Syntactic Atlas of Dutch Dialects – Commentary – Volume I
1 Complementisers
1.1 Introduction
1.1.1 Complementisers
1.1.2 Absence and doubling of complementisers
1.1.3 Complementiser agreement
1.1.4 Historical development
1.2 Discussion of the literature
1.3 Discussion of the maps
1.3.1 Complementisers
1.3.2 Complementiser agreement; general
1.3.3 Complementiser agreement first person singular
1.3.4 Complementiser agreement second person singular
1.3.5 Complementiser agreement third person singular
1.3.6 Complementiser agreement first person plural
1.3.7 Complementiser agreement second person plural
1.3.8 Complementiser agreement third person plural
1.3.9 Plural –n in complementiser agreement
1.3.10 Complementiser agreement –e in plural
1.3.11 Complementiser agreement in comparatives
1.4 Literature on complementisers and complementiser agreement
1 Complementisers
1.1 Introduction
1.1.1 Complementisers
Complementisers belong to the category of conjunctions. Conjunctions are words that conjoin constituents within a sentence. A coordinating conjunction, such as en 'and' and of 'if' puts the constituent preceding it at the same level as the constituent following it. Examples are Willem en Maurits waren goede vrienden 'Bill and Morris were good friends' and Spreekt hij een taal of een dialect 'Does he speak a language or a dialect?' In these examples Willem en Maurits and een taal of een dialect are plural noun phrases, the subject and the object of their clauses respectively. These plural phrases each consist of two equivalent constituents conjoined by en 'and' and of 'if'. Coordinating conjunctions fall outside the scope of this chapter.
This chapter is about subordinating conjunctions, henceforth complementisers. A complementiser also has the function to connect phrases within a sentence, but in this case the constituent following it is subordinated to the constituent preceding it. A complementiser connects two constituents that are not equivalent. A further difference with coordinating conjunctions is that complementisers are used almost exclusively to introduce subordinated clauses. The most typical complementiser is dat 'that', in sentences like Ik weet dat Sinterklaas dit jaar niet komt 'I know that St Nicholas will not come this year.' The clause introduced by dat 'that' is the direct object of the verb weten 'know'. We are dealing with subordination here, also because the possibility to have a dat-clause depends on the verb in the main clause. It is impossible to say *Ik ken dat Sinterklaas dit jaar niet komt, even though weten and kennen are synonyms (* = ungrammatical).
The form of the complementiser depends on the function of the clause that it introduces. In a 'normal' subordinate clause, we find dat 'that' (cf. map 14a), but when the subordinate clause is an indirect yes-no question we use the complementiser of 'if/whether' (notice that of is both a subordinating and a coordinating complementiser), as in Willem vraagt of zijn dialect wel goed Nederlands is 'Bill is asking if his dialect is good Dutch’. Subordinating of 'if' can also be used in a comparison, as in Het lijkt wel of er een inbreker in de tuin staat 'It looks as if there is a burglar in the garden' (cf. map 14b). There are many complementisers that have an adverbial meaning. A conditional clause can be introduced by als 'if': Als de baas het goed vindt, spreek ik vandaag dialect 'If the boss agrees I will speak dialect today' (cf. map 15a). A temporal clause can be introduced by complementisers such as toen 'then', terwijl 'while', nadat 'after' or nu 'now', in sentences like Toen/Terwijl hij op vakantie was, is zijn dialect afgeschaft 'When/While he was on vacation his dialect was abolished’. Other complementisers such as omdat 'because', zodat 'so as to', opdat 'in order that' and mits 'provided' denote various relationships of meaning between the embedded clause and the main clause.
Infinitival clauses can be introduced by a complementiser too. In the unmarked case, the complementiser om 'for' is used, as in Sinterklaas beloofde mij om dit jaar in Spanje te blijven 'St Nicholas promised me to stay in Spain this year’. The sole function of the complementiser om 'for' is here to connect the two sentences; it does not contribute to the semantics of the clause (cf. map 18b). This is different in infinitival clauses introduced by complementisers such as na 'after', alvorens 'before' or zonder 'without'. In the following sentence, the complementiser na ‘after’ has a temporal meaning: Na zijn dialect afgeleerd te hebben, sprak Willem uitsluitend nog goed Nederlands 'After he had unlearned his dialect, Bill only spoke good Dutch'.
1.1.2 Absence and doubling of complementisers
As discussed above, a complementiser has various functions. Firstly, it indicates that an embedded clause will follow. Secondly, it connects the main clause and the embedded clause, and thirdly it determines the kind of connection between these two. The latter function applies in particular when the complementiser has a specific lexical meaning (temporal, causal, conditional, etc.). Given this meaning, we expect the presence of such complementisers to be obligatory. This is different for the complementisers dat 'that' en om 'for', that have no or hardly any meaning. It is therefore not surprising that these complementisers can be left out in certain cases. This is particularly clear for the complementiser om 'for'. When om 'for' introduces a purpose clause, it is obligatorily present, as in Willem ging naar huis om Fries te leren 'Bill went home to learn Frisian' (cf. map 18a). When om 'for' does not have a clear meaning, it can be left out, as in Willem besloot (om) Fries te leren 'Bill decided to learn Frisian' (cf. map 18b). This is different for the complementiser dat 'that'. In English and German, the complementisers that and dass can be absent in many cases. This is much less so in Dutch and its dialects. Only when a Wh-word or relative pronoun is immediately preceding the position of the complementiser, is it possible or necessary to leave out the complementiser, as in Dat is de taal die (dat) ik als kind heb verworven 'That is the language that–singNonNeut (that) I acquired when I was a kid' (cf. map 16b) or Ik vraag me af hoe (of/dat) hij dat gedaan heeft 'I wonder how (if/that) he has done that' (cf. map 16a).
Related to this issue is how we should analyse the optional absence of dat 'that' in a sentence like Voor(dat) hij Nederlands leerde, sprak hij Kerkraads 'Before he learnt Dutch he spoke Kerkraads'. Parallel to the cases discussed above, we can analyse this as a case in which the complementiser has been left out. This would mean that voordat consists of two elements, namely, a preposition voor 'before' and a complementiser dat 'that'. The latter can be left out because it does not contribute to the meaning of the clause. Another option would be to analyse voor and voordat as two different complementisers. Both analyses have their own problems. On the one hand, the second analysis is supported by the fact that dat 'that' can not be left out in similar complementisers such as nadat 'after', opdat 'in order that' and zodat 'such that'. On the other hand, the fact that dat 'that' can often be added to a complementiser in colloquial speech, as in nu dat 'now that', zolang dat 'as long as' and dan dat 'than that' speaks in favour of the first analysis. A choice between these two analyses is beyond the scope of this atlas, but it should be clear that the notion of complementiser is not unambiguous.
A curious phenomenon in Dutch is the occurrence of superfluous complementisers. In relative and embedded interrogative clauses, one or even two superfluous complementisers may follow the relative pronoun or Wh-word. A sentence like Ik weet niet hoe of dat hij die taal geleerd heeft 'I don't know how if that he learnt that language' may not belong to the written standard language, but is quite common in spoken language. In addition to the standard variant Ik weet niet hoe hij die taal geleerd heeft 'I don't know how he learnt that language', we also find the following variants: Ik weet niet hoe of hij die taal geleerd heeft 'I don't know how if he learnt that language' and Ik weet niet hoe dat hij die taal geleerd heeft 'I don't know how that he learnt that language'. Map 16a shows whether this variation is geographically determined.
It is clear that the possibility to leave out the complementiser when there is no preposition, relative pronoun or Wh-word preceding it, as in English and German, is highly limited and can only be found in some of the Dutch dialects. We are dealing here with sentences like Jan denkt Fries vindt iedereen een moeilijke taal lit. 'John thinks Frisian finds everyone a difficult language' and Jan denkt hij spreekt Fries 'John thinks he speaks Frisian' (cf. map 17b).
1.1.3 Complementiser agreement
In many Dutch dialects, the complementiser may agree with the person and number features of the subject of the clause, just like the finite verb. This seems to be a rather unique phenomenon cross-linguistically. Standard Dutch does not exhibit it.
The facts are rather complex. A relevant example is found in the SAND data from Midsland, on the island of Terschelling. The informant provided us with the following sentences:
(1) a. Gienien mag 't zien dus ik fien dastou 't ook net zien meist.
no one may it see so I find dat-2Sing-you it too not see may
'No one is allowed to see it so I consider that you are not allowed to see it either.'
b. Beloof–st mij dast dat boek nooit meer verstoppen zoust?
promise-2Sing me that-2Sing that book never more hide would
'Do you promise me that you will never hide that book again?'
On the basis of this and similar phenomena, we conclude that dastou consists of three elements: dat+st+dou. Assimilation processes yield dastou. We are dealing here with the complementiser dat 'that', the inflectional suffix –st (that we also find on the verb beloven 'promise' in (1b)) and the second person singular subject pronoun dou. (1b) shows that this pronoun can be left out.
This type of phenomena does not only occur in the northern dialects but in many other areas, as is illustrated by the examples in (2).
(2) a. Anek profijtig leef, dan levik gelijk as mijn ouders zou willen.
if-1Sing-I frugally live then live-I like as my parents would want
'When I live frugally I live like my parents want me to.'
b. Gene mag het zieje dus ik ving dastu het ook niet
no one may it see so I find that-2Sing-you it also not
zieje mags.
see may
'No one is allowed to see it so I consider that you are not allowed to see it either.'
c. Ich denk daste veel moes weg gooie.
I think that-2Sing much must away throw
'I think you had to throw way a lot.'
In (2a), recorded in Sint-Niklaas (Oost-Vlaanderen), we see a combination of complementiser + inflection + pronoun 1 singular (als + n + ‘k). (2b) is an example of 2 singular complementiser agreement from Heerlen (Dutch Limburg) and (2c) is an example from Bree (Belgian Limburg).
A further complication is that several dialects have subject doubling in addition to complementiser agreement. This phenomenon is found in Belgium. It is discussed in chapter 3. An example of subject doubling is given in (3), from Sint-Niklaas (cf. 2a).
(3) Anekik profijtig leef, dan levik gelijk as mijn ouders zou willen.
if-1Sing-I-I frugally live, then live-I like as my parents would want
'When I live frugally I live like my parents want me to.'
Anekik, in (3), arguably consists of a complementiser (a = 'if, when'), agreement (–n) and a doubled subject pronoun (ekik).
We do not only find complementiser agreement with the complementisers dat 'that' and als ‘if’, but also with other complementisers. Some examples with the complementisers nu 'now' and toen 'then' are given in (4), and in (5) we find agreement in Wh-clauses in which the Wh-word is not followed by a complementiser. In (6), we have a sequence consisting of Wh-word + complementiser + inflection + pronoun (waar 'where' + dat 'that' +–st-Agr2Sing + du 'you'.
(4) a. Noustu klaar bis kanstu gaan. (Borgharen)
now-2sing-you ready are can-2sing-you go
'Now that you are ready you may go.'
b. Het gebeurde toenstu weggunst (Spijkerboor)
it happened when-2sing-you left-agr
'It happened when you left.'
(5) Ik weet woarstou geboren bis. (Kloosterburen)
I know where-2sing-you born are
'I know where you were born.'
(6) Ik weet woardastu geboren bist. (Onstwedde)
I know where-that-2sing-you born are
'I know where you were born.'
Complementiser agreement occurs with non-pronominal subjects (7) and all pronominal subjects (8).
(7) a. 'k Gelove wel da Lisa zo schoon is of Anna. (Oostende) I believe affirm that Lisa as beautiful is as Anna
'I do believe that Lisa is as beautiful as Anna.'
b. Hij gelooft dan Bart en Peter sterker zijn (Doel) he believes that-plur Bart and Peter stronger are
as Geert en Jan.
than Geert and John
'He believes that Bart and Peter are stronger than Geert and Jan.'
(8) a. Ank zin heb dan doenk et. (Hulst)
when-1sing-I lust have then do-1sing-I it
'When I feel like doing it, I do it.'
b. Ast gezond leefst, leefst nog langer. (Groningen)
when-2sing healthily live-2sing live-2sing even longer
'If you live healthily, you live even longer.'
c. At hee nog drie jaar leeft dan leefter langer (Eibergen)
if-3singhe yet three year lives then lives-he longer
as zie vader.
than his father
'If he will live three more years, he will have lived longer than his father.'
d. At zie zo gevaarlijk leeft leefse niet lang meer. (Eibergen)
if-3singshe so dangerous lives live-she not long more
'If she is living so dangerously she won’t have long to live.'
e. Azzeme sober leve leve me gelukkig. (Monster)
if-1plur-we frugally live live we happily
'If we live frugally, we live happily.'
f. An judder zo losbandig leve (Oostburg)
if-2plur you-plur so fast live
g. Anze voor ulder werk leven, levenze nie (Oostveld)
if-3plur-they for their work live live-3plur-they not
voor udder joenges ja
for their children yes
'If they live for their work they don't live for their children.'
i. And’r zo weinig mense van de boerenstiel leven, ... (Oostende)
if-plur-there so few people from the agriculture live
'If so few people are living from agriculture, ....
However, it is not the case that all dialects with complementiser agreement have a full paradigm. On the contrary, only some dialects in (Zeeuws-)Vlaanderen have an almost complete paradigm, whereas most other dialects have complementiser agreement only for one, or a couple of specific members of the paradigm, often second person singular.
There appears to be a correlation between the form of the suffix on the complementiser and the form of the suffix on the finite verb. This has been shown in several of the examples above. This generalisation goes back to Van Haeringen (1958) and has been stated explicitly in Hoekstra and Smits (1997) as the inversion generalisation: when a complementiser agrees with the subject, the agreement suffix is identical to the suffix on the finite verb in inversion contexts. We see this in the sentences (1b), (4a) and (8a). In these examples, the suffix on the complementiser (–st, –st and –n, respectively) is identical to the suffix on the verb in inversion contexts (beloofst, kanstu and doenk). To test the validity of this generalisation this atlas contains a number of maps that correlate complementiser inflection with verbal inflection. In addition, these maps show whether the suggestion in Goeman (1979) is correct that only the inflection paradigm of verbs with a monosyllabic infinitive (e.g. gaan ‘go’) is relevant for the inversion generalisation.
1.1.4 Historical development
In Middle Dutch, the conditions on the use of the complementiser dat 'that' were less strict than in contemporary Dutch. Dat 'that' could be left out in declarative subordinate clauses (Bouman 1918), even when the finite verb was clause final. In addition, dat 'that' could follow Wh-words and relative pronouns (cf. Stoett 1923: 225-227), similar to many contemporary dialects of Dutch in Belgium (maps 16a,b). The comparative complementiser of 'as if' could introduce a verb-second clause in Middle Dutch (Stoett 1923: 230). Map 14b shows that this option exists in some contemporary Antwerp dialects too.
Very little is known about the historical development of complementiser agreement. A systematic investigation of Middle Dutch texts has not yet been undertaken. Goeman (1997b) notes that complementiser agreement seems to be much less frequent in Middle Dutch texts than might be expected on the basis of the current dialects. Only the second person singular –ste ending shows up in late Middle Dutch. Nevertheless, Goeman does not want to conclude that complementiser agreement is a recent development, as there are a number of attestations which are much older, such as plural agreement in manuscript D of van Maerlant's Der Naturen Bloeme (Brugge, 1287 or earlier), and in a manuscript from Delft, dating from around 1434.
1.2 Discussion of the literature
The literature on complementiser doubling and the occurrence of complementisers following Wh-words is very limited. The ANS mentions these phenomena and notes that there are various regional options (of 'if' in the Netherlands, dat 'that' in Belgium and Noord-Brabant). Hoekstra (1993a, 1993b) discusses sentences with three complementisers in dialects in West-Vlaanderen and Frans-Vlaanderen and in Frisian. He concludes that there are three structural complementiser positions in embedded clauses, each with its own contribution to the meaning (cf. also Hoekstra and Zwart 1994). Taking into account data from Zeeland dialects, Sturm (1996) argues that ofdat 'if-that' is one complementiser. Bennis (1997) draws a parallel between Wh-word + of + dat and the construction wat voor een boeken lit. what for a books, 'what kind of books'. He analyses voor een 'for a' as a case of complementiser doubling. Bouman (1918), Stoett (1923) and Weerman (1989) show that, in Middle Dutch, the complementiser dat 'that' could occur after toen 'then', terwijl 'while', indien 'provided' and tenzij 'unless'. This option still exists in some of the Dutch dialects. The occurrence of dat 'that' following relative pronouns is discussed in Pauwels (1958) for the Aarschot dialect and for Middle Dutch in Verwijs and Verdam (1889). For a recent analysis in an Optimality Theoretic framework, see Dekkers (1999). The optionality of the complementiser dat 'that' in declarative subordinate clauses in Middle Dutch is discussed in Bouman (1918). The occurrence of the complementisers van 'of' and voor 'for' in infinitival clauses where Standard Dutch would have om is discussed in Ryckeboer (1983) and Gerritsen (1991). For the use of the complementiser van 'of' in infinitival clauses see also Van Craenenbroeck (2000).
An extensive historiography of complementiser agreement research is provided in Goeman (1997). On the basis of a literature survey Hoekstra and Smits (1997; henceforth H&S) discuss the geographic distribution of complementiser agreement and the pertinent analyses and generalisations that have been proposed. They also address the distinction between dialects with person agreement and dialects with number agreement on the complementiser. For a brief discussion of the geographic distribution and a map of complementiser agreement in the Netherlands (with the exception of Friesland) and Belgium, see Goeman (1980b). Vanacker (1949) provides an extensive description of the distribution of plural complementisers in Vlaanderen, with maps.
The geographic distribution that H&S describe is strikingly similar to the findings of the SAND project (cf. maps 19a-21a). The language area can be divided into a western part with complementiser agreement, interrupted by Goeree-Overflakkee (Zuid-Holland), Schouwen-Duiveland and Noord-Beveland (Zeeland), an area where no complementiser agreement is present. In the Dutch part of this western area, we primarily find number agreement. The eastern part of the language area also has complementiser agreement, primarily for person. Between this western and eastern part there is central area without complementiser agreement, consisting of the provinces of Belgian Brabant, Antwerpen, Noord-Brabant, Utrecht and a part of Gelderland. The only difference between our findings and those of H&S is the almost complete absence of complementiser agreement in Drenthe in our material. The close similarity between H&S's findings and our own is all the more remarkable because H&S base themselves on data from the literature from the first half of the 20th century. In any case, on the basis of the SAND data we can conclude that complementiser agreement is still present at the beginning of the 21th century in the generation aged between 55 and 70 years.
The earliest analyses of complementiser agreement are functional. Beckering Vinckers (1872) considers complementiser agreement to be phonetic anticipation, a speech error in which the verbal inflection that should come later is realised on the complementiser. However, Goeman (1980a) argues that in speech errors the distance between the position where the morpheme is realised and the position where it should have been realised is usually not longer than two words. In contrast, the distance between an inflected complementiser and the clause final finite verb can be much longer. Van Haeringen (1958) thinks that complementiser agreement is syntactic anticipation that helps to bridge the distance between the subject and the finite verb of the embedded clause, but he himself notes that complementiser agreement also occurs when the subject and the finite verb are adjacent. Both anticipation analyses assume that the suffix on the complementiser is a copy of the suffix on the finite verb. H&S show that this copying generalisation is wrong. On the basis of van Haeringen's (1958) description of eastern Dutch dialects with double verb inflection (in dialects with double verb inflection the inflection on the finite verb in inversion contexts differs from the inflection on the finite verb in regular order and in embedded sentences), they formulate the inversion generalisation which states that the suffix on the complementiser is identical to the suffix on the finite verb in inversion contexts. Goeman (1979) adds to this that only the inflection on verbs with a monosyllabic infinitive (e.g. gaan 'go') is relevant, not verbal suffixes in general. As previously discussed, this atlas shows to what extent the inversion generalisation is correct and whether there is a difference between monosyllabic and polysyllabic verbs.
An important complication for the analysis of complementiser agreement is that it often occurs with an enclitic subject pronoun. It is often hard to decide whether a particular sequence of sounds involves a suffix on the complementiser, an enclitic subject or both. This is particularly the case for the morpheme –de (2 singular) in Brabantish dialects (De Schutter 1997), the morpheme –st (2 singular) in Frisian and its dialects (De Haan 1997) and for -me and -men (1 plural) in some Belgian dialects.
Important issues in recent syntactic analyses include the possible correlation between the verb second property of the West-Germanic languages (the property that the finite verb is in second position in main clauses) and the occurrence of complementiser agreement (cf. Hoekstra and Maracz 1989 and Zwart 1993, 1997), the question as to whether complementiser inflection represents tense information (H&S 1997, 1999), the question as to whether the complementiser agrees with the subject or the finite verb (Carstens 2003), the difference between complementiser agreement before pronominal and non-pronominal subjects (De Vogelaer 2003), locality conditions on complementiser agreement (Ackema and Neeleman 2001, Van Craenenbroeck and Van Koppen 2003) and the possibility of complementiser agreement with only the first conjunct of a coordinated subject (Van Koppen 2005).
Typological research (Corbett 1991, Siewierska 2004) shows that complementiser agreement is an extremely rare phenomenon (cf. also Goeman 1997b).
1.3 Discussion of the maps
1.3.1 Complementisers
1.3.1.1 Complementiser of declarative clause (map 14a) (map in dynaSAND)
The SAND material includes a large number of sentences containing a declarative subordinate clause introduced by the complementiser dat 'that'. A selection of these sentences is given in (9). The informants were asked to translate these sentences. The complementisers in (9) and many other examples in this chapter are in italics to indicate that the form of this complementiser may vary across dialects.
(9) a. Ik denk dat Marie hem zal moeten roepen.
I think that Mary him will must call
'I think that Mary will have to call him.'
b. Ze geloven dat wij rijker zijn als zij.
they believe that we richer are than they
'They believe that we are richer than they are.'
c. U gelooft dat Lisa even mooi is als Anna.
you believe that Lisa as beautiful is as Anna
'You believe that Lisa is as beautiful as Anna.'
d. Ik weet dat Eddy brood wil eten.
I know that Eddy bread wants eat
'I know that Eddy wants to eat bread.'
Map 14a shows the geographic distribution of the different forms for dat 'that' in sentence (9a). There is not much dialectal variation. Four Frans-Vlaamse dialects have e, Stavenisse in Zeeland has a. The rest of the dialects has da or dat. The Belgian dialects have da, with the execption of a large part of Belgian Limburg. In the Netherlands, da is dominant in Noord-Brabant and to a lesser extent in Zeeland. It also occurs sporadically in the central-eastern and northeastern part of the language area. The remaining parts of the language area have almost exclusively dat. The geographic distribution of da and dat is remarkably constant across the various test sentences. Hence, (9a) is representative of the other test sentences.
The most important reason to include this map in this atlas is that the localisation of the da-area may be helpful to establish whether a certain dialect has –t inflection on the complementiser. When a dialect has dat before a 3 singular subject pronoun but da before all other subject pronouns, then it is not unlikely that dat consists of da and a –t suffix. This is particularly true for dialects that have da before vowel-initial names, as in (9d). In such dialects, the da - dat alternation cannot be captured with a phonological rule 'delete final –t of dat before consonants'. In addition, it is clear from the literature (cf. Goeman 1999: 414) that t–deletion before vowels is extremely rare in the Belgian dialects. As map 14a shows, half of the dialects with da before consonants also have da before vowels.
1.3.1.2 Complementiser of comparative of 'if' clause (map 14b) (map in dynaSAND)
The subject clause of lijken 'seem' in (10a) is introduced by the complementiser of 'if'. We call this instance of the complementiser ‘comparative of’. Its occurrence is not restricted to subject clauses of lijken 'seem' (10b) and it can also introduce clauses with a different syntactic function, such as the adverbial clause in (10c). Southern dialects often lack the verb lijken 'seem'. Many Belgian dialects use the construction 't is net of' 'it's as if' instead.
(10) a. Het lijkt wel of er iemand in de tuin staat.
it looks affirm if there someone in the garden stands
'It seems like there is someone standing in the garden.'
b. Het is of je een emmer leeg gooit.
it is if you a bucket empty throw
'It is as if you empty a bucket.'
c. Hij doet / praat of hij de baas is.
he does /talks if he the boss is
'He behaves / talks as if he were the boss.'
Map 14b shows that the language area has seven different forms for comparative of 'if'. Most of the dialects have the same form as the standard language. In the Netherlands, the form of is highly dominant, with the exception of Friesland where of only occurs in the peripheral parts of the province. The rest of Friesland has dat or at instead of of. There is one case of as, in Lemmer, and one case of complementiser doubling, of at in Oosterbierum. Because the informants were asked to translate this sentence, the map does not provide an overview of all possible forms for each location, but an overview of the preferred forms.
Complementiser doubling, such as of dat, also occurs in a large part of Noord-Brabant and in some places in Utrecht and Zeeland. In Belgium, of dat is the most frequent form, except Belgian Limburg that has of. In addition, in Belgium dat often replaces of. This happens in all Dutch speaking provinces of Belgium, in particular in Brabant. Frans-Vlaanderen does not have dat instead of of. In three locations, it has et. In Belgian Limburg and the province of Antwerpen, the use of dat instead of of is very rare. Interestingly, in a number of dialects of or as is followed by a verb-second clause:
(11) ’t Is precies of der staat enen in den hof (Nieuwmoer)
it is precisely if there stands one in the garden
‘It seems like there is someone standing in the garden.’
This phenomenon is found seven times in an area consisting of the northern part of the province of Antwerpen and part of Waasland, in the northeastern part of Oost-Vlaanderen. We also find it twice in southeastern Vlaams-Brabant and twice in Friesland. The rareness of as + verb-second in Friesland is surprising because this construction occurs frequently in every day speech. This could be a task effect. Apparently, the informants translated this sentence without changing the original word order.
1.3.1.3 Complementiser of conditional clause (map 15a) (map in dynaSAND)
The SAND data contains conditional clauses introduced by als ‘if’ for all members of the paradigm of subject pronouns and also for lexical subjects. The informants were asked to translate the test sentences. This yielded a detailed picture of the different forms for als ‘if’. Some of the test sentences are given in (12). Like map 14a (section 1.3.1), this map is relevant for the attestation of complementiser agreement.
(12) a. Als Rudy nog leeft dan leeft Leo ook nog.
if Rudy still lives then lives Leo also still
‘If Rudy is still alive then Leo must be alive too.’
b. Als Pieter en Liesje in het paradijs leven dan leven
if Pieter and Liesje in the paradise live then live
Rosa en Frans in de hel.
Rosa and Frans in the hell
‘If Pieter and Liesje are living in paradise then Rosa and Frans are living in hell.’
c. Als jullie zo losbandig leven dan leven jullie nooit zo lang als ik.
if you so fast live then live you never as long as I
‘If you live that fast then you will not live as long as I.’
We find no fewer than 9 different forms for als ‘if’. Many of these forms look very similar and may be phonological alternants. We will not try to decide here whether this is the case. The largest part of the language area, almost 200 dialects, has as. We also find a minority of places with als in this area. In Friesland and southeastern Gelderland, we find at. Five locations in southeastern Dutch Limburg have wen. In the southwestern part of the language area, we find e in Frans-Vlaanderen and o, a, a/o in West-Vlaanderen. Zeeuws-Vlaanderen (southern Zeeland) also chooses a/o. In dialects with a/o a and o are in free variation. Finally, along the western border of Oost-Vlaanderen and in one location in West-Vlaanderen (Brugge) we find os.
1.3.1.4 Complementiser in comparatives (map 15b) (map in dynaSAND)
In Standard Dutch, the complementiser in a comparative may be als ‘as’ or dan ‘than’ (cf. the examples in (13)). We call these words ‘complementisers’ even though they are followed by a pronominal constituent, rather than a (visible) clause. Possibly, als/dan function as prepositions in these kinds of contexts. This would explain why the pronouns following them get non-nominative case in many varieties.
(13) a. Zij gelooft dat jij eerder thuis bent dan ik.
she believes that you sooner home are than I
‘She believes that you’ll be home sooner then me.’
b. Jullie geloven jammer genoeg niet dat zij armer zijn als jullie.
you believe unfortunately not that they poorer are as you
‘Unfortunately, you don’t believe that they are poorer than you.’
According to normative grammars, the correct form is dan ‘than’ in Standard Dutch. However, in the great majority of dialects we find als ‘as’. This form dominates the entire language area except West-Vlaanderen and Oost-Vlaanderen. The form dan ‘than’ is also frequent, often as an alternative for als. In Zuid-Holland and southern Noord-Holland where the dialects are most closely related to the standard language, dan ‘than’ is usually the only possible form. However, because we used a translation task, this does not mean that als ‘as’ is excluded as a possible option. A lot of variation is found in the southern part of the language area. Southeastern Dutch Limburg has wie. Four places in Antwerpen have gelijk (as). In Frans-Vlaanderen, West-Vlaanderen and western Oost-Vlaanderen, of is the dominant form.
1.3.1.5 Finite complementiser(s) following question word (map 16a) (map in dynaSAND)
In Standard Dutch, a subordinate clause with an initial Wh-word does not contain a complementiser (14a). In colloquial Dutch, a complementiser is possible, and the same holds for many of the dialects. We find three variants (14b-d). All of the sentences in (14) were presented to the informants in their own dialects and they were asked whether these sentences occurred in their dialects. In (14) and on the map, we abstract away from the precise form of the complementiser in the position of dat. In some of the dialects, at or as occurs after Wh-words.
(14) a. Vertel maar niet wie zij had kunnen roepen.
tell just not who she had can call
‘You had better not tell her who she could have called.’
b. Vertel maar niet wie dat zij had kunnen roepen.
tell just not who that she had can call
‘You had better not tell her who she could have called.’
c. Vertel maar niet wie of zij had kunnen roepen.
tell just not who if she had can call
‘You had better not tell her who she could have called.’
d. Vertel maar niet wie of dat ze had kunnen roepen.
tell just not who if that she had can call
‘You had better not tell her who she could have called.’
Map 16a answers the question as to whether the different variants show a regional distribution or simply belong to the common language. We find clear differences between Belgium and the Netherlands. In Belgium, wie dat ‘who that’ is the most common construction, whereas in the Netherlands, this is wie ‘who’ without a complementiser. In Vlaanderen (Zeeuws-Vlaanderen, Oost-Vlaanderen, West-Vlaanderen and Frans-Vlaanderen), wie dat ‘who that’ is the only possibility, whereas in the eastern half of Dutch speaking Belgium (Antwerpen, Vlaams-Brabant and Belgisch Limburg), the complementiser can be left out in most dialects. In the Netherlands, we find wie dat ‘who that’ primarily in Zeeland (in particular in Zeeuws-Vlaanderen), Noord-Brabant and Friesland. The construction wie of ‘who if’ is restricted to the Netherlands and in many places it functions as an alternative for wie ‘who’ without a complementiser. It does not show a clear geographic pattern, occurring a couple of times in every Dutch province. The construction with two complementisers (wie of dat ‘who if that’) is much less frequent than the other three constructions. It does not occur in Belgium, with the exception of Brugge (West-Vlaanderen). In Noord-Brabant and Friesland the construction also occurs only once, in Steenbergen and Lemmer. The remaining Dutch provinces all have some occurrences of wie of dat ‘who if that’.
1.3.1.6 Finite complementiser(s) following relative pronoun (map 16b) (map in dynaSAND)
Map 16b shows the geographic distribution of dialects in which a relative pronoun is followed by a complementiser. Four different constructions are represented: locative relative clauses (sentence 15a), subject relatives (sentence 15b), free relatives (15c) and object relatives (15d). The informants were asked to translate these sentences into their own dialects. We actively tested the occurrence of complementisers. This map abstracts away from the different forms that the complementiser may have (cf. maps 84a,b, 88a and 89a for different forms of the complementiser)
(15) a. de bank waar op da ze zaten (Geraardsbergen)
the bench where on that they sat
‘the bench on which they were sitting’
b. Dat is de man die as het verhaal verteld hef. (Staphorst)
that is the man who that the story told has
‘That is the man who has told the story.’
c. Wie da geld eet moet mij maar wa geve. (Zandhoven)
who that money has must me just what give
‘Whoever has got money should give me some.’
d. Dat is de man die as ze geroepen he. (Stavenisse)
that is the man who that they called have
‘That is the man who they have called.’
There are two core areas that have a complementiser following a relative pronoun. The first is in the south and includes Zeeuws-Vlaanderen (southern Zeeland), Waasland (northeastern Oost-Vlaanderen) and the province of Antwerpen. Most measuring points in this area have a complementiser in three and sometimes even in all four of the constructions. The same holds for the second core area, Friesland, where more than half of the measuring points has a complementiser in all four constructions. When we compare the distribution per construction, we see that complementisers occur much more frequently in the locative relative (sentence 15a) than in the other three constructions. We find the locative relative with a complementiser in all Dutch speaking provinces of Belgium, Frans-Vlaanderen and Dutch Limburg. Within the Netherlands, this phenomenon is restricted to a zone along the eastern border from Limburg up to, but not including, Groningen, and also in Zeeland and Friesland. The distribution of subject relatives with a complementiser is very similar to that of object relatives with a complementiser. In the Netherlands, they hardly show up outside the core areas described above. In Belgium, the measuring points in the core area have both constructions. Outside this area we find a number of places where subject relatives are given with a complementiser but object relatives without one. Free relatives with a complementiser occur in the core areas and in West-Vlaanderen and Oost-Vlaanderen.
1.3.1.7 Correlation finite complementiser following question word and relative pronoun (map 17a)
The goal of map 17a is to answer the question as to whether the dialects with a complementiser following a Wh-word are the same as the dialects with a complementiser following a relative pronoun. The data of maps 16a,b have been combined here, and there is no differentiation for relative construction type. We see that almost all measuring points in Belgium may have a complementiser both after Wh-words and after relative pronouns. In the Netherlands, this holds for the zone along the eastern border up to Groningen, and alsofor Friesland and Zeeland. The rest of the Netherlands only has complementisers after Wh-words. There are no clear areas where complementisers are totally excluded.
1.3.1.8 Finite complementiser absent (map 17b) (map in dynaSAND)
When a subordinate clause in Standard Dutch does not start with a Wh-word or a relative pronoun, the complementiser cannot be left out. Some dialects have constructions in which this is possible. The relevant constructions are given in (16). In sentence (16a), the subordinate clause has main-clause word order (verb second) and there is a fronted constituent in first position. In (16b), the subordinate clause has main-clause word order too, with the subject in first position. The testing of these sentences was delicate because they are possible in Standard Dutch as well provided that there is an intonational break following geloof ‘believe’ and denk ‘think’. In such cases, the clause following these verbs is not a subordinate clause but a main clause in direct speech. Therefore, these sentences were presented to the informants with a context (the part between brackets) and the field workers and assistant interviewers were instructed to read the sentence as one prosodic unit, i.e., with the intonation of one sentence and without a break.
(16) a. (Ik zei nog tegen haar): Ik geloof deze jongen vinden ze allemaal
(I said to her): I believe this boy find they all
wel aardig
well nice
‘(I said to her), I believe they all like this guy.’
b. (Ik zei nog tegen haar): Ik denk hij is weg.
(I said to her): I think he is gone.
Map 17b shows that these constructions (and hence the possibility to leave out the complementiser) do not occur in Belgium. In the Netherlands, a complementiserless subordinate clause with a fronted constituent (16a) occurs primarily in Friesland and furthermore in an eastern zone along the border of the language area. This may be due to the influence of Standard German which also allows this construction. A complementiserless subject initial clause with main-clause word order (16a) primarily occurs in Noord-Holland and Friesland. The map clearly shows that there is no correlation between the two phenomena.
1.3.1.9 Obligatory infinitival complementiser (map 18a) (map in dynaSAND)
The complementiser om ‘for’ introduces infinitival clauses and cannot be left out when the infinitival clause is a purpose clause, as in sentence (17). The informants were asked to translate this sentence, and to give possible alternatives for om.
(17) Heb je genoeg mensen om hooi van het land te halen?
have you enough people for hay from the land to get
‘Do you have enough people to bring in the hay?’
The variation attested here involves the choice of the complementiser. Map (18a) shows that most dialects in Belgium can use both voor and om in these contexts. West-Vlaanderen is an exception, as most dialects have voor as the only option there. Frans-Vlaanderen almost exclusively has om. There is not much variation in the Netherlands. The dialects of Dutch Limburg behave as do most dialects in Belgium. Noord-Brabant also has four locations with two complementisers, and there are four eastern locations with only voor. This map is strikingly similar to the map given in Ryckeboer (1983) which is based on dialect data dating from 1922-1976. Ryckeboer’s map shows the different complementisers used in the sentence Ik moet ossebloed drinken om te verkloeken lit. I must ox-blood drink for to recover ‘I have to drink ox-blood to recover.’ On this map, which is restricted to Dutch speaking Belgium including Frans-Vlaanderen, Frans-Vlaanderen is a homogeneous om area whereas a large part of West-Vlaanderen is a homogeneous voor area. The rest of Belgium has both voor and om. Voor is the dominant form in Vlaams-Brabant and Belgian Limburg. Ryckeboer ascribes the use of voor to influence of Standard French (pour).
1.3.1.10 Optional infinitival complementiser (map 18b) (map in dynaSAND)
In example (18), the clause om te komen werken is the actual subject of is ‘is’. The infinitival clause does not denote a purpose and many speakers of Standard Dutch can leave out om ‘for’.
(18) Het was aardig van Jan om te komen werken.
it was kind of John for to come work
‘It was kind of John that he came and worked.’
The informants were asked to translate the sentence in (18) and to indicate whether it was possible to replace the complementiser by a different form or to leave it out. Map 18b shows another big difference between Belgium and the Netherlands. In all Dutch-speaking provinces of Belgium, the complementiser van ‘of’ is highly frequent. There are dialects where this is the only form. This holds for West-Vlaanderen, Vlaams Brabant and Limburg and also for Frans-Vlaanderen. There are also dialects that have om in addition to van. These two groups of dialects do not show a clear geographic pattern. The complementiser van does not occur in the Netherlands. A second difference between north and south is the complementiser voor, which is restricted to the southern part of the language area. In this case, Dutch Limburg patterns with the Belgian provinces; the rest of the Netherlands does not have voor. The complementiser voor is most frequent in Dutch and Belgian Limburg and in West-Vlaanderen and Oost-Vlaanderen. There are not many dialects that only have voor; usually om is possible too. In West-Vlaanderen, we find four locations that allow all three of the complementisers. Only eighteen dialects allow omission of the complementiser. Almost half of these are located in the southeastern part of the language area, i.e., Vlaams-Brabant and Belgian Limburg, and Dutch Limburg. There are five dialects with complementiser omission in the periphery of Friesland, three in Holland and two in Gelderland. The majority of the dialects in which the complementiser can be left out have om as an alternative.
1.3.2 Complementiser agreement; general
1.3.2.1 Complementiser agreement; synthesis (map 19a)
In Standard Dutch, the inflection on the finite verb is determined by the number and person of the subject. In many dialects of Dutch the same holds for the complementiser. Map 19a provides on overview of all dialects that have complementiser agreement in at least one member of the paradigm. In this and all other complementiser agreement maps, all attestations of complementiser agreement, both from the oral and the telephonic interviews are taken together. The material includes both translation and indirect judgement tasks. Cases of complementiser agreement in sentences that were presented for other purposes are included as well. The SAND data do not provide an answer to the question as to whether there are dialects in which complementiser agreement is obligatory. In many measuring points the informants give equivalent sentences with and without complementiser agreement. This could mean that complementiser agreement is optional, but also that the variant without complementiser agreement is part of the regional or standard language. Map 19a gives two perspectives on the data. The colored areas are the result of kernel estimation (cf. section 0.3.2 and Wattel & Van Reenen 1994). The white diamonds are dialects that have complementiser agreement, and the black diamonds are dialects that do not have it.
Complemeniser agreement roughly divides the language area into three areas: a western area with complementiser agreement, a central area without it, and an eastern area with complementiser agreement. The province of Drenthe is an exception within this eastern area. There are some very homogeneous complementiser agreement areas. In Vlaanderen (i.e., Frans-Vlaanderen, West-Vlaanderen, Oost-Vlaanderen en Zeeuws-Vlaanderen), all measuring points except one have complementiser agreement. Another homogeneous complementiser agreement area is found in the southeastern part of the language area. It consists of the lower-half of Dutch Limburg and the eastern half of Belgian Limburg. Other core areas can be found in the north and the east. In the north, this involves an area consisting of the upper part of Noord-Holland, a diagonal strip in Friesland and the whole of Groningen. In the east, this is an area consisting of eastern Overijssel and Gelderland.
1.3.2.2 Complementiser agreement; all forms (map 20a)
Map 20a provides an overview of all complementiser agreement suffixes that have been attested. It is the form of the suffixes that is relevant here, not the person or number information that they provide. All attested suffixes also occur as verbal agreement suffixes in the language area. There are five different forms. The suffix –t is the most frequent. We find it in a southwestern area that includes Vlaanderen (i.e., Frans-Vlaanderen, West-Vlaanderen, Oost-Vlaanderen and Zeeuws-Vlaanderen), the northwestern part of the province of Antwerpen and some places in Zeeland and western Noord-Brabant. The suffix –t also occurs in eastern Overijssel and Gelderland and in some southern dialects of Dutch Limburg and Belgian Limburg. The second most frequent suffix is –n. This suffix is found almost exclusively in the southwestern part of the language area described above for the suffix –t. There are two rather isolated attestations in the Netherlands, in Strijen (Zuid-Holland) and Lopik (Utrecht). The suffix –s(t) is almost as frequent as –n. It is realised as –s in some places and contexts and as –st in others. In many cases it is impossible to distinguish these two. We therefore treat them as one suffix here. We find the suffix –s(t) in the eastern part of the language area: Friesland, Groningen, some eastern dialects in Overijssel and in Belgian Limburg and Dutch Limburg. In most of these dialects, –s(t) is the only form that occurs, being used in 2 singular. The suffix –e (schwa) is frequent in the Netherlands in Holland, Overijssel and Gelderland. In Belgium this suffix only occurs in Oost-Vlaanderen. The suffix –(n)t occurs only three times, in Heerlen, Kerkrade and Nieuwenhagen (Zuid-Limburg).
1.3.2.3 Complementiser agreement; all persons (map 21a)
Map 21a shows, for each dialect, which members of the paradigm have audible complementiser agreement. Doubtful cases in 2 singular and 1 plural are not included on this map. (cf. map 25a and maps 29a, 35a-c). Since the individual members of the paradigm are discussed in the sections that follow, we only discuss frequent combinations here. One such combination is 1 plural and 3 plural. We find many dialects with this combination in the Netherlands, in particular in Holland, Gelderland and Overijssel. A second frequent combination is 3 singular and 3 plural. This combination shows up in particular in West-Vlaanderen and a little less in Oost-Vlaanderen. These dialects may be characterised as dialects with agreement for third person. Typical for the southwestern edge of West-Vlaanderen and for Frans-Vlaanderen is the combination of 1 singular, 3 singular, 3 plural, in two cases combined with 1 plural (Oostende and Izenberge). The combination of 1 singular, 3 singular, 3 plural is found several times in Oost-Vlaanderen and the neighbouring part of West-Vlaanderen and twice in the east of the Netherlands. Dialects with agreement in more than three members of the paradigm are rare but do exist. In addition to the two dialects just mentioned, they can be found in Zeeuws-Vlaanderen, in different combinations. In Hoek and Hulst only 2 singular is lacking. In Oostburg 2 singular and 3 plural are missing. In southern Oost-Vlaanderen we find two places (Bevere and Meilegem) where only 1 singular and 2 singular are lacking, and one location (Merelbeke) which has no inflection in 2 singular and 1 plural.
The attested patterns are not clear enough to distinguish dialects with only person agreement, only number agreement, and dialects with a full paradigm. The dialects with inflection in 1 plural and 3 plural may perhaps be characterised as dialects with plural inflection if we assume that 2 plural is not always a real plural. Evidence from Standard Dutch supporting this view is that we have jullie hebt ‘you guys have-2sing’ in addition to jullie hebben ‘you guys have-2plur’, and that the form of 2 plural possessive and reflexive pronouns is identical to their 2 singular equivalents, namely je. Another potential cause of the incompleteness of inflection paradigms is that we cannot decide on the basis of the SAND material whether the complementiser has a zero suffix in certain cases. If we took identity with the verbal paradigm as a criterion instead of overt inflection, then there could be more dialects with a complete paradigm.
1.3.3 Complementiser agreement first person singular
1.3.3.1 Complementiser agreement 1 singular; synthesis (map 22a)
Complementiser agreement 1 singular occurs in sentences like (19).
(19) a. Als ik zuinig leef, leef ik zoals mijn ouders willen.
if I frugally live live I like my parents want
‘If I live frugally, I live like my parents want me to.’
b. Ik geloof dat ik groter ben dan hij.
I believe that I taller am than he
‘I believe that I am taller than he is.’
Map 22a provides all cases of 1 singular complementiser agreement. Locations with a white diamond have complementiser agreement, locations with a black diamond do not. We see that Frans-Vlaanderen is the core area. The dialects investigated there have complementiser agreement without any exception. Southwestern West-Vlaanderen links up with this core area. In addition, we find some dialects with complementiser agreement in and around Zeeuws-Vlaanderen (Waasland). An isolated case is Merelbeke, in central Oost-Vlaanderen. This is probably not complementiser agreement, as the same morpheme (–n) shows up in other members of the paradigm as well.
1.3.3.2 Complementiser agreement 1 singular (map 23a) (map in dynaSAND)
This map is also based on the sentences in (19). The map compares the occurrence of complementiser agreement on dat ‘that’ and als ‘if’ and it also shows that there is only one 1 singular suffix, –n. Examples are dan ik ‘that-1Sing I’ and an ik ‘if-1Sing I’. Dialects in the core area have agreement both on dat and als. In Zeeuws-Vlaanderen and its surroundings we find three dialects with agreement only on als.
1.3.3.3 Correlation complementiser and verb agreement 1 singular (map 23b)
Map 23b compares the presence of the suffix –n on the complementiser (indicated with C in the map key) and the presence of the same suffix on finite verbs in present tense inversion contexts. We see that twelve out of eighteen dialects with 1 singular complementiser agreement –n also have gaan ik (= ga +–n ‘go +–n’). This does not completely confirm the generalisation suggested in the literature (cf. Goeman 1979) that the suffix on the complementiser is identical to the suffix on monosyllabic verbs in inversion contexts, although there is a clear tendency in the right direction. It is clear from this map that the suffix on polysyllabic verbs in inversion contexts does not play any role. Only four of the dialects with complementiser agreement also have –n on leev ‘live’.
1.3.4 Complementiser agreement second person singular
1.3.4.1 Complementiser agreement 2 singular; synthesis (map 24a)
This map shows that complementiser agreement 2 singular is a northeastern and southeastern phenomenon. Such agreement occurs on the complementiser in the following types of sentences.
(20) a. Ze gelooft dat jij eerder thuis bent dan ik.
she believes that you earlier home are than I
“She believes that you will be home sooner than me.’
b. Als je gezond leeft dan leef je langer.
if you healthily live then live you longer
‘If you live healthily you will live longer.’
c. Ik weet waar je geboren bent.
I know where you born are
‘I know where you were born.’
d. Het gebeurde toen je wegging.
it happened when you left
e. Nu je klaar bent mag je gaan.
now you ready are may you go
‘Now that you are ready, you may go.’
All of these sentences involved translation tasks. Map 24a gives all the dialects that have at least one case of complementiser agreement. Places with a white diamond have complementiser agreement, locations with a black diamond do not. We should add to this that it is not entirely clear whether we are dealing with complementisers in (20c-e), since waar ‘where’, toen ‘then’ and nu ‘now’ can also occur as independent constituents and there are dialects in which they are followed by a complementiser. All measuring points in the province of Groningen and all measuring points in Friesland except three (Oosterbierum, Hollum and Anjum) have complementiser agreement. In Twente (in the east of Overijssel), we find three dialects with agreement: Rossum, Noord-Deurningen and Losser. In the southeast, almost all dialects in Dutch Limburg have complementiser agreement except three dialects in northern Limburg (Maasbracht, Maasbree and Meterik). The eastern half of Belgian Limburg has complementiser agreement without any exception.
1.3.4.2 Complementiser agreement 2 singular (map 25a) (map in dynaSAND)
Map 25a uses squares in different colours for the different complementisers in the sentences in (20). The map shows that inflection on the different complementisers strongly correlates. The large majority of the dialects with 2 singular complementiser inflection have inflection on four complementiser types (dat ‘that’, waar ‘where’, toen ‘then’, nu ‘now’). In all cases, we are dealing with the suffix –s(t), sometimes followed by e (schwa). Since the suffix is hard to detect on complementisers like als ‘if’ that have a final /s/, we do not consider inflection on als here.
Despite the fact that in many cases the subject pronoun can be left out (cf. chapter 2, map 41a), it is fairly certain that this involves genuine complementiser agreement and not a clitic pronoun. The same suffix –s(t) occurs on the finite verb in clause-final position, a position that is never followed by clitic pronouns (cf. De Haan 1997 and De Schutter 2003). In subject-initial main clauses, –s(t) follows the finite verb too (cf. map 25b). As none of the dialects involved lies in an area with subject doubling, –s(t) can only be analysed as complementiser agreement.
In addition to these clear cases of complementiser agreement, map 25a also shows the distribution of the more doubtful case of the morpheme –de as it occurs in adde ‘if-2Sing’ and dadde ‘that-2Sing’. On the basis of the SAND data, we cannot decide whether this morpheme is agreement, a clitic pronoun or a suffix –d combined with a clitic pronoun e. (cf. De Schutter 1997, 2003; Van Craenenbroeck and Van Koppen 2003). The dialects with this morpheme therefore have not been included in the synthesis maps of this chapter. We present them here as potential cases of complementiser agreement.
1.3.4.3 Correlation complementiser and verb agreement 2 singular (map 25b)
Map 25b compares the presence of a morpheme –s(t) or –de on the complementiser (C in the map key) with the presence of these morphemes on the present tense finite verb in inversion contexts, both for the verb gaan 'go' and the verb leven 'live' (gaa–st do, leef–st do, ga–de gij, leef–de gij). We see that the correlation is almost perfect for the suffix –s(t). The following generalisation holds for all measuring points in Friesland, Groningen, and Dutch Limburg: if a dialect has the complementiser agreement suffix –s(t), then it also has this suffix on the finite verb, regardless of whether this verb is gaan 'go' or leven 'live'. Outside this area there are a number of dialects that favour the stricter hypothesis that only the suffix on monosyllabic verbs is relevant (cf. Goeman 1979). The dialects of Rossum, Noord-Deurningen and Losser (Twente, Overijssel), Beekbergen (Gelderland) and Hasselt (Belgian Limburg) have –s(t) on ga but not on leev. In these dialects leev does not have a suffix in inversion contexts.
The dialects with –de on the complementiser all have –de on the finite verb in inversion contexts. Map 25b also clearly shows that there is an area where –de occurs on the finite verb in inversion contexts but not on the complementiser. This area consists of eastern and central Noord-Brabant, the province of Antwerpen and neighbouring parts of Oost-Vlaanderen and Vlaams-Brabant.
Finally, it is noteworthy that many dialects in Oost-Vlaanderen and Vlaams-Brabant have –de only after leev but not after ga or complementisers.
1.3.5 Complementiser agreement third person singular
1.3.5.1 Complementiser agreement 3 singular; synthesis (map 26a)
Map 26a provides an overview of the locations that have 3 person singular complementiser agreement. We restrict ourselves to masculine subject pronouns, since the SAND data do not allow a reliable description of the distribution of complementiser agreement preceding feminine and neuter subject pronouns. Complementiser agreement 3 singular masculine occurs in sentences like:
(21) a. Je gelooft toch niet dat hij sterker is dan jij?
you believe yet not that he stronger is than you
'You don't believe that he is stronger than you, do you?'
b. Als hij nog drie jaar leeft leeft hij langer als zijn vader.
if he still three years lives lives he longer as his father
'If he lives another three years, he will have lived longer than his father.'
In all cases the suffix is –t. It is difficult to determine whether we are really dealing with complementiser agreement here. Alternative analayses are: (i) the complementiser underlyingly has a final /t/ that is deleted before consonants and hence can only be heard before vowels; (ii) /t/ is the first sound of the subject pronoun; (iii) /t/ is inserted to resolve an illicit sequence of two vowels.
For this map, we used the following strategy. If, in a dialect, 3 singular is the only member of the paradigm with /t/ as the final consonant of the complementiser and the dialect has no weak or strong 3 singular masculine pronouns starting with /t/ in regular order, then we assume that /t/ is complementiser agreement. If /t/ were to show up between two vowels we would expect (d)a–t–ek in 1 singular where we get (d)a–k. Moreover, in many dialects /t/ does not show up before a name that starts with a vowel (da Eddy, not da–t Eddy; cf. map 14a). Since we only include dialects with initial pronouns not starting with /t/, the chance that /t/ is the initial consonant of the pronoun is small. A complication is that many of the dialects that potentially have 3 person singular inflection may have /t/ in inversion contexts after is 'is', zou 'would' and past tense verbs: is-t-ij, zou-t-ij and leefde-t-ij (cf. map 43b). This /t/ does not show up in regular order (except with zou), which may indicate that /t/ is not inflection here, unless we are dealing with so-called double agreement dialects, a phenomenon that does exist in the Dutch language area (cf. section 1.2). By way of precaution, we only use the als-sentences (21b) for this map, since als ‘if’ almost certainly does not have an underlying final /t/ (cf. map 15a for different forms of als; Friesland has at, but this form shows up with other members of the paradigm too). Notice in passing that the area with als +–t (3 singular) is a proper subpart of the area with da +–t (3 singular) (cf. map 27a).
Map 26a shows that we find 3 singular inflection in areas that also have complementiser inflection in other members of the paradigm. The large majority of the measuring points in Vlaanderen have 3 singular agreement. This southwestern core area extends to the northwestern part of the province of Antwerpen, western Noord-Brabant and a small part of Zeeland. Other areas with agreement are found in Belgian Limburg, Dutch Limburg and the eastern part of Overijssel and Gelderland.
1.3.5.2 Complementiser agreement 3 singular (map 27a) (map in dynaSAND)
Since there is no variation of form in 3 singular masculine, map 27a only shows the difference between inflection on als ‘if’ and on dat ‘that’. As 27a shows, the area of dialects with 3 singular complementiser inflection on –t may be larger than the area depicted on map 26a. This is because the area with als+–t is part of the area with da+–t, i.e. the area where all members of the paradigm have da except 3 singular which has da+–t and where da probably does not have a final /t/ underlyingly (cf. section 1.3.1.1 and 1.3.5.1).
Southern locations with agreement outside the southwestern core area include Ossendrecht and Steenbergen in Noord-Brabant, Goes and Scherpenisse in Zeeland, Borgloon and Tongeren in Belgian Limburg and Kerkrade, Nieuwenhagen and Heerlen in Dutch Limburg. In the eastern part of the language area, we find agreement in Enter and Vriezenveen (Overijssel) and Groenlo and Eibergen (Gelderland). An isolated case is Roswinkel (Drenthe), on the border with Groningen.
1.3.6 Complementiser agreement first person plural
1.3.6.1 Complementiser agreement 1 plural; synthesis (map 28a)
Complementiser agreement 1 plural appears in sentences like (22).
(22) a. Als we sober leven, leven we gelukkig.
if we frugally live live we happily
'If we live frugally, we live happily.'
b. Ze geloven dat wij rijker zijn dan zij.
they believe that we richer are than they
‘They believe that we are richer than they are.’
The doubtful cases a-me and a-men have not been included in this map (cf. map 29a). We find complementiser agreement 1 plural in three core areas. In the southwest, there is an area consisting of Oost-Vlaanderen (excluding the northwestern part), Zeeuws-Vlaanderen and the north of West-Vlaanderen. There are some attestations in other locations in West-Vlaanderen too. The second core area consists of a large part of Zuid-Holland and Noord-Holland above the Noordzeechannel. The third core area involves a large part of Overijssel and northeastern Gelderland. Finally, there is a small area in southern Dutch Limburg which has agreement.
1.3.6.2 Complementiser agreement 1 plural (map 29a) (map in dynaSAND)
Map 29a provides all different forms of 1 plural complementiser agreement on als ‘if’ and dat ‘that’. We distinguish between forms that are almost certainly agreement suffixes and more doubtful cases. Agreement suffixes are: –e (schwa, as in as–e we 'if-1plur we', dat–e we en dad–e we 'that-1plur we'), –n (as in a–n we 'if-1plur we' and da–n we 'that-1plur we') and –(n)t (as in wen–(n)t vir 'if-1plur we'). Potential agreement suffixes are –me and –men in (d)a–me wij 'that/if-agr/pron we', (d)a–men wij 'that/if-agr/pron we', but these may also be enclitic pronouns (cf. sections 1.3.9, 2.3.5.2.2 and 3.1.3.2). Forms like damme (henceforth da-me) and amme (henceforth a-me) without a following subject pronoun may also be the result of assimilation on the basis of da–n we 'that-1plur we' and a–n we 'if-1plur we'. Such dialects could then be rated among the dialects with agreement suffix –n.
Schwa inflection is typical for the Netherlands. We find it in Zuid-Holland and Noord-Holland, Gelderland and Overijssel.
The suffix –n on da as in da–n we 'that-1plur we' is only attested in West-Vlaanderen and Oost-Vlaanderen, in ten locations, and in one location in the province of Antwerpen, Sint-Lenaarts. In none of these locations do we find a–n we 'if-1sing we', but almost all of them have a-me 'if-agr/pron' and da-me 'that-agr/pron'. This supports the plausibility of the assimilation analysis described above. In the entire southwestern area, there are numerous dialects with a-me 'if-agr/pron’ and da-me 'that-agr/pron' but no an we 'if-1plur we' or dan we 'that-1plur we'. It therefore does not seem too bold to suppose that this area, consisting of Vlaanderen and the western part of the province of Antwerpen, is a massive 1 plural complementiser agreement area.
The form a-men 'if-agr/pron' occurs fourteen times: eleven times in Oost-Vlaanderen and three times in Zeeland. Three of these dialects in Oost-Vlaanderen have an we 'if-1plur we' and one has dan we 'that-1plur we'. For these instances as well, it is not unlikely that da-men is the result of assimilation.
Finally, the form went 'if-1plur' occurs twice, in Kerkrade and Nieuwenhagen (Dutch Limburg). This form can be analysed as wen+–t 'if +-t' or wen+–nt 'if +–nt'. Which of the two analyses is the correct one cannot be decided on the basis of the SAND data. If the generalisation is correct that the suffix on the complementiser is identical to the suffix on monosyllabic verbs in inversion contexts, then the correct analysis is wen+–nt 'if +–nt' (cf. 1.3.6.3). However, we have seen in section 1.3.3.3 (map 23b) that this generalisation does not hold for all dialects.
1.3.6.3 Correlation complementiser and verb agreement 1 plural (map 29b)
Map 29b compares the complementiser agreement suffixes (C in the map key stands for complementiser) with the verbal agreement suffixes. For complementiser agreement on –e (schwa), which is found in the Netherlands exclusively, the map does not lead to a clear conclusion. The form ga–ne we 'go-1plur we' is attested in Friesland, where we don't find 1 plural complementiser agrement. Possibly, the relevant form is possibly ga+–e. Presumably –e inflection on a word that ends in a vowel is excluded (cf. de lila–(*e) jurk 'the lilac-(agr) dress' vs. de paars–*(e) jurk 'the purple-(agr) dress'). These Dutch dialects, almost without exception, have a schwa on leev (lev–e we 'live-1plur we') and, in the majority of cases, gaan we 'go-1plur we'.
The southwestern dialects with an –n suffix on the complementiser all have ga+–n (1 plural) in inversion contexts, which conforms to the generalisation that the complementiser agreement suffix is identical to the suffix on monosyllabic verbs in inversion contexts. Many of these dialects also have–n after leev 'live', or they have men. None of the dialects with men following C have men on ga 'go', and only two of them have men on leev. If the generalisation just mentioned is correct, then this shows that men is not complementiser agreement. This obviously does not exclude the possibility that the initial consonant of men is the result of assimilation of /n/ and /w/. As for the verbal suffixes the map shows that the great majority of the dialects have –e on leev 'live' and –n on ga 'go'. Dialects with a different behaviour are found in West-Vlaanderen, Groningen and Drenthe, where a nasal occurs on both types of verb. The suffix –(n)t is found on ga 'go' in Nieuwenhagen, but not on leev. In the second dialect with –(n)t on the complementiser (Kerkrade), this suffix neither occurs on leev nor on ga.
1.3.7 Complementiser agreement second person plural
1.3.7.1 Complementiser agreement 2 plural; synthesis (map 30a)
Complementiser agreement for 2 plural occurs in the following kind of sentences:
(23) a. Als jullie zo losbandig leven dan leven jullie nooit zo lang als ik.
if you so fast live then live you never so long as I
'If you are living so fast, you will not be living as long as I do.'
2 plural complementiser agreement is extremely rare. Map 30a shows that it occurs in southern Zeeland, Zuid-Holland and the south of Dutch Limburg. Strikingly, some core complementiser-agreement areas, such as the southwestern part of Belgium and the central eastern part of the Netherlands, have no complementiser agreement for 2 plural. A possible explanation for the rareness of this phenomenon is that 2 plural does not always seem to be a real plural in the varieties of Dutch (cf. section 1.3.2.3 for arguments). It is also possible that we are dealing with zero inflection here. In Vlaanderen and Brabant, the finite verb does not have audible inflection in inversion contexts.
1.3.7.2 Complementiser agreement 2 plural (map 31a) (map in dynaSAND)
Map 31a distinguishes agreement on als 'if' and dat 'that' and shows which suffixes can express 2 plural. Four dialects in Zeeland have an –n suffix. It occurs on als 'if' in Oostburg, Hoek and Hulst (Zeeuws-Vlaanderen) and on dat 'that' in Oostburg and Goes. Three dialects have a–e suffix (schwa). It occurs on als 'if' in Katwijk and Sliedrecht (Zuid-Holland) and in Hulst. In Katwijk, –e also occurs on dat 'that'. Notice that Hulst has both –n and –e as a complementiser agreement suffix. Finally, in the south of Dutch Limburg we find three dialects, Heerlen, Nieuwenhagen and Kerkrade, with the agreement suffix –(n)t (cf. section 1.3.6.2 and 1.3.6.3 for the question as to whether the suffix is –t or –nt). This shows up in the form wentir. We assume that this form must be analysed as wen+–(n)t+dir 'if-2plur-you'. The morphemes wen and dir occur in other positions independently. Devoicing of /d/ should then be the result of assimilation with the (final) consonant of the suffix -(n)t.
1.3.7.3 Correlation complementiser and verb agreement 2 plural (map 31b)
Map 31b shows in which cases the suffix on the complementiser (C in the map key) is identical to the suffix on the finite verb in inversion contexts. The reason that the measuring points in West-Vlaanderen are almost devoid of any symbols is that only audible suffixes have been included. For example, most measuring points in West-Vlaanderen have ga 'go' as the form of the finite verb in inversion contexts, without any suffix.
The four dialects in Zeeland with a complementiser agreement suffix –n also have this suffix on ga 'go' but not on leev 'live'. These dialects conform to the generalisation that the suffix on the complementiser is identical to the suffix on monosyllabic verbs, not on verbs in general. However, the two dialects in Zuid-Holland with complementiser agreement –e do not obey the generalisation, as they have ga+–n. Only one of the three Limburgian dialects with -(n)t on the complementiser has ga+–nt, the others do not have a suffix either on ga 'go' or leev 'live'.
1.3.8 Complementiser agreement third person plural
1.3.8.1 Complementiser agreement 3 plural; synthesis (map 32a)
Complementiser agreement for 3 plural occurs in sentences like the following:
(24) a. Jullie geloven toch niet dat zij armer zijn dan jullie?
you believe yet not that they poorer are than you
'You don't believe that they are poorer than you, do you?'
b. Als ze voor hun werk leven, dan leven ze niet voor hun
if they for their work live then live they not for their
kinderen.
children.
'If they live for their work, they don't live for their children.'
Map 32a provides an overview of all dialects with 3 plural complementiser agreement in a position preceding a pronoun, the expletive pronoun er and nominal subjects (cf. also maps 33a,b, 34a). Again it is Vlaanderen that has complementiser agreement almost without exception. In the Netherlands, many places in the west (Zeeland, Zuid-Holland and Noord-Holland) and in Overijssel and eastern Gelderland have complementiser agreement. We also find two places with complementiser agreement in the south of Dutch Limburg.
1.3.8.2 Complementiser agreement 3 plural (map 33a) (map in dynaSAND)
Map 33a shows the distribution of the various suffixes that express 3 plural complementiser agreement and distinguishes between agreement on als 'if' and on dat 'that'. In the southwestern complementiser agreement area, all dialects except two have –n both on dat and on als. In the west of the Netherlands, there are two isolated dialects with –n on dat, namely Strijen and Lopik. The other Hollandic complementiser agreement dialects and three eastern dialects (Epe, Doesburg and Kilder) have –e, but this is only found on als 'if'. Five dialects in the eastern part of the Netherlands have a –t on als 'if': Brummen, Aalten, Groenlo, Eibergen, and Enter. The latter three dialects also have –t on the complementiser in 3 singular (map 27a). Nieuwenhagen and Heerlen (Dutch Limburg) have -(n)t on wen 'if' both in 3 singular and 3 plural.
1.3.8.3 Complementiser agreement 3 plural, non-pronominal subject (map 33b) (map in dynaSAND)
Map 33b presents the dialects that have complementiser agreement before a non-pronominal subject, which, in this case, happen to be two coordinated proper names.
(25) a. Hij gelooft dat Louis en Jan sterker zijn als Geert
he believes that Louis and John stronger are than Geert
en Peter.
and Peter.
'He believes that Louis and John are stronger than Geert and Peter.'
b. Als Pieteren Liesje in het paradijs leven, dan leven
if Pieter en Liesje in the paradise live than live
Rosa en Frans in de hel.
Rosa and Frans in the hell
'If Pieter and Liesje are living in paradise then Rosa and Frans are living in hell.'
There is a large number of dialects in West-Vlaanderen and Oost-Vlaanderen that exhibit agreement in this construction, in most cases on als 'if' (21) but also quite a few with agreement on dat 'that' (14). Six of these dialects have agreement on both types of complementisers. In the east of the Netherlands, it is, once again, the dialects of Aalten, Groenlo and Enter that have an agreement suffix –t.
1.3.8.4 Complementiser agreement 3 plural preceding pronouns, expletives and noun phrases (map 34a) (map in dynaSAND)
There are three environments in which 3 plural complementiser agreement occurs: before subject pronouns (sentence 24b), before expletives (sentence 25b) and before noun phrases (sentence 26). This map shows the correlation between these three environments.
(26) Als er zo weinig mensen van de landbouw leven, dan
if there so few people of the agriculture live then
leven er veel mensen van werk in de fabriek.
live there many people of work in the factory
'If there are so few people making a living from agriculture, then there must be many people making a living from industry.'
The number of dialects with complementiser agreement before subject pronouns is the largest, 44. They are all situated in the southwestern part of the language area. Eight of these dialects do not show agreement in the other two environments. Seven of these eight dialects are spoken in Zeeuws-Vlaanderen and Oost-Vlaanderen, the eighth is spoken in Frans-Vlaanderen (Waals-Kappel). There are 40 dialects with agreement before expletive er 'there'; 36 of them also have agreement before subject pronouns. There are four dialects which only have agreement on als 'if' before er 'there': Steenvoorde (Frans-Vlaanderen), Roeselare and Moorsele (West-Vlaanderen) and, far removed from the southwestern core area, Lopik (Utrecht). Notice that these four dialects do have –n on da 'that' (cf. map 33a). The number of dialects with agreement before a noun phrase (21) is half as small as the number of dialects with agreement before er 'there'. With the exception of Nieuwpoort (West-Vlaanderen), all of these dialects also have agreement in the other two environments. Putting aside this exception, we can conclude that dialects with agreement before non-pronominal subjects always have agreement before pronominal subjects, but not vice versa.
Map 34a only shows dialects with an –n suffix. In the eastern part of the Netherlands, we find a couple of dialects with –t before expletive subjects.
1.3.8.5 Correlation complementiser and verb agreement 3 plural (map 34b)
Map 34b shows in which cases the 3 plural suffix on the complementiser is identical to the suffix on ga- 'go' and leev- 'live' in present tense inversion contexts. For the suffix –n, this is indeed the case. Dialects that have this complementiser agreement suffix are spoken in the southwestern part of the language area. The great majority of these dialects also have–n on the two verbs. Only in the east of Oost-Vlaanderen do we find a number of dialects for which this correlation does not hold (the data for Frans-Vlaanderen are incomplete and therefore not discussed here). The dialects with suffix –e on the complementiser, all located in in the Netherlands, also have–e on leev- 'live', but not on ga- 'go'. However, as was noted before, this does not necessarily forces us to reject the generalisation that the complementiser agreement suffix is identical to the suffix on ga 'go' and leev 'live' in present tense inversion contexts, if there is an idosyncratic restriction on –e after vowels (cf. 1.3.6.3, map 29b). Most of the dialects with complementiser agreement on –t, in the east of the Netherlands, conform nicely to the generalisation. Two of these dialects (Losser and Doesburg), don't have–t on ga 'go' and leev 'live'. Finally, Heerlen and Nieuwenhagen (Dutch Limburg) with -(n)t demonstrate once more that the generalisation applies to monosyllabic verbs, as both these dialects have ga+–nt, but leev+–e.
As regards verbal agreement, the map shows some interesting regional differences. Dialects in Groningen and on the periphery of Drenthe have –n on both verbs. Dialects in central Drenthe, Overijssel and eastern Gelderland usually have–t. The southwestern part of the language area uniformly has –n, and the rest of the dialects, in fact, the large majority, have–n on ga and –e on leev.
1.3.9 Complementiser agreement –n in plural
As was noted in section 1.3.6.2 it is unclear whether the 1 plural forms da+me 'that-agr/pron’ and da+men 'that-agr/pron' contain a suffix, an enclitic pronoun or both. The form da+me could be the result of place assimilation of a plural suffix /n/ to /m/ in the sequence da+–n+me. The form dammen may be analysed in the same way, although it has been claimed that the final /n/ is agreement here. In this section, we check if there is a correlation between having damme or dammen on the one hand and having –n in 2 and 3 plural on the other hand. It might be that there are dialects in which plural agreement always has to be marked on the complementiser with –n, just as this is the case for verbs in certain dialects.
1.3.9.1 Complementiser agreement –n in plural; synthesis (map 35a)
Map 35a gives all dialects with a plural form of –n. This suffix is highly frequent for 3 plural complementiser agreement (55 dialects), occurring in the southwestern part of the language area and in two isolated locations in the Netherlands (cf. section 1.3.8.2). 13 of these 55 dialects also have –n in 1 plural. These do not constitute a clearly demarcated area. It should be noted that 1 plural –n does not occur in Frans-Vlaanderen. There are four dialects that have –n in 1 plural but not in 3 plural. They are all located in the periphery of or outside of the southwestern core area: Oostburg (Zeeuws-Vlaanderen), Aalst (Oost-Vlaanderen), Sint Lenaarts (Antwerpen), and Monster (Zuid-Holland). Finally, there are only four dialects with –n in 2 plural. Two of them have –n in all plural forms: Hoek and Hulst (Zeewus-Vlaanderen). Oostburg (Zeeuws-Vlaanderen) has –n in 1 and 2 plural. Goes (Zeeland) has –n in 2 and 3 plural.
1.3.9.2 Correlation (d)amme (1 plural) and –n in 2 and 3 plural (map 35b)
Map 35b answers the question as to which dialects with (d)amme have –n on the complementiser in 2 and 3 plural. It is clear that the dialects with –n in 3 plural are a proper subset of the dialects with (d)amme, putting aside the two attestations in the Netherlands. They can be found in Frans-Vlaanderen and West-Vlaanderen and in northern and western Oost-Vlaanderen. We may assume that these dialects have a generalised plural –n on the complementiser, if we also assume that 2 plural is not a real plural in these dialects. At the eastern edge of the southwestern plural agreement area we find dialects in Oost-Vlaanderen, Vlaams-Brabant and Antwerpen that have (d)amme but no –n in 2 and 3 plural. This shows that (d)amme does not need to be a case of complementiser agreement, unless we are dealing here with dialects that only have–n in 1 plural. For three dialects in Zeeland (Goes, Hoek and Hulst) we can safely assume that (d)amme contains complementiser agreement. As we have seen in the previous section, a–n may show up in 1, 2 and 3 plural in these dialects.
1.3.9.3 Correlation (d)ammen (1 plural) and –n in 2 and 3 plural (map 35c)
As map 35c shows, the form (d)ammen is much less common than (d)amme. It is found in eastern Oost-Vlaanderen and in a number of places in Zeeland. In most places with (d)ammen, there is no plural agreement –n on complementisers. Therefore, this map does not give us any reason to assume dat (d)ammen always contains inflection. In Hoek and Hulst (Zeeuws-Vlaanderen), (d)ammen is an alternant of (d)amme.
1.3.10 Complementiser agreement –e in plural; synthesis (map 36a)
The goal of 1.3.9 was to see if generalised plural complementiser agreement on –n exists. In this section, we ask if there are any dialects in which plural complementiser agreement is always expressed by –e (schwa). After all, there are quite a few dialects (e.g., Standard Dutch) in which plural agreement on polysyllabic finite verbs is always –e. Map 36a shows that generalised plural –e does not exist in the southwestern complementiser agreement area. There are 15 dialects in the Netherlands which have –e in 3 plural. Eleven of these also have–e in 1 plural. We could take these to be dialects with generalised plural –e on the complementiser, if we assume that 2 plural is not really plural. The latter is not implausible given the alternation jullie hebb–en 'you-guys have-plur' and jullie heb–t 'you-guys have-2sing' in Standard Dutch and the fact that 2 plural reflexive and possessive pronouns in Standard Dutch are homophonous with the 2 singular personal, reflexive and possessive pronoun je. Finally, in the east of the Netherlands we find 11 dialects which have only a –e on the complementiser in 1 plural.
1.3.11 Complementiser agreement in comparatives (map 36b) (map in dynaSAND)
Complementiser agreement also shows up in comparatives, for example on als ‘as’ in sentences like (27).
(27) We geloven dat jullie niet zo slim zijn als wij.
we believe that you not as smart are as we
‘We believe that you are not as smart as we are.’
Map 36b provides an overview of all cases of complementiser agreement in comparatives found in the SAND data. In the singular, we only find agreement for the second person, on –s(t) and the forms ofte and dante. It does not come as a surprise that the dialects with –s(t) are located in the 2 singular agreement area described in section 1.3.4. It involves two dialects in Friesland (West-Terschelling and Kollum) and six dialects in Dutch and Belgian Limburg (Grote Spouwen, Eigenbilzen, Gronsveld, Rijckholt, Nieuwenhagen, and Kerkrade). The 2 singular forms ofte/dante occur in five locations in Oost-Vlaanderen (Deinze, Lovendegem, Gent, Oostakker, and Zaffelare). In 1 plural, –e-inflection is found in the form asse 'than-1plur' in four places in Oost-Vlaanderen (Zaffelare, Berlare, Ninove, and Geraardsbergen) and in four places in the east of the Netherlands (Nieuw-Scheemda, Wechterholt, Groenlo and Winterswijk). A potential case of –e inflection is omme, found in Eeklo, Lovendegem and Strijpen (Oost-Vlaanderen), which could be a combination of o+–n+me and assimilation. There are two forms for 2 plural: ofte/dante and went. The first is found in Lovendegem, Gent and Oostakker (Oost-Vlaanderen). The second we find in Kerkrade and Nieuwenhagen (Dutch Limburg). In some dialects, the forms ofte/dante are not restricted to 2 singular and 2 plural. Because they do not distinguish between persons or number in such dialects, we do not take them to be complementiser agreement.
1.4 Literature on complementisers and complementiser agreement
Abraham, W. (1996). 'German Standard Pronouns and Non-Standard pronominal Clitics:
Typological Corollaries.' In R. Lippi-Green & J. Salmons (eds.) Germanic Linguistics. Amsterdam: Benjamins. 1–14.
Ackema, P. & A. Neeleman (2001). 'Context-sensitive spell-out and adjacency.' Ms.
Utrecht University and University College London.
ANS (1997). W. Haeseryn, et al (eds.). Algemene Nederlandse Spraakkunst. Groningen:
Wolters-Noordhoff.
Bakkes, P. (1996). Variatie en verandering in het Montforts. Taalstructurele en
sociolinguïstische aspecten van een veranderend dorpsdialect. Amsterdam: P.J. Meertens-Instituut voor Dialectologie, Volkskunde en Naamkunde.
Bayer, J. (1984). 'Towards an explanation of certain that-t phenomena: The COMP-node
in Bavarian.' In W. de Geest & Y. Putseys (eds.) Sentential complementation. Dordrecht/Cinnaminson: Foris Publications. 23–32.
Beckering Vinckers, J. (1872). 'Fonetische voorbarigheid, een middel ter verklaring van
smijns.' Taal- en Letterbode 3. 165–171.
Bennis, H. (1997). 'Voegwoordvariaties.' In A. van Santen & M. van der Wal (eds.) Taal
in Tijd en Ruimte. SNL. 353–364.
— (2000). 'On the Interpretation of Functional Categories.' In H. Bennis, M.
Everaert & E. Reuland (eds.) Interface Strategies. KNAW Verhandelingen,
afdeling Letterkunde, Nieuwe Reeks, deel 180. Amsterdam: KNAW. 37–52.
Bennis, H. & L. Haegeman (1984). 'On the status of agreement and relative clauses in
West-Flemish.' In W. de Geest & Y. Putseys (eds.) Sentential complementation. Dordrecht/Cinnaminson: Foris Publications. 33–53.
Bouman, A.C. (1918). Bijdrage tot de syntaxis der 'dat'-zinnen in het Germaansch. Diss.
Utrecht.
Boland, E.M. (1959). 'Vervoegde voegwoorden in het Oosten (II).' Driemaandelijkse
bladen NS11. 39.
Carnel, D. (1891). Le dialecte flamand de France. Parijs: Bouillon.
Carstens, V. (2003). 'Rethinking Complementizer Agreement: Agree with a Case-
checked Goal.' Linguistic Inquiry 34. 393–412.
Corbett, G. (1991). Gender. Cambridge: Cambridge University Press.
Corbett, G. (2000). Number. Cambridge: Cambridge University Press.
Cosijn, P.J. (1872). 'Assimilatie in het Nederlandsch.' Taal- en letterbode 3. 270–274.
Daan, J. (1958). 'Assewe – assese.' Taal en Tongval 10. 62.
De Boer, R. (1952). 'Vervoeging van voeg- en voornaamwoorden.' It Beaken. Tydskrift
fan de Fryske Akademy 14. 18–22.
De Haan, G. J. (1994). 'Inflection and Cliticisation in Frisian, –sto, –ste, –st.' North
Western European Language Evolution 23. 75–90.
— (1997). 'Voegwoordcongruentie in het Fries.' In E. Hoekstra & C. Smits
(eds.) Vervoegde voegwoorden. Amsterdam: P.J. Meertensinstituut. 50–67.
Dekkers, J. (1999). Derivations and evaluations. On the syntax of subjects and
complementizers. Diss. University of Amsterdam. LOT Dissertations 21.
De Meersman, A. (1985). 'Een VSO-maneuver in de Zuidnederlandse dialecten.' In
H. Ryckeboer, J. Taeldeman & V.F. Vanacker (eds.) Hulde-album Prof. dr. Marcel Hoebeke. Gent: RUG, Seminarie voor Nederlandse taalkunde. 123–131.
Den Besten, H. (1989). Studies in West Germanic Syntax. Amsterdam/Atlanta: Rodopi.
De Rooij, J. (1965). Als – of – dat. Een semantisch-onomasiologische studie over enkele
subordinerende conjuncties in het ABN, de Nederlandse dialecten en het Fries,
vergelijkend-synchronisch beschouwd. Assen: Van Gorcum & Comp. N.V.
— (1967). 'Lummel dat je bent.' Tijdschrift voor Nederlandse Taal- en
Letterkunde 83. 108–121.
De Schutter, G. (1990). 'De studie van de syntaxis van Nederlandse en Friese dialecten.'
Taal en Tongval Themanummer 3. 10–47.
— (1994). 'Voegwoordflectie en pronominale clitisering waarin Vlaams en
Brabants bijna elkaars tegengestelden zijn.' Taal en Tongval 46. 108–131.
— (1997). 'Incorporatie-in-C in de Vlaamse en Brabantse dialecten.' In E.
Hoekstra & C. Smits (eds.) Vervoegde voegwoorden. Amsterdam: P.J.
Meertensinstituut. 31–49.
— (2003). 'De studie van de Nederlandse en Friese dialectsyntaxis sedert
1990. 'Een bibliografisch overzicht.' Taal en Tongval Themanumer 15–16.
Dialectsyntaxis in bloei. 10–43.
De Visser, M. (1988). 'In pear klitisearringsferskynsels yn it Frysk.' In S. Dyk & G.J. de
Haan (eds.) Wurdfoarried en Wurdgrammatika. Leeuwarden: Fryske Akademy. 175–222.
De Visser, M. & A. Goeman (1979). 'Voegwoord, relatief partikel en persoonsvorm in
een dialect.' Taal en Tongval 31. 222–241.
De Vogelaer, G. A. Neuckermans & G. Vanden Wyngaerd (2002). 'Complementizer
Agreement in the Flemish Dialects.' In S. Barbiers, L. Cornips & S. van der Kleij (eds.) Syntactic Microvariation. Meertens Institute Electronic Publications in Linguistics 2.
De Vogelaer, G. (2003). 'Person marking in Dutch dialects.' In B. Kortmann (ed.)
Dialectology meets typology. Berlijn: Mouton de Gruyter. 181–210.
De Vries, W. (1940). 'Congruerende voegwoorden.' Tijdschrift voor Nederlandse Taal-
en Letterkunde 59. 78–79.
Dyk, S. & J. Hoekstra (1987). 'In oersjoch fan 'e studzje fan 'e Fryske syntaksis.' In
S. Dyk & J. Hoekstra (eds.) Ta de Fryske syntaksis. Leeuwarden: Fryske Akademy. 7–37.
Frank, J. (1898). 'Schriften zur limburgischen Sprache und Litteratur.' Taal en Letteren
8. 49–52, 105–109, 135–138, 337–341, 387–392, 415–421, 463–475, 503–515.
Gerritsen, M. (ed.) (1991). Atlas van de Nederlandse Dialectsyntaxis (AND), Deel I.
Amsterdam: P.J. Meertensinstituut.
Goeman, A. (1979). 'Naschrift' [bij De Visser & Goeman 1979]. Taal en Tongval 31.
240–241.
— (1980a). 'COMP-agreement.' In W. Zonneveld & F. Weerman (eds.)
Linguistics in the Netherlands 1979. Dordrecht: Foris Publications. 291–306.
— (1980b). 'Werkwoordsuitgangen bij voegwoorden in de bijzin.' Onze Taal 49. 35.
— (z.j.). 'Structurele aspecten: mogelijkheden tot en beperkingen op
voegwoordvervoeging.' Ms. Ongepubliceerd.
— (1997a). 'Historiografie van het onderzoek naar voegwoordvervoeging: een
bibliografisch overzicht (1821–1996).' In E. Hoekstra & C. Smits (eds.) Vervoegde voegwoorden. Amsterdam: P.J. Meertensinstituut. 112–145.
— (1997b). 'De zeldzaamheid van COMP-agreement.' In E. Hoekstra & C.
Smits (eds.) Vervoegde voegwoorden. Amsterdam: P.J. Meertensinstituut. 87–111.
— (1999). T-deletie in de Nederlandse dialecten. Kwantitatieve analyse van
structurele, ruimtelijke en temporele variatie. Diss. VU Amsterdam. LOT Dissertations 26.
— 'Structurele aspecten van de morfologie van voegwoorden: mogelijkheden en
beperkingen, morfologisch gewicht en MCGG.' In Véronique de Tier, Magda
Devos & Jacques van Keymeulen (eds.) Nochtans was scherp van zin. Huldealbum Hugo Ryckeboer. Gent: Vakgroep Nederlandse taalkunde.
Haegeman, L. (1990). 'Subject Pronouns and Subject Clitics in West Flemish.' Linguistic
Review 7. 333–363.
— (1992). Theory and description in generative syntax. A case study in West
Flemish. Cambridge: Cambridge University Press.
Hoekema, T. (1955). 'Konjugaesje, kumulphoby of folksetymologyske analyse fan
enklisefoarmingen?.' It Beaken 17. 13–18.
Hoeksema, J. (1999). 'Recensie van 'Vervoegde Voegwoorden' van E. Hoekstra & C.
Smits.' Driemaandelijkse bladen 51. 28–34.
Hoekstra, E. (1993). 'Some implications of number agreement on COMP.' In F.
Drijkoningen & K. Hengeveld (eds.) Linguistics in the Netherlands 1993. Amsterdam: Benjamins. 61–68.
— (1993a). 'On the Parametrisation of Functional Projections in CP.' In A.J.
Schafer (ed.) Proceedings of NELS. Amherst: GLSA. 191–204.
— (1993b). 'Dialectal variation inside CP as Parametric Variation.' In W. Abraham & J. Bayer (eds.) Dialektsyntax. Opladen: Westdeutscher Verlag. 161–179.
Hoekstra, E. & C. Smits (1997). 'Vervoegde voegwoorden in de Nederlandse dialecten:
een aantal generalisaties.' In E. Hoekstra & C. Smits (eds.) Vervoegde voegwoorden. Amsterdam: P.J. Meertensinstituut. 6–30.
— (1999). Everything you always Wanted to Know about Complementizer Agreement. Ms. Meertensinstituut. Ongepubliceerd.
Hoekstra, E. & J.-W. Zwart (1994). 'De structuur van CP. Functionele Projecties voor
Topics en Vraagwoorden in het Nederlands.' Spektator 23. 191–212.
Hoekstra, J. (1997). 'Pro-drop, clitisering en voegwoordcongruentie in het
Westgermaans.' In E. Hoekstra & C. Smits (eds.) Vervoegde voegwoorden. Amsterdam: P.J. Meertensinstituut. 68–86.
Hoekstra, J. & L. Maracz (1989). 'On the Position of Inflection in West-Germanic.'
Working Papers in Scandinavian Syntax 44. 75–88.
— (1989). 'Some implications of T-to-C-movement in Frisian.' In H. Bennis & A. van Kemenade (eds.) Linguistics in the Netherlands. Dordrecht: Foris. 81–90.
Klatter, J. (1933). 'Dialectstudie en Syntaxis.' Onze Taaltuin 2. 73–84.
Koelmans, L. (1968). 'Vervoegde voegwoorden in het dialect van Bruinisse.' De Nieuwe
Taalgids 61. 195.
Paardekooper, P.C. (1961). 'Persoonsvorm en voegwoord.' De Nieuwe Taalgids 54.
296–301.
— (1993). 'En(de)/inan 'indien' enz.: een uniek uitstervingsproces.'
Taal en Tongval 45. 50–79, 115–134.
Pannekeet, J. (1995). Het Westfries: inventarisatie van dialectkenmerken. Wormerveer:
Noord-Holland.
Pauwels, J.L (1958). Het dialect van Aarschot en omstreken. Belgisch Universitair
Centrum voor Neerlandistiek.
Platzack, C. (1992). 'Complementizer agreement and argument clitics.' Working
Papers in Scandinavian Syntax. 25–54.
Roukens, W. (1947). De taal der Limburgers. Nijmegen: De Koepel.
Ryckeboer, H. (1983). 'Voor te + infinitief. Verkenning van de dynamiek van een
dialectisme.' Taal en Tongval 35. 83–90.
Siewierska, A. (2004). Person. Cambridge: Cambridge University Press.
Simons, L. (1889). Het Roermondsch dialect. Gent: Koninklijke Vlaamse Academie.
Stoett, F.A. (1923). Middelnederlandsche spraakkunst; Syntaxis. Den Haag: Martinus
Nijhoff.
Sturm, A. (1996). 'Over functionele projecties.' Nederlandse Taalkunde 1. 191–207.
Teirlinck, Is. (1924). Klank- en vormleer van het Zuid-Oostvlaandersch dialect. Gent:
Koninklijke Vlaamse Academie.
Tiersma, P.M. (1999). Frisian Reference Grammar. Leeuwarden: Fryske Akademy.
Vallaeys, A. (1997). Spraakkunst van het Poperings. Poperinge: Stad Poperinge.
Vanacker, F. (1949). 'Over enkele meervoudsvormen van voegwoorden.' Taal en
Tongval 1. 32–45, 77–93, 108–112.
— (1970). 'Een paar punten uit een vergelijkend syntaxisonderzoek van
enkele Zuidnederlandse dialecten.' In Zijn akker is de taal. Faculteitenreeks 11.
Den Haag: Bert Bakker. 291–303.
Van Craenenbroeck, J. (2000). 'Complementerend van. Een voorbeeld van syntactische
variatie in het Nederlands.' Nederlandse Taalkunde 5. 133–163.
— (2004). Ellipsis in Dutch dialects. Diss. Leiden University. LOT Dissertations 96.
Van Craenenbroeck, J. & M. van Koppen (2003). 'Congruentie en lokaliteit in de
Nederlandse dialecten.' Taal en Tongval. Themanummer 15: Dialectsyntaxis in bloei. 63–82.
Van den Hoek, Th. (1971). 'Opmerkingen over zinscomplementatie.' Studia neerlandica
7. 189–215.
Van den Toorn, M.C. et al. (1997). Geschiedenis van de Nederlandse Taal. Amsterdam:
Amsterdam University Press.
Van der Horst, J.M. (1990). 'Is weglating van het voegwoord dat typisch voor 19e-
eeuwse vrouwen?' In J.B. den Besten et al. (eds.) Vragende wijs: vragen over tekst, taal en taalgeschiedenis: bundel aangeboden aan Leopold Peeters bij zijn afscheid als Hoogleraar Historische Taalkunde van het Nederlands aan de Universiteit van Amsterdam. Amsterdam: Rodopi. 95–103.
Van der Meer, G. (1991). 'The 'conjugation' of subclause introducers: Frisian –st.' North-
Western European Language Evolution 17. 63–84.
Van Es, G.A. (1950). 'Dialectologie en syntaxis.' In anoniem (eds.) Album aangeboden
aan Prof. Dr. L. Grootaers. Leuven: St Alfonsdrukkerij. 205–225.
Van Ginneken, J. (1938). 'De vervoeging der onderschikkende voegwoorden en
voornaamwoorden.' Onze Taaltuin 8. 1–11.
— (1939). 'De vervoegde voegwoorden en de Slavische conditionals.' Onze
Taaltuin 8. 33–41.
— (1954). Drie Waterlandse dialecten. Dl. I: Grammatica, phonologie,
klankleer. bez. door A. Weynen. Alphen a/d Rijn: Samson.
Van Haeringen, C.B. (1939). 'Congruerende voegwoorden.' Tijdschrift voor Nederlandse
taal- en letterkunde 58. 161–176.
— (1958). 'Vervoegde voegwoorden in het Oosten.' Driemaandelijkse bladen NS 10. 115–124.
Van Helten, W. (1882). 'Bijdragen tot de Dietsche grammatica II. Een tweede persoon
enkelvoud op –tst(e) of –ts.' Tijdschrift voor Nederlandse Taal- en Letterkunde 2. 49–51.
Van Koppen, M. (2005). One probe, two goals. Aspects of Agreement in Dutch dialects.
Diss. Leiden University. LOT Dissertations.
Verwijs, E. & J. Verdam (1889). Middelnederlandsch woordenboek (deel 2). Den Haag:
Martinus Nijhoff.
Weerman, F. (1989). The V2 Conspiracy. A synchronic and diachroninc analysis of
verbal positions in Germanic languages. Dordrecht/Providence: Foris.
Weijnen, A. (1940). 'Congruerende verbindingswoorden.' Tijdschrift voor Nederlandse
Taal- en Letterkunde 59. 75–77.
— (1966). Nederlandse dialectkunde. Assen: Van Gorcum & Comp N.V.
— (1971). Schets van de geschiedenis van de Nederlandse syntaxis. Assen:
Van Gorcum & Comp. N.V.
Zwart, C. J. W. (1993). Dutch syntax. A minimalist approach. Diss. Groningen
University: Groningen Dissertations in Linguistics.
— (1993). 'Clues from dialect syntax: complementizer agreement.' In W.
Abraham & J. Bayer (eds.) Dialektsyntax. Opladen: Westdeutscher Verlag. 246–
270.