The Dutch spoken between 1150 and 1500 is referred to as Middle Dutch. During this period there was no standardization of grammar. Authors generally wrote in their own dialects. Very often it is possible to tell from the language whether a text comes from Limburg, Brabant, or Holland. There was a lot of variation in the spelling. Words were often written as they were pronounced: lant (land), hi vint (he finds). The sound determined the spelling, irrespective of the basic word. This is no longer the case with modern Dutch where land, has a voiceless ‘d’, but is written thus because the infinitive is landen, and hij vindt (he finds) has dt as it is 3rd person singular, thus stem (vind) +t .
|Karel ende Elegast (lines 1-6)|
|Fraeye historie ende al waer|
Mach ic v tellen hoort naer
Het was op enen auontstont
Dat karel slapen begonde
Tengelem op den rijn
Dlant was alle gader sijn.
A problem with the Latin alphabet was that it was not easy to make a distinction between long and short vowels (a - aa). Various solutions were found. At the beginning of the thirteenth century the word jaar (year) was spelt jar but other variants soon appeared: jaer and jair and later jaar or even yaer and iaer.
Another feature of Middle Dutch is that articles or prepositions were often joined on to the word they belonged to: tjaer (the year) and dlant (the land), as in the accompanying fragment from Karel ende Elegast. The text translates: “I will tell you a marvelous story, and a true one. Listen! One evening Charles was sleeping in Ingelheim on the Rhine. All the land you could see was his.”
There were also regional differences. Thus a clerk in Amsterdam in the fourteenth century would usually write lant, but in Utrecht he would write land. The modern system of double consonants for shortening the vowels was also known: compare tellen (short e) with slapen (long a) in the extract.
The invention of printing led to a more standardized approach.
Siegenbeek thought that the spelling should reflect refined Dutch pronunciation, taking into account the uniformity, etymology, and analogy. From the Siegenbeek spelling reforms, we get the modern Dutch ij (called lange ij (long y) as distinct from the usually identically pronounced ei, called korte ei (short ei)). The word for iron ijzer used to be written yzer. Other spellings from Siegenbeek include: berigt (modern Dutch: bericht / report), blaauw (blauw / blue), Dingsdag (dinsdag / Tuesday), gooijen (gooien / to throw), magt (macht / power), kagchel (kachel / stove), koningrijk (koninkrijk /kingdom), muzijk (muziek / music) and zamen (samen / together).
Siegenbeek's spelling never achieved real popularity. In particular the poet Willem Bilderdijk fought against it, largely out of personal spite. He produced some of his own spellings which were popular in the 1830s and 1840s including the modern kachel (stove), plicht (instead of pligt /duty) and gooien (to throw). However, other spellings of his did not last: andwoord (antwoord / answer), hair (haar /hair/her), ontfangen (ontvangen / to receive), thands (thans / at present) and wareld (wereld / world).
In the Flemish speaking areas in the south of the Netherlands the Siegenbeek spelling was always unpopular. After Belgium declared independence in 1830 the spelling was denounced as “Hollandish” and “Protestant”. The spelling situation was quite chaotic with much discussion about whether to use a or ae, oo or oó, ee or eé, ei or ey, ui or uy, ambt or ampt, u or ue, and about the spelling of verbs.
In 1836 the Belgian government offered a reward for a proposal for a new spelling. In the end the jury, headed by Jan Frans Willems, produced their own suggestion in 1839 which remained quite close to the Siegenbeek spelling in use in the Netherlands. They retained their own spelling of a few words such as kaes (cheese), ryden (to ride) and vuerig (fiery). The Willems spelling was given royal approval on January 9 1844.
The spelling used today both in the Netherlands and in Flanders (Dutch-speaking Belgium) is based on an orthography originally intended only for use in a dictionary. An ambitious project was proposed in 1851 at the Taal- en Letterkundig Congres (Linguistic and Literary Congress) in Brussels at which both the Netherlands and Flanders were represented. The project aimed to produce a large dictionary: Woordenboek der Nederlandsche Taal (WNT) (Dictionary of the Dutch Language), incorporating vocabulary of the past centuries.
There was a problem with this project: which spelling was to be used for the dictionary? There were three spelling systems in use at the time: the Willems spelling in Belgium, the Siegenbeek spelling in the Netherlands, and several variants of Bilderdijk’s system. It would have been unacceptable to have used a mixture of these systems. In addition, the Siegenbeek system did not address certain issues such as when compounds were to be written as one word or the interpolation of letters in between. It was proposed to create a special dictionary spelling.
This spelling was established by the linguists Matthias de Vries and L.A. te Winkel. In 1863 Te Winkel published the results in De grondbeginselen der Nederlandsche spelling. Ontwerp der spelling voor het aanstaande Nederlandsch Woordenboek (The foundations of Dutch spelling. Project for the spelling of the forthcoming Dutch Dictionary). The spelling of De Vries and Te Winkel combined elements of the three current systems, providing a much needed solution to the chaos. By 21 November royal decree in Belgium had accepted the decision. In 1866 De Vries and Te Winkel produced Woordenlijst voor de spelling der Nederlandsche taal (Vocabulary for the spelling of the Dutch language), which can be seen as a precursor of today’s Groene Boekje (Green Booklet).
The Netherlands were slower in accepting the De Vries and Te Winkel spelling. Schools continued to use the Siegenbeek spelling until 1870, and in government circles it was not until 1883 that the new spelling started to be used. The De Vries and Te Winkel spelling eventually led to a large degree of uniformity of spelling in the Netherlands and Belgium.
Teachers and linguists continued to object to certain features of the spelling. It was thought that too great an emphasis was being given to etymology. Why lezen (single vowel) but heeten (double vowel)? There were etymological reasons for this, but it was hard to teach as it did not reflect pronunciation. R.A. Kollewijn produced an article in 1891 Onze lastige spelling. Een voorstel tot vereenvoudiging (Our awkward spelling: a proposal for simplification). He emphasized the need for spelling to relate to pronunciation, therefore mensch (person/human) and Nederlandsch ought to become mens and Nederlands, Russisch (Russian) he thought should be spelt Russies and moeilijk (difficult) moeilik.
In 1916 a Dutch commission looked into the possibility of a compromise between De Vries and Te Winkel and the Kollewijn spelling. This gradually led to adaptations: on 1 september 1934 the minister for Education, Marchant, accepted most of Kollewijn’s proposals. The Netherlands and Belgium were starting to diverge once again.
The Marchant spelling included:
The endings '-isch' (as in logisch (logical)) and '-lijk' (mogelijk (possible)) remained unchanged. Kollewijns proposals '-ies' and '-lik' remain popular in some circles as a "progressive" spelling to the present day.
During World War II the governments of the Netherlands and Flanders decided to look for a way to restore the unification of spelling based on De Vries and Te Winkel. This led to the introduction of a simplification of the Marchant spelling being introduced in Flanders in 1946 and in the Netherlands the following year. A Flemish-Dutch committee compiled a vocabulary which was published in 1954 in a green volume entitled “Woordenlijst van de Nederlandse taal” (Vocabulary of the Dutch language), which became known as “het Groene Boekje” (the Green Booklet).
There was still dissatisfaction after 1954. Uncertainty arose about many words which had alternative spellings: one version was the preferred spelling (e.g. actie (action)), the other was the permissible or progressive spelling (e.g. aktie). The Dutch generally used the former, the Belgians the latter. Another problem was the speed at which Dutch was developing new vocabulary for which the 1954 dictionary was of no help for spelling definition.
In 1994, after much discussion, the Committee of Ministers of the Nederlandse Taalunie which had been founded in 1980 issued a new spelling decree. In the new Groene Boekje which they published the alternative “progressive” spellings were abolished (it was now actie) and there were new rules about the n linking the compounds of words (pannekoek (pancake) became pannenkoek and bessesap (currant juice) became bessensap). This came into force in 1996.
In 1994 it was agreed that the vocabulary of het Groene Boekje should be reviewed every ten years without changing the actual rules. On 15 October 2005 the first of these revisions appeared. Only one rule concerning exceptions was made (the so-called Dandelion Rule): paardebloem (dandelion) and vliegezwam (fly agaric) became paardenbloem and vliegenzwam in order to conform to other similar compounds, e.g. paardenstaart (horse-tail) and vliegenmepper (fly swatter). (Notice that these ‘n’s are not normally pronounced.)
Apart from this there are a few individual changes. Here are some of the most important ones:
Although government and educational establishments have to conform, some newspapers and other publications in The Netherlands are refusing to use the new spelling and have released Het Witte Boekje (The White Booklet) as an alternative to the latest edition of Het Groene Boekje. This "white spelling" allows more than one spelling in several cases, like when it comes to linking ns, hyphens and capitalisation. It has, in return, been accused of being even more inconsistent than Het Groene Boekje. In Flanders, the white spelling is not used by any large media.