RANDSTAD DUTCH


Cor van Bree (University Leiden NL)


1. Introduction[1]


What is Randstad Dutch? I use Dutch as a umbrella notion. Dutch includes Standard Dutch and the
dialects, in the widest sense, that are spoken in the domain of this language and/or are closely
related to it.[2] Randstad is the name given to a conurbation in the west of the Netherlands. This
conurbation consists of cities and towns that have been growing together, so that they now form as
it were a shape of a frame which encloses an agricultural region with some scattered towns and
villages, the so called Green Heart. See map 1. Among these towns are Alphen aan den Rijn, not far
from Leyde, and Zoetermeer, a sattelite town of The Hague. The frame runs from Haarlem southwards
via Leyde and The Hague to Rotterdam and from this city south-east to the socalled Drechtsteden
(Dordrecht, Zwijndrecht, Sliedrecht etc.) and north-east to Utrecht. From Utrecht it runs north to
Amsterdam and then westwards to Haarlem, our point of departure.

The Randstad including the Green Heart covers large areas of the provinces of South Holland and
North Holland and the western part of the province of Utrecht. It excludes North Holland above the
river het IJ and the socalled South Holland islands and waarden ‘holms, bottoms’ in the south. The
Randstad does not extend beyond the city of Utrecht although we may doubt if we should not to
include the city of Amersfoort, north-east of Utrecht. Outside its borders there are mainly
agricultural regions in the north and the south and a region covered with forests in the east. In
the west the Randstad borders on the North Sea.

In this contribution I shall leave the eastern branches, between Amsterdam and Utrecht and between
Rotterdam and Utrecht, out of consideration, because these branches have their own dialectal
backgrounds and need a separate treatment.

(For the sake of clarity: I do not use the term Holland for the entire Netherlands but only for the
western provinces North and South Holland. Holland became also the name for the whole country
because it was its the most important part. Utrecht is the name of a city as well a province of
which this city is the capital.)


2. The old Holland dialect


Originally Holland dialects were spoken in the region described above. Relicts are present
especially in two seaside resorts, Scheveningen (now a borough of the Hague) and Katwijk aan Zee
(not far from Leyde). Both are well-known old fisher villages. I give a few examples from the
dialect of Katwijk: the bilabial w, a palatal pronunciation of the aÛ (both can be heard in the
word for 'water': [ÃÏ:t«r], and the consonant cluster sk- instead of sch- [sX-]: skool, Standard
Dutch school ‘school’ In the coastal regions in general we can find relicts of an old dialect
spoken there about 800 and closely related to early stages of English and Frisian. The terms
Ingveonic, Coastal Germanic or North Sea Germanic are used for this relative [betrekkelijke]
language unity within the West-Germanic dialect continuum. Bregge for 'bridge' is an example. It
contains an [E] in the northern coastal regions, e.g. in Katwijk, and an [I] in the southern
regions: brigge, e.g. in the province of Zeeland. The two variants contain unrounded vowels in
comparison with Standard Dutch brug with [{]. Frisian has a long [E]: brêge, and English [I}:
bridge. Compare also the toponyms Terbregge lit. ‘near the bridge’, the name of a village near
Rotterdam, and Brigdamme, lit. ‘bridge-dam’, on the island of Walcheren in the province of Zeeland.

The old Holland dialect has disappeared to a great extent. However, that does not mean that
Standard Dutch is now spoken everywhere and by every groups and in all situations. To explain this
we have to see how Standard Dutch came into being.


3. The rise of Standard Dutch


The rise of Standard Dutch started in the 16th century. In that century Antwerp was the most
important city in the Netherlands. It looked as if a standard language would come into being on the
basis of the Brabant dialect of this city. The year 1585 marks an important turn of fate in the
history of the Netherlands. At the time the Netherlands were at war with Spain and Parma, the
leader of the Spanish army, besieged and captured the city. This victory meant the definitive
separation of the northern and the southern Netherlands.[3] The northern Netherlands became a free
republic, the Republic of the Seven United Netherlands, the origin of the present Kingdom of the
Netherlands (Nederland). The southern provinces remained under Spanish rule and later on they fell
to Austria and, at the end of the 18^th century, to France; they became the historical basis of
present-day Belgium (België).

Many southern emigrants settled in the northern provinces for religious reasons: they were
protestants and would not live in the catholic south. Among them there were rich merchants who
brought prosperity into the northern Netherlands. The political, economic and cultural centre
shifted from the south to the north; in terms of cities: from Antwerp to Amsterdam. As a result
Standard Dutch was formed on the basis of the Holland dialect, especially the dialect of the upper
class in the big cities. Therefore Standard Dutch can be called Regentenhollands in origin, lit.
‘the Holland dialect of the regents (i.e. the governing aristocrats)’ (Van Haeringen 1936, 308).
The poet Joost van den Vondel, in his Aenleidinge ter Nederduitsche Dichtkunste ‘Introduction to
Dutch Poetry’ (1659), especially mentioned Amsterdam and the Hague in this connection. According to
him Standard Dutch in statu nascendi was the language of the rich merchants of Amsterdam and the
Court of the “stadthouder”, the Prince of Orange, in the Hague. Of course the grammarians played an
important role in establishing the norms of the standard language. They conformed to the language
use of the higher classes, but these classes in their turn conformed to the norms established by
the grammarians in their turn.


4. Written and spoken standard language


At first a written form of the standard language came into being that was strongly influenced by a
written tradition of southern origin. Dutch was  written for the first time about 1200, in Flanders
(Bruges, Gand) and later on in the Brabant region (Antwerp, Malines, Brussels) too. This led to a
written standard on a southern basis. The Holland authors conformed to this standard to a great
extent, but gradually they used more and more Holland elements. An oral form of the standard needed
more time to come into being; it was strongly influenced by the written standard. The form breg(ge)
e.g. was not in the written language and therefore it was not allowed in the spoken standard
ultimately. The form brug became the standard form in both written and oral Standard Dutch. In
other cases the Holland form won out at least in the spoken language, e.g. gooien 'to throw'
instead of southern werpen. Another example is the pronunciation of final n after schwa. This n is
still written but mostly it is not pronounced (except in some dialect regions): dage(n) 'days',
regene(n) 'to rain’. Sometimes the old Holland elements also won out in the written language;
compare the diminutive endings containing -(t)je instead of southern -ke: e.g. bakje, diminutive of
bak ‘tray’, instead of bakske. However, the newer Holland endings containing -ie were not allowed
as in bakkie.

A complicated example is the development of the sound ei [Ei]. Originally the different spellings
<ij> and <ei> expressed different pronunciations, resp. [Ei] and [ai] (older resp. [i.] and [Ei]).
Although the grammarians tried to prevent the merger of these sounds, they did not succeed;
however, they succeeded in maintaining the different spellingss. Moreover, they succeeded in
establishing the “civilised” pronunciation of the phoneme. The “correct” pronunciation is still
[Ei] and not [ai], the second being considered substandard. The grammarians rejected the
pronunciation [ai] because neither the spelling <ei> nor the spelling <ij> suggested a
pronunciation [a](cf. Hellinga 1968, chapter 2). However, this was probably a rationalisation in
hindsight: the secret reason was that [Ei] was the pronunciation of the higher classes and [ai]
that of the lower ones.


5. Grammatical argumentations


In general, the grammarians chose the forms of the higher classes, but defended them with different
kinds of argumentation (cf. Van der Wal 1996). Authority was an important motivation: elements from
the (more authoritative) written language were preferred (compare brug) and pronunciations were
chosen in conformity with (the visible and therefore influential) spelling (compare the preference
for [Ei]. In the opinion of the grammarians, especially the language of famous authors (Joost van
den Vondel, Pieter Corneliszoon Hooft) deserved imitation. Another important motivation was
mellifluence, whether the sound was “beautitul” or “pleasant”: -(t)je sounded better than –ke (cf.
Van der Wal & Van Bree 1994, 211-12) and [Ei] better than [ai] (cf. Hellinga 1968, chapter 2). Of
course, mellifluence is a very strange motivation, because phonemes are not beautiful or ugly as
linguists know. But here we have an interesting paradox: just because they are neutral in se, they
are very suitable to be connected with value judgments.

The most important motivation was avoidance of ambiguity. Ambiguity is one of the greatest concerns
of standardising grammarians. Compare again the maintenance of the difference in spelling between
<ei> and <ij> (cf. Hellinga again). A famous lexical example is the distinction between liggen 'to
lie' en leggen 'to lay' and the distinction between kennen 'to know' and kunnen 'to be able to'. In
the Holland dialect these distinctions are not present anymore: there are two verbs, legge 'to lie'
and 'to lay', and kenne 'to know' and 'to be able to'.[4] A morphological example is the
distinction between the personal and the possessive pronoun: whereas the standard language has
different forms, e.g. mij/me 'me' and mijn/m'n 'my' (before the slash the strong, behind it the
weak forms containig a schwa), the Holland dialect has only one form: mijn/me (mijn pronounced as
[main] or [mE:n]). The translators of the Statenvertaling, the famous Bible translation of 1637,
noted in the Latin registration of their decisions: numquam mijn, ut vulgus hic loquitur 'never
[use] mijn [instead of mij] as common people here say' (Van der Wal 1995, 35). They clearly
dissociated themselves from the “vulgus”.

Distinguishing oneself from the lower classes played an important role is also proved by the
following case. In Middle Dutch the same form was used for the personal and the reflexive pronoun,
namely hem or haer: hi wast hem/si wast haer 'he washes him/himself, she washes her/herself'. In
the Holland dialect this ambiguity was solved by a new form for the reflexive, namely zijn eige
(pronunciation: z«n eige) or haar eige (pronunciation: d«r eige). The grammarians chose another
solution, namely a form of High German origin: zich, which had penetrated into the written language
via Bible translations under German influence (cf. Van der Wal & Van Bree 1994, 214-15). The
grammarians solved the ambiguity but they disqualified the form of the vernacular at the same time.
An interesting example of interplay of a linguistic and a sociolinguistic factor.


6. Standard syntax


So far I have given examples from four different levels: phonology, orthography, lexicon and
morphology. Are there also syntactic examples? There are syntactic phenomena that are accepted in
spoken but not in written Standard Dutch, e.g. the so called left dislocation as in die man DIE zei
tegen me dat ik weg moest gaan lit. 'this man HE said to me that I had to go'. That this
construction is not accepted in written language may be due to the fact that it is experienced as
redundant. There are also phenomena that are rejected both in spoken and in written Standard Dutch,
e.g. the use of a double negation as in ik heb NOOIT GEEN brief van hem gekregen, lit. 'I have
NEVER NO letter from him received' = 'I never received a letter from him'. The double negation is
rejected because it is felt to be illogical, because mathematically - x - = +![5] Moreover, there
might be influence from classical Latin involved. Both constructions occur on a large scale in the
Dutch linguistic area.

There are not many syntactic phenomena that are characteristic of the Holland dialect or parts of
it. An example is conditional al in e.g. AL is het morgen mooi weer, dan ga ik naar het strand lit.
'IF it is beautiful weather, I shall go to the beach' (cf. Van Bree & Van der Hee 2002-3 and 2002).
In Standard Dutch the conjunction al can only be used as a concessive conjunction. A speaker of
Standard Dutch would be inclined to interpret the sentence in question as 'although it is etc.'
which would be a strange communication.

Although many phenomena known from the Holland dialect (and from many other dialects) are not
accepted in Standard Dutch, at least in written Standard Dutch, the standard syntax has
nevertheless a Holland origin. Perhaps southern phenomena spread to the north during the Middle
Ages, e.g. the use of partitive (referring) er as in heb je schone overhemden? ja, ik heb ER een
paar lit. 'do you have clean shirts? yes, I have (OF THEM) a couple'. However, when standardisation
started, these phenomena had become indigenous Holland dialect. Later southern developments did not
reach the Holland region anymore, e.g. the use of er referring to a singular: heb je koffie? ja, ik
heb ER nog lit. 'do you have coffee? yes, I have still (OF IT)'. On the other hand, the south did
not participate in later northern developments such as the replacement of the causal auxiliary doen
by laten. Causal doen is still in use in the south but in the north it is restricted to rather
formal written language: ik DEED hem vertrekken instead of ik LIET hem vertrekken 'I had leave
him'. (Cf. for syntax Van Bree 2007.)


7. Selection and acceptation


The American sociolinguist Einar Haugen (cf. Haugen 1966) distinguishes four stages (we can also
speak of aspects) of standardisation: selection, codification, elaboration of function, and
acceptation. In this context we are interested in selection and acceptation. As regards election,
it is useful to distinguish macroselection and microselection. Macroselection concerns the question
which dialect was chosen to be the standard language. Microselection concerns the question which
concrete forms were chosen and which rejected. We have seen that the elements selected mostly have
a Holland origin but among them there are also southern (and High German) elements.

As I remarked, acceptation of the written language was in the lead. In an 1825 story, an Amsterdam
grandmother of a rather high social level tells her grandchildren a fairy tale, in which she still
uses a language that contains many features of the old Holland dialect (Van der Wal & Van Bree
1994, 285-86). An explanation may be that in the higher classes French functioned as standard
language (Burke 2005, 20). But gradually Standard Dutch became the daily spoken language of the
higher social classes in the big cities and spread from the higher to the lower classes. Later on
the same happened in the countryside. There arose a contrast between the classes: more standard
language at the higher social levels than at the lower ones. In addition, a similar contrast arose
between the cities and the countryside. Such contrasts had not exist to such an extent before.


8. The new Holland dialect


In many respects the lower classes preserved the old Holland dialect forms. All examples given
above can still be heard. This also holds for the phonological ones: the [ai]-pronunciation of
<ei>/<ij> is still very frequent just like the ou [Ou] in nou 'now' and douwe 'to push' instead of
Standard Dutch nu and duwen (both with [y]). However, at the phonological level new developments
took place. The lower classes introduced new pronunciations, different from the old Holland dialect
as well as from the standard language. This development should be considered in connection with the
radical social changes of the era of industrialisation. The old hierarchy of standen (social
positions) developed into a new hierachy of socio-economic classes. In the old hierarchy everyone
had her or his own position authorised by God. The members of the different groups respected each
other, shared the same norms and, to a great extent, the same language, i.e. the same dialect. In
the new hierarchy a social gap existed between employees and employers. On the other hand, there
was a growing unity, caused by improved instruction, increase of real and social mobility and
cooperation on the shop floor. Contrast and unity were reflected in the language, the unity at the
phonological and the contrast at the phonetic level. Both classes shared the same phonemes but did
not pronounce them in the same way. Compare Commandeur 1988-9 and cf. also Van der Wal & Van Bree
1994, chapter 12.

Let me illustrate this with an example. The old Holland form veul 'much, many' (with eu [¼.])
disappeared (it was still used in the countryside for a long time) and was replaced by veel (with
ee [e.]), a form of southern (more precisely: south-western) origin and already present in written
standard language. The [e.] seems to be a rather unstable sound: it easily moves to a higher
position (>[i]) or is diphthongised (>[e.’]). Cf. Labov 1994, 116 and Van Bree 1996, 98. The second
development took place in western Dutch: ee is pronounced slightly diphthongised (whereas the
eastern ee is a pure monophthong like in German). In the vernacular of the Randstad the result is a
clear diphthong, approaching ei [Ei]. The same holds for the two other mid high long vowels, the
rounded front vowel eu and the back vowel oo. These vowels are approaching ui [¿i] and ou [Ou]
respectively. The diphthongisation began in the cities. An old attestation (perhaps the oldest one)
is from Leyde and dates from 1841, In Studenten-typen, a sketch by Klikspaan (pseudonym of Johannes
Kneppelhout): Heirejei instead of Herejee lit. ‘Lord Je(sus)’ (a disguised oath) and laiven
(containing an remarkably extreme diphthong!) instead of leven ‘life’. The diphthongisation was so
successful in this city that it also took place before r (before which, in general,
diphthongisation is excluded in Standard Duch): compare Heir(ejei). From the cities the Randstad
diphthongisation expanded to the more agricultural regions. I have the impression that it is
stronger in the north, especially around Amsterdam, than in the south. However, the first
attestation is from the middle region (Leyde) as we have seen.

Summarizing, we have to take into account three variants: the old Holland variant, now
disappearing: veul [¼.], the standard variant: veel [e.’], and the new Holland variant, approaching
veil [Ei].


9. The Superstandard


However, the picture is still more complicated. We have to take into account a fourth variant, a
superstandard variant. The highest classes, old nobility and also intellectuals, dissociate
themselves from the lower classes (and perhaps also from the middle classes) in a hypercorrect way.
This way of speaking is called geaffecteerd lit. 'affected' or bekakt ‘snooty’. It is comparable
with posh English in England.[6] People who speak this kind of Dutch are so afraid of speaking
diphthongised ee, eu and oo that they avoid the slightest trace of it. Therefore, they pronounce
veel, neus and groot with pure monophthongs: [e.], [¼.] and [o.]. (Such tendencies to distinction
are a well-known phenomenon and were already discussed in Van Haeringen 1962.)[7]

Substandard Holland Dutch is considered not to be appropriate for formal situations. If we carry
out a Labovian research in which we try to elicit different speech styles (casual style and more
formal styles), we see that speakers of the new Holland dialect try to avoid their socially
stigmatised features in more formal style. In a formal style the speaker is more aware of how
he/she has to express him/herself. Because affected Dutch is spoken by very elitist people, we
expects its use to increase in more formal styles. This expectation did not prove to be correct in
research carried out at Leyde University: in a more formal style bekakt-speakers produced fewer
superstandard features than in a more informal style. They try to speak more “normally”, probably
because they are aware of the fact that, since democratisation started about 1970, their way of
speaking is not very much appreciated anymore. Of course, it is characterised by culturedness, but
it lacks generalness, and according to Smakman (2006, 277-278) there seems to be the need in the
Netherlands not to stand out. In a more informal style bekakt-speakers speak as if they were “among
us”.[8] Cf. Van Bree & Van Dunné 2004.

Affected Dutch is often named Haags or Leids. Perhaps The Hague (where many members of the old
nobility lived) and Leyde (a rather elitist university city) are the places of origin. However, the
names are confusing because affected Dutch can now be heard in the whole country.


10. The Randstad continuum


The developments described above have led to a continuum, in which there are no clear boundaries
between the “varieties”. In the more peripheral parts of the Dutch linguistic area (in Flanders as
well as in the northern, eastern and southern regions of the Netherlands) the situation is
different: there is still a discontinuum with a clear difference between dialect (or regiolect) on
the one hand and standard language (or a regional variety of the standard language) on the other
hand. Because the old Holland dialect was the most important source of Standard Dutch, there
existed a continuum in the west from the beginning.

The Randstad continuum reaches from extreme Substandard on the one hand to extreme Superstandard on
the other hand. People living in the Randstad are aware of only one language (Dutch) although they
may also be aware of striking differences. Persons who mostly speak substandard Holland Dutch do
not think the queen (who speaks Superstandard) to speak a different language, and neither does the
queen. Above I treated a number of tendencies, tendencies to diphthongise ee, eu and oo (in
Substandard), to monophthongise them (in Superstandard) or to diphthongise them within certain
limits (in neutral standard language), etc. The intensity of these tendencies depends on style,
gender and social position (the last-mentioned now more a question of education than of income and
housing). However, we should not interpret this causal relation too simply: on the one hand
people’s way of speaking is determined by situation and class or group, but on the other hand
speakers can use their way of speaking to indicate how they interpret the situation and to which
class or group they want to appear to belong.

In any case, the linguistic situation of the Randstad is a dynamic one, more precisely: a stable
dynamic situation. The “varieties” that are involved show remarkable stability. Especially
remarkable is the stability of the Substandard: although those who speak it are aware of the fact
that their way of speaking is not appreciated very much by higher classes, they nevertheless stick
to it, because it expresses their identity and their mutual loyality (Labov: covert prestige). The
superstandard seems to be stable too although its “market value” has decreased (cf. section 12).

This stable dynamics, however, does not exclude the possibility of some elements undergoing
changes. Even the standard language is not a fortress that cannot be captured. Let us take as an
example the substandard pronunciation of ei <ei, ij>: [ai]. This pronunciation (and other
substandard pronunciations) can be heard ever more in the speech of young educated people, but we
have to wait and see if this will become the new norm (cf. Stroop 1997)[9] It is also possible
that, as a result of the modern processes of democratisation and informalisation, the norm will
allow more variation than formerly: the norm has become more diverse or more variants are being
tolerated although they do not belong to the norm proper.

In any case, the Substandard can be an important source of innovations in Standard Dutch. A
morphological example is the past tense of raden ‘to guess, to advise’: the old strong form ried
has become very old-fashioned whereas the new weak form raadde is now the norm. Another important
source of innovation is slang: a word such as lullig ‘shitty’ (a derivation from lul ‘prick’) is
more and more acceptable.[10]


11. Review of the vowels


Below I shall give a review first of the short and then of the long vowels including the
diphthongs.[11] For each group of vowels there is a diagram with comments. I shall give fixed
values of the (neutral) standard phonemes, but, of course, these can show some variation and
tendencies to change (cf. section 10). These fixed values represent the norms and serve as a frame
of reference. The arrows symbolise tendencies: they are directed at the (covert) norms of
Substandard and Superstandard. They can also be read in the opposite direction, symbolising
tendencies which manifest themselves in a more formal style. Together the values and arrows
symbolise the dynamics of the Randstad situation.[12]

For the Substandard cf. Commandeur 1988-9 and, also for the old Holland sounds, Van Bree 2004, for
the Superstandard Van Bree & Van Dunné 2004 and for Standard Dutch Smakman 2006 (and the literature
mentioned in these publications).


11.1. The short vowels


There seems to be a general and natural tendency for short vowels to become more open (lower)
(Labov 1994, 116). However, in substandard Holland Dutch there is the opposite tendency, and the
(more natural) tendency of fronting, both conditioned by following consonants. In Amsterdam and
surroundings a* and o* are palatalised before tensed n, t, d and st: pan > [pa’n] ‘pan’, kat >
[ka’t] ‘cat’, (we) hadde(n) > [ha’d«] ‘we had’, kast > [ka’st]; bon > [bo’n] ‘voucher’, pot >
[po’t] ‘pot’, vodde(n) > [fo’d«] ‘rags’, post > [po’st] ‘post’. Cf. map 2 for the spread of this
phenomenon south of Amsterdam. Something similar holds for e*, which approaches [I], and maybe for
u* [{], especially before the same consonants. More to the south, in Rotterdam and surroundings,
the a* becomes more velar (> [O]) before labial or velar. Rotterdam is more or less pronounced as
Rotterdom; zag je dat? 'did you see that?' sounds as zoggie dot? [zO¯i dOt?]. In the same region u*
and o* can be pronounced higher and dinke [I] instead of denke)n) ‘to think’ can be heard, with [I]
instead of [E] before n plus consonant.

The superstandard reaction is to pronounce the vowels in question more open. Especially the more
open pronunciation of e*, u* [{] and o* is very striking. A well-known sentence mocking The Hague
pronunciaton is: in Den Hèèg gèèt men tannissen met kannissen instead of neutral Standard Dutch in
Den Haag gaat men tennissen met kennissen 'in the Hague people go playing tennis with
acquaintances' (for the èè in The Hèèg and gèèt see next section).


11.2. The long vowels and diphthongs


Properly speaking, the three diphthongs do not fit in the diagram phonetically. For the sake of
convenience they have been placed in the centre of the diagram. About the high vowels nothing
important can be noticed: they are very stable and do not show much variation. An exception is
substandard ou in nou and douwe instead of nu and duwen (see section 8). The mid high vowels were
discussed in section 8. In addition I mention palatalisation of oo before r in Superstandard: voor
‘for, before’ [vo:r] sounds as veur [v¼:r], and the superstandard pronunciation of e.g. militair
‘military’ with [I:] instead of [E:] for <ai>.

The diphthongs show two ways of substandard pronunciation. In the first place there is the old
Holland way of pronouncing them as rather extreme diphthongs, with great phonetic distances between
their first and second parts: ei <ij, ei> > [ai], ui [¿i] > [Ai], ou <ou, au> > [au]. In the second
place there is the (probably more recent) way of pronouncing them as socalled secondary
monophthongs: ei > [E:], ui > [¿:], ou > [A:]. Secondary monophthongisation is characteristic of
the cities of Amsterdam (which has [a:] besides [E:]), Harlem, Leyde, The Hague, Delft and
Dordrecht. However, only the Hague shows monophthongisation of ou. Maybe the secondary monophthongs
were originally (unsuccessful) attempts to pronounce neutral standard ei, ui and ou. An old
attestation (perhaps the oldest) is from the south-east quarter of old Amsterdam, from 1874: vaaf,
with [a:], instead of vijf, with [Ei]. From the same year is the attestation of the Hague [¿:] and
[E:] in (wat) reuist (die) zzèè, instead of (wat) ruist (die) zij! lit. ‘how rustles this silk!’.
(Cf. Winkler 1874, p. 90 and 130 respectively.) The countryside mostly has the old Holland
pronunciation. It is remarkable that this is also the pronunciation of the big city of Rotterdam.
This may be due to the fact that many immigrants from the surroundings came to this city in the
second half of the 19th century. There are, however, indications (cf. De Reus 1991) that the
extreme diphthongs are again finding their way into the other big cities. The superstandard
reaction is to pronounce ei, ui and ou as little diphthongised as possible, in the direction of ee,
eu and oo. Of course, the directions [E.], [¿.] and [A.] are excluded because these would entail
pronunciations approaching the results of the stigmatised secondary monophthongisation.

A rather unstable vowel is the low and medial aÛ. In South Holland the substandard pronunciation is
[a.] like in the standard language, the so called clear aÛ (not too velar and not too palatal); in
the North Holland part of the Randstad, especially Amsterdam, the pronunciation is [A.], a so
called dark, rather velar aÛ. Sometimes, a nasalised [a.] can be heard, especially in the speech of
women (cf. for Amsterdam Schatz 1986). This [aà.] sounds not very standard. Maybe the nasalisation
makes it easier to keep the aÛ central where it properly belongs and to avoid deviations to the
palatal or the velar side. The superstandard pronunciation is slightly palatal: [Ï.], e.g. in Hèègs
(cf. section 11.1). It is difficult to explain this. It might be a reaction to the dark aÛ, but the
palatal pronunciation is not restricted to the northern part of the Randstad region. Still more
peculiar is the fact that especially the Hague is known for the palatal aÛ whereas the old dialect
of Scheveningen (now part of the Hague) has a similar aÛ! Perhaps a dark aÛ, [A.] or [O.], is
considered to be rustic (Dutch: boers), because it is the pronunciation of many Dutch agricultural
regions. The palatal aÛ of the Scheveningen fishermen is (was) not open to objection, perhaps
because they are or were a rather closed community with few outside contacts. Moreover, their “aÛ”
has not quite the same pronunciation as the superstandard aÛ: [e.] and [Ï.] respectively.


11.3. The consonants


As regards the consonants I restrict myself to a few remarks.


1. The r has many different pronunciations in Standard Dutch. At least three should be mentioned:
the alveolar, the uvular and the so called Gooise r. The third r is called so because people think
that it has been spread via radio and television programmes which are broadcasted from studios in
het Gooi, a region southeast of Amsterdam. Its precise nature is not yet clear: perhaps it is a
retroflex r. It is restricted to the codas of syllables. It is a very striking sound, which is
especially heard in the speech of younger, well educated people. Recent research (cf. Smakman 2006,
235-243) shows that both the alveolar and the uvular r are accepted in Standard Dutch. The
acceptance of the Gooise r, in any case a clear Gooise r, is dubious, because there is a preference
for a non-prominent r in the coda.

In substandard Dutch there is also a lot of variation. The dialect of the Hague is characterised by
an uvular r approaching [X]. The difference with [X] <g, ch> is maintained by pronouncing the
latter extra fortis (Van Gaalen & Van den Mosselaar 1998). The dialects of Leyde and Rotterdam have
an r that is related to the Gooise r, except that it can also be heard at the beginning of a word.
In Superstandard there is a preference for a moderate uvular r.

For the pronunciation of the r cf. Bezooijen 2006, for especially the Gooise r for instance
Bezooijen 2004.


2. The l has a rather velar (“thick”), [u]-like pronunciation in substandard Holland Dutch,
especially in the coda. This pronunciation is more striking in the north than in the south. The
result is e.g. that it may be difficult to hear whether vernield ‘destroyed’ or vernieuwd
‘renovated’ is meant. The superstandard reaction is to pronounce a “thin” l like in German, with
quite a lot of tension.


3. The v and z (voiced or lax fricatives in Standard Dutch) are pronounced voiceless or tense in
substandard Holland Dutch. The g, a more uvular than velar fricative, is often tense in Standard
Dutch too and can be rasped in both varieties. However, to much rasp sounds “ugly”. The voiceless
or tense pronunciation is more striking in the north than in the south. The north may be the region
from which it spread; in any case, Winkler (1874, I, 78) already speaks of an expectorating g, for
which he especially blames the inhabitants of Harlem. In research done with students, Harlem had
41.7% devoicing, Rotterdam 39,9% and Vlaardingen, west of Rotterdam, 26%. In other research
Uithoorn (just north of the boundary between North and South Holland) had 53% devoicing in
opposition to Ter Aar and Nieuw-Vennep (just south of it) with only 14%. On average, the g is more
voiceless or tense than v and this fricative, more than z (Smakman 2006, 250). As regards place of
articulation, especially Amsterdam is known for a rather palatal s: [s#]. It is possible that this
s originates from the Dutch as it was spoken by Jewish people (Den Besten & Hinskens 2005, 293).
The superstandard reaction can be the slis-r ‘lisped r’ almost like [¹] (the english hard
th-sound). The so called soft g (voiced and lax) is characteristic of Flanders and the southern
regions of the Netherlands. Recent research shows that is not felt problematic in Standard Dutch
(Smakman 2006, 227).[13]


12. Conclusion


The picture I sketched above, has been based partly on research and partly on impressions. The
impressions are the author’s, who was born in the Randstad (in Vlaardingen, west of Rotterdam) and
has always lived there (in Oegstgeest, near Leiden, from his 19^th year). He usually speaks neutral
Standard Dutch but, in a sense, Substandard Holland Dutch and Superstandard Dutch belong to his
registers. The research was carried out by him and by his students (University Leiden), all
speakers of Standard Dutch from the Randstad, by carefully listening to informants. An additional
phonetic analysis might be interesting, but that I have to leave to the phoneticians. (Cf. for
Standard Dutch Smakman 2006).


BIBLIOGRAPHY


Bartsch 1987: Renate Bartsch, Norms of language, London & New York.

Den Besten & Hinskens 2005: Hans den Besten & Frans Hinskens, ‘Diversificatie van het Nederlands
door taalcontact’, in: Nederlandse Taalkunde 10, 283-309.

Bezooijen 2004: Renée van Bezooijen, ‘De Gooise r: wie ziet er wat in en waarom?’, in: Taal en
Tongval. Themanummer 17 Taalvariatie en groepsidentiteit onder red. van Arjan van Leuvensteijn,
Roeland van Hout & Henk Aertsen), 2004, 86-108.

Bezooijen 2006: Renée van Bezooijen, Het raadsel van de ratelaar. Rondom de r in het Leewarders en
het Fries, Ljouwert/Leeuwarden.

Van Bree 1975: C[or] van Bree, ‘Regionale varianten van de Sjwa’, in: M.J.M. de Haan & B. van Selm,
Studies voor Zaalberg enz, Leiden, 53-72.

Van Bree 1996: Cor van Bree, Historische Taalkunde, 2d impression, Leuven & Amersfoort (also to
consult digitally: www.dbnl.nl).

Van Bree 2004: Cor van Bree, Zuid-Hollands, The Hague.

Van Bree 2007: Cor van Bree, ‘Syntactische configuraties in het nederlandse taalgebied’,
forthcoming in: Neerlandica Wratislaviensia.

Van Bree & Van Dunné 2004: Cor van Bree & Essemie van Dunné, ‘Variatie in geaffecteerd Nederlands’,
in: Johan de Caluwé etc. (eds.), Taeldeman, man van de taal, schatbewaarder van de taal, Gand.

Van Bree & Van der Hee 2002-3 and 2003: Cor van Bree & Jeroen van der Hee, ‘Conditioneel al’, in:
Taal en Tongval themanummer 15-16, 158-183 and 55, 124-137.

Burke 2005: Peter Burke, Towards a Social History of Early Modern Dutch, Amsterdam.

Commandeur 1988-9: P.N.M. Commandeur, ‘Het ontstaan van het modern Amsterdams. Over het verband
tussen maatschappelijke ontwikkeling en taalontwikkeling’, in: Taal en Tongval 41, 159-171 and 42,
1-21.

Van Haeringen 1936: C.B. van Haeringen, ‘Hollands’, in: De Nieuwe Taalgids 30, 302-309.

Van Gaalen & Van den Mosselaar, “Kèk mè nâh”...”. Plat & bekakt Haags, Den Haag.

Van Haeringen 1960: C.B. van Haeringen, Netherlandic Language Research, Leiden,

Van Haeringen 1962: C.B. van Haeringen, ‘Gedistingeerdheid in taal’, in: C.B. van Haeringen,
Gramarie, 1962, 24-37, first published in: De Nieuwe Taalgids 42, x-x.

Hagen  1990: A.M. Hagen, ‘Groepsportret van het Nederlands’, in: Onze Taal 2/3, 14-44.

Haugen 1966: Einar Haugen, ‘Dialect, language, nation’, in: American Anthropologist 68, 922-935.

Hellinga 1968: W. Hellinga, De opbouw van de Algemeen Beschaafde Uitspraak van het Nederlands, in:
W. Hellinga, Bijdragen tot de geschiedenis van de nederlandse taalcultuur, Arnhem 1968; originally
thesis Gemeentelijke Universiteit Amsterdam 1938.

Labov 1994: William Labov, Principles of Linguistic Change, Vol. 1: Internal Factors, Oxford UK &
Cambridge USA.

De Reus 1991: Annelies de Reus, ‘Diftongen in het Delfts dialect’, in: Taal en Tongval 43, 137-158.

Schatz 1986: H.F. Schatz, Plat Amsterdams in its social context. A sociolinguistic study of the
dialect of Amsterdam, Amsterdam.

Smakman 2006: Dick Smakman, Standard Dutch in the Netherlands. A Sociolinguistic and Phonetic
Description, thesis Radboud University Nijmegen.

Stroop 1998: Jan Stroop, Poldernederlands – waardoor het ABN verdwijnt, Amsterdam.

Van der Velde 1996: Hans van de Velde, Variatie en verandering in het gesproken
Standaard-Nederlands (1935-1993), thesis Katholieke Universiteit Nijmegen.

Van der Wal 1995: M[arijke] J. van der Wal, De moedertaal centraal. Standaardisatie-aspecten in de
Nederlanden omstreeks 1650, The Hague.

Van der Wal 1996: Marijke van der Wal, ‘De grammaticus en de schoolmeester: externe taalnormen en
taalverandering’, in: Cor van Bree & Ariane van Santen (eds.), Leidse mores. Aspecten van
taalnormering. Leiden, 13-29.

Van der Wal & Van Bree 1994: Marijke [J.] van der Wal & Cor van Bree, Geschiedenis van het
Nederlands, 2d impression, Aula, Utrecht.

Winkler 1874: Johan Winkler, Algemeen Nederduitsch en Friesch Dialecticon, The Hague.
________________________________

[1] I thank Guy Tops (Wilrijk, Belgium) for the correction of my English.

    [2] The Frisian dialects in the northern part of the Netherlands do not belong to Dutch because
they are (more) closely related to Standard Frisian although they are spoken in a region in which
Standard Dutch is used too. However, the Flemish dialects in the north-west corner of France belong
to Dutch: although they are spoken in the domain of Standard French, they are closely related tot
Standard Dutch.

[3] I have to leave aside the history of Dutch, especially Standard Dutch, in the south. I only
mention that the South has adopted the northern standard language since the second half of the
19^th century.


[4] The principal parts in the Holland dialect are: legge – lag – gelege; kenne – kon – gekend,
those in Standard Dutch: liggen – lag – gelegen, leggen – legde – gelegd; kennen – kende – gekend,
kunnen – kon – gekund.

[5] In many languages, e.g. the Slavonic languages, double (or triple etc.) negations are
acceptable in the standard variety. If the negation is repeated an odd number of times (compare
Czech nikdy nikde nikomu nic nekupuje lit. ‘I never nowhere for no one nothing not-buy’ = I never
buy anything for anyone anywhere”), the argumentation that the outcome is illogical does not apply,
because it is negative.

[6] Kloeke (1951) suggests that superstandard Dutch can be traced back to the Regentenhollands
(i.e. of the 17^th and 18^th centuries; cf. section 3) and that neutral Standard Dutch is in origin
the language of people who abandoned their dialect and switched to the standard language in later
periods. In his opinion influence from French on the superstandard pronunciation is possible,
because there was a lot of bilingualism (i.e. in the 17^th, 18^th and 19^th centuries).

[7] The Superstandard is also characterised by a number of lexical preferences and sometimes by the
use of substandard elements in combination with a posh accent, as if its speakers wanted to express
this way that they are so distinguished by birth that they think to be allowed to use these
elements. Perhaps the language of the very highest class (nobility, especially in the countryside)
was in origin fairly substandard. There is a well-know case of a former governor of the province of
Zeeland (Jonkheer ‘Esquire’ De Casembrot) who pronounced clearly diphthongised ee, eu and oo.

[8] That the social value of the Superstandard has decreased also becomes clear when pre-war films
or radio programmes are compared with post-war ones. Whereas the Dutch of the former sounds very
affected, the Dutch of the latter contains quite a few substandard features (for a study of these
differences cf. Van de Velde 1996). What happened to the well-known author Adriaan van Dis is
revealing in this respect: he was not admitted as a television news presentator because of his posh
accent (Hagen 1990).

[9] Cf. e.g. Stroop 1997.

[10] For Standard Dutch a great variety of terms is used. A well-known term is General Civilised
Dutch (Algemeen Beschaafd Nederlands, ABN): ‘Dutch spoken by civilised people and accepted as the
general norm (at least for formal situations) even by people who do not speak it in a proper way’.
In Dutch linguistics there has been a lively but also confusing debate about the suitability of the
terms and the precise nature of their referents. Cf. Van Haeringen 1960, 53-55.

[11] I leave aside the schwa, which can have an [I]-like pronunciation in some Holland dialects as
well as in the Holland substandard. Cf. Van Bree 1975.

[12] I leave aside reductions and assimilations e.g. in [plisi] for politie ‘police’(Standard Dutch
[politsi]) and [kAàs] for kans (Standard Dutch [kAns]), further “svarabhakti” vowels as in [kEr«k],
which vowels can be heard more clearly in Substandard than in Standard Dutch.

[13] I have the impression that the voiced plosives b and d too are pronounced with more tension in
Substandard than in Standard Dutch. They are approaching p and t. Maybe the distinction between b
and d on the one hand and p and t on the other hand is maintained by pronouncing the latter with
still more [nog meer] tension. (There is no voiced velar plosive in Dutch except in cases of
assimilation, e.g. in zaKdoek [zAgduk] ‘handkerchief’.)