pit (seed) - pitten (seeds), and long i is written ie e.g. mier (ant) - mieren (ants).

This alternation of double and single vowels and consonants to indicate long vowels, although illustrated above only by contrasting the singular and plural of nouns, is also found in verbs and adjectives.^2. For example, infinitives all end in -en and can look as follows: varen (to go by ship), leren (to learn, teach), kopen (to buy) and huren (to hire). They are conjugated in the present tense as follows:

ik leer	wij leren	ik koop	wij kopen
jij leert	jullie leren	jij koopt	jullie kopen
hij leert	zij leren	hij koopt	zij kopen	etc.

The same alternation of a/aa, e/ee etc. also occurs in adjectives when inflected e.g. kaal (bald) - kale, geel (yellow) - gele etc.

Double consonants always indicate that the preceding vowel is short e.g. laf (cowardly) - laffe, spatten (to splash) - ik spat, potten (pots) - pot. It is impossible to have a double consonant (i.e. the same consonant) at the end of a Dutch word. For example, the second and third persons of the verb in the present tense normally take a -t ending but a verb like zitten (to sit), whose stem already ends in t, forgoes the ending e.g. ik zit, jij zit, hij zit. A second t would be superfluous to indicate the correct pronunciation.

At this point we touch on a spelling problem which has tormented Dutch

1. Two notable exceptions to this otherwise hard and fast rule are waarom (why), a compound of two separate words, and tweede (second), where the dropping of an e could obscure its relationship to twee (two).

2. The diminutive forms of café and auto (car) are interesting examples of Dutch spelling: cafeetje, autootje i.e. long vowels in closed syllables must be indicated by doubling. Also the plural of foreign words ending in a, o and u illustrates an interesting variant i.e. firma's, auto's and paraplu's, replacing *firmaas, *autoos and *parapluus where once again the doubling of the vowel in such obviously foreign words, although in accordance with the spelling rules of Dutch, would disturb the eye too much. The apostrophe thus literally stands in place of a letter which has dropped out.

[p. 38]

school-children, and many adults as well, for a long time. Because final d is pronounced as t, confusion often arises as to which letter is required. Nowhere is the problem greater, however, than in the spelling of verbs. Whereas zitten, as illustrated above, does not require a second t in the second and third persons singular, a verb such as verbranden (to burn, trans.) does require t because the sight of dt at the end of a word does not disturb the Dutch eye i.e. ik verbrand, jij verbrandt, hij verbrandt, wij verbranden. The dt cluster sounds, however, like a single t. The issue is further complicated by the past participle i.e. verbrand (see p. 63). There are simple rules for knowing the correct spelling of t, dt or d but nevertheless the Dutch make many mistakes in this regard. It is one of the few instances where the spelling is not phonemic but partially determined by grammar.

Another consistency between the spoken word and spelling is the way in which a final f or s following a long vowel or diphthong^3. become v and z when an e follows in other forms of the word i.e. f and s are then voiced in intervocalic position and their voiced equivalent in the alphabet replaces them e.g. duif (dove) - duiven (doves), vies (dirty) - vieze (inflected form), ik reis (I travel) - wij reizen (we travel). As the examples illustrate, this spelling change also occurs, as do the a/aa etc. changes, in nouns, adjectives and verbs and consequently pervades the whole of Dutch. In effect, as far as the f/v and s/z alternation is concerned, the Dutch are simply expressing in writing something which we also say in English but do not always attempt to reflect in the spelling: compare the pronunciation of ‘house’ and ‘houses’, ‘roof - rooves’.

The above summary of the logic behind the spelling of Dutch is similar to what one would find in any basic grammar of the language for foreigners. There are, however, several more peculiarities and even some inconsistencies which are usually left unexplained, but which are both interesting and important.

The ij^4.: this ligature, along with the double vowels, is the most distinctive feature of written Dutch and enables a novice to recognise a printed page immediately as being Dutch. Although the Dutch for historical reasons have two spellings for the diphthong [εi], namely ei and ij, the latter spelling, because of its origin in long i and also because of the way it is formed in handwriting^5., has become totally confused with the letter y which to all intents and purposes hardly exists in Dutch except in very few foreign words e.g. baby, typisch (pron. i).

The new diphthong ij has been present in Dutch since the end of the Middle Ages, but the spelling ij, to indicate the length of the original long i has been present since the Middle Dutch period i.e. 1100-1500 (see p. 93). In the course of time speakers of Dutch, or one should say writers of Dutch, have come to regard the ligature as

3. This spelling change is also often found after certain consonants e.g. wolf - wolven (wolf - wolves), vers - verzen (verse - verses), gans - ganzen (goose - geese) but kers - kersen (cherry - cherries) and mens - mensen (person - people).

4. This is called lange [εi], a reference to the length of the j in writing, to distinguish it from ei.

5. For the history of the ij spelling for the diphthong [εi] see p. 40. The letter j, which the Romans did not know, was originally a scribal variant of the letter i in much the same way as we used to have a long and a short s even in English e.g. beʃt and was. Thus the ligature ij goes back to a double i, the tradition we still employ with a, e and u in Dutch today.

[p. 39]

one letter, pronounced [εi], This attitude is already evident in Middle Dutch manuscripts and the earliest printed texts (latter half of the fifteenth century) as is the habit of leaving the dots off, thus writing a y instead. Consequently, when saying the alphabet most Dutch people pronounce the letter y as [εi], although the letter y is strictly speaking called i grec or ypsilon. An important ramification of this alternative way of regarding the ligature ij (which by the way is a separate key on a Dutch typewriter and when used in a situation where capitalisation is necessary both the I and the J are capitalised or else a capital Y is used) is that library catalogues, encyclopaedias and telephone books arrange all words containing ij alphabetically under y^6.; this is not the case in Belgium, however. Such inconsistency can often be confusing, even to the Dutch. Whatever attempts have been made to insist that ij is not y, the practice continues and seems to be here to stay, even though dictionaries now all place ij under i.^7. I recall having seen the French loan word bijouterie (jewellery counter) in various stores spelt byouterie and yet this is an instance where the pronunciation is definitely i + j i.e. [i.ʒ]. For years there has been debate as to whether ij belongs under i, under y or should be regarded as a 27^th letter, occurring even after z in the alphabet. Swedish and Spanish, to name but two notable examples, know similar problems with additional letters and have found various solutions to them.

Finally a word about several other minor peculiarities of Dutch spelling. There are two spellings for the gutteral fricative [X], namely ch and g. The distinction in spelling is a remnant of a time when the latter was a voiced fricative, as is still the case in southern dialects, but in the north both are now pronounced as voiceless fricatives. The falling together of those two sounds is one of the factors which has given rise to the commonly heard comment ‘Dutch is such a gutteral language, isn't it?’ (see chapter on pronunciation p. 51).

Dutch also knows the grave and acute accents in its spelling, and not only in French loan words. The acute accent in particular is often used to show emphasis, where an English text may underline or use bold type; it is also used to distinguish homonyms where confusion can arise e.g. een -a, één - one; voor - for, vóór - in front of/before.

Dutch makes extensive use of the diaeresis. The Dutch name, deelteken (lit. part sign, also called trema), explains its function. When two vowel signs representing two separate syllables may lead the reader to pronounce them as one sound (whether as a long vowel or a diphthong), the vowel of the second syllable bears a diaeresis to indicate this e.g. zoëven, financiën, tweeëntwintig. It is sometimes used on o's as well, but will usually only be found on e's.

Syllabification in Dutch also differs from English. Whereas English splits words according to the semantics of the component parts e.g. regist-er, Dutch splits entirely according to sound, e.g. har-ten (hearts), mees-ter (master), lo-pen (to walk). The rule is simply that the break-away syllable(s) must always begin with a consonant; where double consonants are concerned, one stays behind and the

6. Afrikaans knows only the y spelling and places it under ij in the alphabet.

7. Van Dale's Dutch dictionary, generally accepted as the ‘Oxford’ of Holland, has a note at the point between the lemmas i-grec and ij to the effect that it is incorrect to regard ij as y in alphabetical lists.

[p. 40]

other precedes the break-away syllable on the new line e.g. belet-ten (to prevent), kat-ten (cats).

Compound nouns are usually always written as one word. In certain specific instances hyphens may occasionally be used e.g. auto-ongeluk (to avoid a cumber-some double oo), West-Duitsland (a geographic name, although the corresponding adjective is written Westduits). There is, however, and I feel I should add unfortunately, an ever growing tendency to write compound nouns as separate words, even when medial sounds are used to complete the compound e.g. klanten service, bibliotheeks uren, stads vernieuwingsgebied. Whether this is simply due to carelessness, influence of English or a genuine belief that there is nothing wrong with such spellings, I have been unable to ascertain. The practice is not just limited to informal writings by simple individuals. The example bibliotheeks uren was taken from an official sign on the door of the university library in Utrecht. Whether compound nouns are written as one word, two words or hyphenated is indeed a problem for all in English; the Dutch had an ideal solution but seem to be heading into the same confusion we find ourselves in. The practice should not be copied.

The recent history of Dutch spelling

The Spelling De Vries en Te Winkel (1860-1934)

From the 1860's through to 1934, the official spelling system of Dutch which had received government backing, was that of Matthijs de Vries and Lammert Allard te Winkel. The former was a professor of Dutch in Groningen and later in Leiden while the latter was a teacher at a Leiden gymnasium. They were working on a definitive dictionary of Dutch for both the northern and southern Netherlands. Such a task was only feasible if there was agreement on spelling and thus in 1863 De Vries and Te Winkel produced Grondbeginselen der Nederlandsche spelling. Both men had a thorough knowledge of philology and thus although they did not break completely with the spelling traditions of their predecessors, they did apply certain historical principles which had hitherto often been ignored or not thoroughly understood. Some of these are still present in the spelling today.

It has already been mentioned that the diphthong [εi] can be spelt either ei or ij, depending on the origin of the word. This distinction was originally a recommendation of De Vries en Te Winkel. They argued that any word containing an [εi] that was derived from an original long i, should be written ij e.g. wijn < wîn, karwij from French carvi (a plant). The ei spelling, they maintained, should represent the [εi] diphthong in all words where the origin was either an original Germanic diphthong or of any origin other than long i e.g. bereid -compare Gothic raidjan, vallei from French vallée.

De Vries and Te Winkel were also guided by etymology when they laid down the rules for e/ee, o/oo and sch. Because in certain dialects, notably Rotterdams and Zeeuws, long e and o were and are pronounced differently according to whether they were derived from an original diphthong or were simply long vowels, this was reflected in the spelling. By comparing the following Dutch words with cognate forms in German, one can see what spelling was required: heeten, beenen, weezen versus heißen, Beine, Waisen but deken, gele, bede versus Decke, gelbe, Bitte;

[p. 41]

droomen, boomen, koopen versus träumen, Bäume, kaufen but boter, zonen versus Butter, Söhne. Similarly, that ruischen was written sch and bruisen was written s (also Nederlandsch, mensch etc.) was also to be traced back to etymology; here again comparison with German clarifies the situation somewhat i.e. rauschen but brausen.

There were, however, many instances where the origin of words was not clear but the main objection was that the common man had no idea of etymology anyway. And as far as the e/ee and o/oo issue was concerned, it was based on a distinction in pronunciation that only a minority of speakers applied; in the ABN of most of the country the two sounds had fallen together long ago.

De Vries and Te Winkel were also guided by historical principles when they, like generations before them, insisted on case endings still being used with articles and adjectives; this was also an artificial distinction which had long since disappeared from the spoken language and which thus caused great difficulty e.g. Ik heb haren vader met zijnen hond in het park gezien.

De Vries and Te Winkel also allotted a masculine or feminine gender to all common gender nouns, once again basing their rules on historical principles which the common man had no knowledge of or natural feeling for. With few exceptions (see p. 62) all common gender nouns had come to be regarded as masculine when they needed to be replaced by pronouns i.e. hij/hem were used in such cases, seldom zij/haar.^8.

It is obvious from the above summary of the basics of the De Vries and Te Winkel spelling that there must have been a great deal of opposition. Nevertheless it received government approval and managed to maintain its authoritative position for some seventy years. Opposition in the nineteenth century culminated in the foundation of the Vereniging tot vereenvoudiging van onze schrijftaal by R.A. Kollewijn in 1893. He cast aside most historical principles and based his spelling on general cultivated speech. This became known as the Spelling Kollewijn. Kollewijn had many followers but his spelling did not gain government approval.⁹

Little official notice was taken of the newer simplified spelling until 1934, and even then only in Holland, not in Belgium. This was during the period of office of the minister of education H.P. Marchant. The so-called Spelling Marchant, which was adopted by royal decree in 1936, did not go quite as far as Kollewijn had recommended. The difficulties of e/ee, o/oo and s/sch were removed and case n's were only to be written where they are heard in ABN i.e. in standard expressions such as op den duur, goedenavond etc. There was, however, one annoying exception to this which was not abolished till 1947: articles and adjectives standing before words designating singular masculine beings or the names of animals which designate only the male of the species, still required an n in oblique cases e.g. bij den

8. Virtually all pre-war books are printed in the spelling of De Vries and Te Winkel but apart from looking somewhat antiquated, their spelling is no bar to understanding the text for a modern reader.

9 When Afrikaans replaced Dutch as one of the two official languages of South Africa in 1925, the simplified spelling of Kollewijn was adopted immediately; South Africa is still far ahead of the Netherlands in the degree to which its spelling is phonetic, particularly with regard to foreign words, a problem which still exists in Dutch.

[p. 42]

man but op de stoel (formerly also op den stoel), van den aap (masculine) but van de muis (feminine).

This final remnant^10. of the etymologically based spelling of De Vries and Te Winkel was eradicated in Holland by the so-called Spellingwet of 1947 and in Belgium by the Spellingbesluit of 1946. In 1947 a committee comprising both Belgians and Dutchmen was given the task of compiling an official spelling list which appeared in 1954 under the title Woordenlijst der Nederlandse Taal. Ever since that book has been the ‘Bible’ of Dutch spelling in Holland and Belgium; it is known colloquially as het Groene Boekje as all editions thereof since 1954 have appeared in the same green hardback cover. Although this is the spelling used by all government and educational establishments, it is not uncommon for older people educated prior to the war still to write in the spelling of De Vries and Te Winkel.

10. This statement is not completely true because the ei/ij and au/ou distinctions which still exist and which also have certain opponents are also etymologically based.

The voorkeurspelling versus the nieuwe spelling

The spelling which the Groene Boekje recommended did not adequately solve the problem of how to spell foreign loan words, known generally in Dutch as bastaardwoorden. Typical entries in the Woordenlijst for foreign words are:

konvokatie; zie convocatie.

convocatie: ook konvokatie.

tekst (but text is not given as a possibility)

textiel (but tekstiel is not given as a possibility)

However examen or eksamen are both permitted.

To leave the choice to the user was a mistake and the fact that the Woordenlijst was not prepared to give the seal of approval to one spelling or the other has been a continual source of confusion and inconsistency ever since. It was clear that the intention was for foreign words to be gradually dutchified, but the compilers were afraid to push people too quickly.

Eventually a Dutch-Belgian commission was appointed to look into the issue further and its suggestions were published in 1967: Rapport van de Nederlands-Belgische commissie van spelling van de bastaardwoorden. Its Eindvoorstellen (final suggestions) were published in 1969. These recommendations formed the basis of the so-called nieuwe spelling whereupon the spelling as it appears in the Woordenlijst, still the only spelling with government approval, became known as the voorkeurspelling (preferred spelling). To exemplify the sort of suggestions made in this report, let the following suffice:

a.	intervocalic s should be written z e.g. prezident, filozofie
b.	th should become t e.g. tee, biblioteek
c.	total replacement of c by k e.g. aktueel, kollege or by s e.g. sent, sijfer, sukses(< succes) The last recommendation was supported, for example, by the fact that foreign words no longer necessarily regarded as such were spelt like that e.g. sigaret, prinses, porselein.
d.	oe should replace ou e.g. doeane, goeverneur but not in words such as douche and journalist where other foreign sounds that cannot be dutchified are still present.

[p. 43]

The commission's report concludes by expressing the hope that a revised edition of the Woordenlijst will appear at some time in the future in which all its recommendations will be adopted. So far this has not happened.^11. However, looking at spelling tendencies in the Netherlands today, there is a definite trend, whether sanctioned by government and educational bodies or not, towards many of the commission's suggestions.

The student of Dutch, whether native born or not, is well advised to follow the Groene Boekje to the letter until such time as it is revised and the many varying alternative spellings that one meets daily in Holland and Belgium are given official recognition. Many avant-garde publications, such as student newspapers, often use the nieuwe spelling, employing spellings such as kado (< cadeau), logies (< logisch) and sometimes go even further than the commission suggested e.g. tejater (< theater), sosjeteit (< sociëteit).

The ongoing controversy about the spelling of Dutch is not, however, restricted to the spelling of loan words. There are those, and the commission touched on this too, who also wish to introduce radical changes into the spelling of indigenous words e.g. abolition of the historical distinction between au and ou (gauw - soon, gouw - province) and ei and ij (hei - heath, hij - he), as well as writing the unstressed endings -lijk and -rijk as -lik and -rik (lelijk - ugly, belangrijk - important).^12. There is also the perennial problem of final d and dt (particularly in verbal conjugations), both being pronounced t and thus a recommendation that they be written t. Official adoption of such radical changes would seem, however, to be even more remote than adoption of the Dutch-Belgian commission's recommendations for the spelling of the bastaardwoorden.

11. Interesting examples of negative reactions to the commission's recommendations can be found in L. Craeybeckx Sluipmoord op de spelling (1972), which, in addition to Craeybeckx' own opinion, incorporates numerous letters from well-known academies and writers of the day.

12. It is interesting to note that in the more radical spelling of Afrikaans au and ou are both written as ou, -lijk and -rijk are also written with a simple i, but the ei/ij distinction has been preserved, the ij being written as y, however.

Bibliography

KEUKEN, G.J. van der De nieuwe spelling. Thieme, Zutphen. A good brief account of the rules of Dutch spelling as decided in 1947.

Woordenlijst van de Nederlandse Taal Staatsdrukkerij- en Uitgeverijbedrijf, The Hague, 1954. The official spelling list of the Dutch language (see p. 42). On page 42 updates of this list with regard to the spelling of foreign words are also mentioned.