Everything you always Wanted to Know about Complementizer Agreement*
Eric Hoekstra & Caroline Smits
Fryske Akademy, Ljouwert and Meertens Institute, Amsterdam

[preliminary version, final version in Proceedings of WECOL 1998]

1. Summary
This paper deals with characteristics of complementizer agreement in Dutch dialects (i.e. dialects spoken in the Netherlands and in the Dutch speaking region of Belgium).1 In section 2, we will discuss a number of generalizations that have been made in relation to the phenomenon. In section 3 we will discuss a number of unsolved questions in relation to complementizer agreement:
(i) Why do some dialects have complementizer agreement, others not?
(ii) Why does complementizer agreement generally have a defective paradigm?
We propose (the beginning of) a solution to the second problem in section 4. In section 5, we will further modify the initial proposal.

2. Complementiser Agreement: Some Generalizations
2.1. What is complementizer agreement?
Complementizer agreement involves the phenomenon by which the agreement ending which appears on the verb also shows up on the complementizer, intro-ducing a subordinate clause.2 As a result, there is not only agreement between the subject and the verb as far as person and number are concerned, but between the subject and the complementizer as well (cf. also De Haan 1997). Although this phenomenon is highly uncommon in the languages of the world, it is quite popular in German and Dutch dialects. Both German and Dutch are Verb-Second languages, and this seems to be a necessary condition for complementizer agree-ment to occur. As said, the present paper only discusses the phenomenon in the Dutch dialects. Consider the following examples of complementizer agreement:

(1) Limburg:
veurtot-s tiech de bruk zuu-s
before-2SG you the bridge see-2SG
 

(2) South Holland:
datt-e ze ziek benn-e
that-3PL they ill are-3PL

2.2. The copy generalization: the agreement ending on the verb in the sentence in question is copied onto the complementizer.
In early studies dealing with complementizer agreement, the stand is taken that the agreement ending of the verb in the sentence in question is simply copied onto the complementizer. This view is reflected by Beckering Vinckers (1872), for instance, who, in an attempt to explain the phenomenon, claimed that the so-called 'conjugated conjunctions' are actually slips of the tongue; a speaker who is uttering a sentence may anticipate the end of that sentence that is the finite verb , and may thus prematurely add the agreement ending of the verb to the conjunction. Beckering Vinckers' explanation of the phenomenon cannot be correct, however, since it completely ignores the systematic nature of the phenomenon: it is only agreement endings that re-appear (and never the markers for tense); these endings only occur on the complementizer (and never on other parts of the sentence, for instance on the subject or the object); moreover, and as will be discussed in greater detail below, complementizer agreement is generally limited to specific forms within the paradigm. To give two examples of the latter: in Frisian, complementizer agreement only occurs in clauses with 2SG subjects; in Limburg dialects, it only occurs in 2SG and 2PL.
Similarly, Van Haeringen (1939) explains the phenomenon by stating that it occurs in an attempt to overcome the distance between subject and finite verb, a distance which can be quite long in subordinate clauses in Dutch. By copying the agreement ending of the clause-final verb onto the clause-initial complementizer, the tension between subject and finite verb is somewhat relieved. Again, this view reflects the idea that the agreement ending of the finite verb in the sentence in question is simply copied onto the complementizer, meaning that the agreement endings occurring on the verb and on the complementizer should be identical. Cf. the following examples which seem to be in harmony with this view ('MP' stands for 'modal particle'):3

(3) North Holland:
azz-e je morgenavond even ankomm-e
when-2SG you tomorrow evening MP come-2SG over

(4) North Holland/South Holland:
toen-e me kwamm-e
when-1PL we came-1PL

(5) North Holland/South Holland:
(ze zegge) datt-e ze ziek benn-e
(they say) that-3PL they sick are-3PL

(6) Flanders (Belgium):
(et werk) da-n de kinders gemaakt e-n
(the work) that-3PL the children made have-3PL

(7) Groningen:
(ik wait nait) of-s toe kom-s
(I know not) if-2SG you (2SG) come-2SG

(8) Groningen:
az-n koin nait in et laand blievm will-n
when-3PL cows (3PL) not in the land stay want-3PL

(9) Overijssel (Enschede):
(ik weet nig) of-s toe kom-s
(I know not) if-2SG you (2SG) come-2SG

(10) Limburg:
(iech waet neet) boe-t ger zee-t
(I know not) where-2PL you (2PL) are-2PL

2.3. The inversion generalization: the ending on the complemen-tizer is identical to the agreement ending on the verb in inver-sion
As becomes clear from the endings on the complementizer in dialects spoken in the eastern parts of the Netherlands, however, the copy generalization, discussed above, cannot be correct. It concerns a vast area (i.e. large parts of the Provinces of Drente, Overijssel and Gelderland) that have a 123PL ending -t (cf. Hol 1955). Cf. in this light the following examples, taken from Van Haeringen (1958):

(11) ik geleuve datt-e wy et mit hum maar es prebeer-t
I believe that-1PL we it with him MP MP try-1PL

(12) azz-e wy de turf niet verkoopn kun-t
if-1PL we the peat not sell can-1PL

In the eastern dialects in question, complementizer agreement only occurs in 1PL. Obviously, in these dialects the agreement ending of the verb, ending in -t , is not identical to the ending on the complementizer, which ends in -e . This means that complementizer agreement does not simply entail the phenomenon by which the agreement ending of the verb in the sentence in question is copied onto the complementizer. As Van Haeringen (1958) rightly noted, it is actually the agreement ending of the verb in inversion (which is the verb form in absentia) which is added to the complementizer. Naturally, this can only be observed by considering dialects in which the agreement ending of the verb in inverted order is different from the ending in non-inverted order. The eastern dialects from which the examples in (11) and (12) are taken represent such dialects. In these dialects the 1PL verb ending in non-inverted order is different from the 1PL verb form in inverted order; in inverted order it ends in -e , and in non-inverted order it ends in -t . Cf. (13):

(13) a. wy speul-t
we play-1PL
b. speul-e wy
play-1PL we

In short, the ending which is added to the complementizer is identical to the agreement ending of the finite verb in inversion. For the eastern dialects under discussion this means that in clauses with a 1PL subject the complementizer ends in -e , hence datt-e /*dat- t and azz-e /*as-t .
 

3. Unsolved mysteries in relation to complementizer agreement
The above inversion generalization does not exhaustively characterize complementizer agreement, however. Actually, there are quite a number of un-solved mysteries in relation to the phenomenon. One of these mysteries concerns the geographical distribution of complemetizer agreement within the Dutch speaking area, another concerns the distribution within the paradigm of verbal endings.

3.1. Defective geographical distribution
Remarkably enough, complementizer agreement does not occur within all regions or dialects within the Dutch area (i.e. the Netherlands and Dutch speaking Belgium). Put differently, although Verb-Second seems to be a necessary condition for complementizer agreement to take place, and although all Dutch dialects meet this condition, the phenomenon does not occur throughout the Dutch speaking area.4
First, we should distinguish between a western part and an eastern part where complementizer agreement does occur, and an area in between where the phenomenon is absent. The latter area involves the Dutch Provinces of Utrecht, North Brabant, and a large part of the Province of Gelderland, and the Belgian Provinces of Antwerp and Brabant.
Second, the western and the eastern part where complementizer agreement does occur can be further subdivided. As far as the western part of the area is concerned, a northern part and a southern part can be distinguished where complementizer agreement does occur. The northern part involves the dialects spoken in the Provinces of North and South Holland (Van Haeringen 1939). The southern part involves the isles of Zuid-Beveland (Hoekstra 1993) and Zeeuws-Vlaanderen (De Visser & Goeman 1979) in the Province of Zeeland, the Flemish dialects in the Province of French Flanders, West Flanders, and the eastern part of East Flanders (Vanacker 1949; De Schutter 1997). Between the northern and the southern part, however, there is a region where complementizer agreement is absent. This involves the isles of Goeree-Overflakkee (in the Province of South Holland), and Schouwen-Duiveland and Noord-Beveland (in the Province of Zeeland).
Similarly, the eastern part of the Dutch speaking area can be divided into two areas where complementizer agreement does occur, and an area in between where it is absent. Specifically, both the northern and the southern part do have complementizer agreement: the northern part concerns the Provinces of Friesland, Groningen, and the eastern parts of the Provinces of Drenthe, Overijssel and Gelderland (Beckering Vinckers 1872; Klatter 1933; Van Ginneken 1939; Van Haeringen 1958; Van der Meer 1991; De Haan 1997); the southern part concerns the Province of Limburg, particularly the southern region (Van Ginneken 1939). Between these two parts, however, there is, again, an area where complementizer agreement is absent. At present, we do not have a ready explanantion for the remarkable distribution of complementizer agreement throughout the Dutch-speaking area.

3.2 Defective paradigmatic distribution
Perhaps even more remarkable is the fact that in the areas and dialects in which complementizer agreement does occur, it is rarely the full paradigm of verbal endings (123SG and 123PL) which appears on the complementizer. Put differently, complementizer agreement generally has a defective paradigm.5 Moreover, it appears that different dialects exhibit different preferences for specific forms within the paradigm.6 Cf. the following list of complementizer agreement in various Dutch dialects (cf. also the examples under (1)-(12) that were given above):

(14) North Holland: 2SG -e + 123PL -e
South Holland: 123PL -e
Zeeland Flanders: 1SG -n + 123PL -n
Friesland: 2SG -st
Groningen: 2SG -st + 123PL -n
Overijssel: 2SG -s
Eastern dialects: 1PL -e
Limburg 2SG -s + 2PL -t

In sum, we can ask ourselves at least the following two questions:
(i) Is there a system in the defectivity of complementiser agreement paradigms?
(ii) Why do some dialects have complementiser agreement, others not?

3.3 Comparing complementizer agreement with verb agreement
In some sense, complementizer agreement seems to be parasitic on verb agree-ment. We find dialects with verb agreement but without complementizer agree-ment. However, we do not find dialects with complementizer agreement but without verb agreement.
Verb agreement in West-Germanic can be either synthetic or analytic. Consider the t-ending in the following paradigm:

(15) Main verb lopen 'to run'
present past
1S loop liep
2S loopt liep
3S loopt liep
PL lope liepe (official PL spelling: -en)

The -t encodes person/number information, on the one hand, and tense information, on the other. Thus the -t says: you're dealing with a singular form, second or third person, and you're dealing with a present tense form. As the -t encodes both person/number information and tense information, it is commonly referred to as a synthetic ending. The plural ending, on the other hand, does not encode tense information. It just says: you are dealing with a plural. Hence it is an analytic inflection.
Thus verbal agreement can be either synthetic or analytic. Does complementiser agreement have all the properties of verb agreement? No, it does not. In order to come to grips with the remarkable paradigmatic distribution of complementizer agreement it is important to stress the fact that the agreement between the complementizer and the verb always involves person and number, never tense (cf. already Van Haeringen 1958). Put differently, complementizer agreement never depends on marking for tense, but on marking for person and number only. The net result is that the agreement ending that is added to the complementizer is the same in both present tense sentences and preterite sentences.
By taking the latter fact as a starting-point, we will propose a new generali-zation which may bring us to a closer understanding of the fact that complemen-tizer agreement generally has a defective paradigm. We will elaborate upon this proposal in the following section.
 

4. A New Generalization
Just like the verb, the complementizer agrees with the subject. A close inspec-tion of complementizer agreement and verb agreement yields an astonishing result. It turns out that the agreement ending that is added to the complementizer is the same in clauses in the present tense and clauses in the past tense. Put differently, the ending added to the complementizer agrees with both the agreement ending of the present tense verb and the preterite verb. This means that the complementizer never expresses tense information. Therefore, complementizer agreement is never synthetic.
This leads us to the following two conditions on complementizer agreement:

(16) The PNT condition
Complementizer agreement can be agreement for Person and Number but it may not express tense

(17) The homophony condition
Complementizer agreement must be homophonous to verbal agreement

The corollary of these two conditions is the following generalization.

(18) The Identity Generalization (to be revised)
Complementizer agreement only occurs when the agreement ending of the inverted verb in the present tense is identical to the agreement ending of the inverted verb in the preterite

This generalization explains the defectiveness of the paradigms of dialects with complementizer agreement that have been discussed in the previous sections. In Frisian, for instance, the agreement ending for 2SG is identical in present tense and preterite (-st ). As expected, Frisian has complementizer agreement in 2SG. However, the agreement ending in 123PL is not identical in present (-e ) and in the preterite (-n ). As expected, Frisian has no complementizer agreement in the plural.
In the Groningen area, however, the ending for 123PL in present tense and preterite verbs is identical (-n ). Indeed, different from Frisian, the Groningen dialects do have complementizer agreement in the plural.
In the eastern dialects that have complementizer agreement in 1PL (cf. section
2.3: wy speult 'we play' but speule wy 'play we', hence datte wy 'that we') we expect, on the basis of the condition above, that the agreement ending reads -e in preterite 1PL, and this is indeed the case. Cf. the following examples taken from Van Haeringen (1958):

(19) a. speul-e wy 'play we' (present 1PL)
b. bet-e wy 'bit we' (preterite 1PL)

In 3PL,7 however, the present tense ending is not identical to the preterite ending. Cf. the following examples (Van Haeringen 1958):

(20) a. speul-t ze 'play they' (present 3PL)
b. beet-n ze 'bit they' (preterite 3PL)

Hence, in 3PL complementizer agreement is absent in the dialects in question.
All in all, the generalization formulated under (18) explains the defective paradigms of the dialects that have been discussed in the present paper. Cf. the table below that sums up the facts:

Table 1: Verb agreement and complementizer agreement in Dutch dialects.

We compare present and past forms within a person/number combination.
present and past => comp agreement
present and past => no comp agreement
North Holland
2SG: -E and -E => comp agreement for -E
3SG: -T and ZERO => no comp agreement
123PL: -E and -E => comp agreement for -E
South Holland
3SG: -T and ZERO => no comp agreement
123PL: -E and -E => comp agreement for -E
Zeeland Flanders
1SG: -N and -N => comp agreement for -N
123PL: -N and -N => comp agreement for -N
Friesland
2SG: -ST and -ST => comp agreement for -ST
3SG: -T and ZERO => no comp agreement
123PL: -E and -EN => no comp agreement
Groningen
2SG: -ST and -ST => comp agreement for -ST
3SG: -T and ZERO => no comp agreement
123PL: -N and -N => comp agreement for -N
Overijssel
3SG: -T and ZERO => no comp agreement
1PL: -E and -E => comp agreement for -E
3PL: -T and -N => no comp agreement
Limburg
2SG: -S and -S => comp agreement for -S
3SG: -T and ZERO => no comp agreement
2PL: -T and -T => comp agreement for -T

The data on complementizer agreement that we have at our disposal at present appear to corroborate our hypothesis that complementizer agreement only occurs if the agreement ending of the present tense verb is identical to the agreement ending of the preterite verb.
Zero-agreement also provides evidence for or against our account, of course. Although it is logically possible that a dialect does not have verb agreement, whereas it does have comp agreement, we predict that if the verb-agreement in inversion is zero, then the comp agreement must also be zero. In all dialects discussed above, the predictions concerning zero-agreement are borne out.

5. Sharpening up the Identity Generalization
We have claimed that the complementizer agreement ending must be homo-phonous with the agreement ending of the verb. But verbs constitute a large class. It is hardly conceivable that a grammatical condition checks on all mem-bers of the class of verbs. It is far more likely that only a characteristic closed subset of all verbs is concerned. Hence we suggest the following revision of the homophony condition:

(21) The Identity Generalization (revised)
Complementizer agreement only occurs when the agreement ending of the inverted auxiliary in the present tense is identical to the agreement ending of the inverted auxiliary in the preterite

We will now go on to provide three pieces of evidence for the Identity Generalization, as restricted to auxiliaries.
1. Consider first the dialects of South Holland. Main verbs do not exhibit a singular/plural opposition in the past whereas auxiliaries do. Interestingly, complementizer agreement behaves as if the main verbs did not exist:

(22) PAST SG and PAST PL
C-agt: ZERO and -E
Aux: ZERO and -E
Weak verbs: -E and -E

Put differently, weakly conjugated main verbs because of their past tense constitute a counterexample to the unrevised homophony condition, but not to the final version.
2. Limburg facts similarly support the claim that the auxiliaries (being a subclass of the strong verbs) are relevant, not just all verbs. In the Limburg dialects, the 2PL verb forms end in -T as far as present tense verbs and irregular past tense verbs (this includes all auxiliaries) are concerned. However, the 2PL of regular PAST tense verbs does not end in -T, but in ø:

(23) Present tense irregular (hence all auxiliaries):
a. kin-t ger 'can you'
b. * kin ger

(24) Present tense regular:
a. woen-t ger 'live you'
b. * woen ger

(25) Past tense irregular (hence all auxiliaries):
a. waor-t ger 'were you'
b. * waor ger

(26) Past tense regular:
a. * woende-t ger 'lived you'
b. woende ger

Not withstanding the fact, then, that the 2PL agreement ending is absent on regular past tense verbs, complementizer agreement does occur in 2PL thanks to the fact that it does occur on irregular preterites among which auxiliaries). This supports the final version of the Identity Generalization.
3. Facts from Standard Dutch provide a third piece of evidence for the proposed revision of the Identity Generalization:

(27) Standard Dutch present past
Weak verbs, 13PL -E and -E
Strong verbs, 13PL -E / -N and -N

This is the reverse case of South Holland and Limburg. In South Holland, the auxiliaries, and only the auxiliaries made possible complementiser agreement. In Standard Dutch, the auxiliaries and only the auxiliaries block complementiser agreement.
Notice that we capitalize on the fact that all auxiliaries are strong verbs. Of course, we could also restrict the Identity Generalization to strong verbs. However, the notion 'strong verb' seems to us a less basic notion than 'auxiliary'. There are several reasons for this. The notion 'strong verb' is restricted to Indo-European, the notion 'auxiliary' is presumably relevant to all languages of the world. Furthermore, there is no relation between the notion 'strong verb' and the notion 'complementizer'. However, there is a relation between the notion 'auxiliary' and the notion 'complementizer' (Paardekooper 1961). We could even go one step further and suppose that not all auxiliaries but only one is relevant, for example, the verb zijn 'be'. We will leave this for future research.
 

6. Notes
* We would like to thank the audience at the Meertens Symposion 1994 on Complementizer Agreement and the audience at the Western Conference on Linguistics 1998 for stimulating questions and discussion.
1. This article is a revised version of Hoekstra & Smits (1997).
2. Never co-ordinate clauses.
3. The examples were taken from Van Ginneken (1939), Van Haeringen (1939, 1958), De Vries (1940), and from a corpus of spoken Maastricht (Provinc of Limburg) gathered in 1997.
4. Note that in Standard Dutch, complementizer agreement is absent as well. It is not unlikely that this should not be attributed to structural factors (cf. section 4 below), but to cultural factors instead. Specifically, it is claimed that due to the strong normativeness of standard languages, natural processes such as complementizer agreement develop far less easy in such languages than in dialects ( which are generally far less normative) (cf. Van Marle 1996).
5. However, for a number of Flemish dialects it is claimed that complementizer agreement is applied throughout the paradigm (cf. e.g. De Schutter 1997; Haegeman 1992).
6. Note, by the way, that another distinction should be made. This involves the fact that within some areas complementizer agreement is optional, whereas in other areas it is obligatory. Specifically, complementizer agreement is generally optional, the only two exceptions being Frisian (2SG) and the Limburg dialects (in 2SG; in 2PL it seems to be optional). We will not deal with the problem of the optionality of complementiser agreement here.
7. At present, we do not have any data at our disposal concerning the agreement ending for preterite verbs in 2PL.
 

7. References
Beckering Vinckers, J. 1872. Phonetische voorbarigheid, een middel ter verklaring van smiins. Taal en Letterbode 3, 165-171.
Ginneken, J. van. 1939. De vervoeging der onderschikkende voegwoorden en voornaamwoorden. Onze Taaltuin 8, 1-11.
Haan, G. de. 1997. Voegwoordcongruentie in het Fries. In. E. Hoekstra & C. Smits (eds), Vervoegde voegwoorden . Amsterdam: P.J.Meertens-Instituut, 50-67.
Haegeman, L. 1992.
Haeringen, C.B. van. 1939. Congruerende voegwoorden. Reprinted in Van Haeringen (1962). Neerlandica . Verspreide opstellen. The Hague: Daamen, 246-259.
Haeringen, C.B. van. 1958. Vervoegde voegwoorden in het oosten. Reprinted in Van Haeringen (1979). Gramarie . Utrecht: HES, 309-318.
Hoestra, E. 1993. Some implications of number agreement on COMP. In F. Drijkoningen & K. Hengeveld (eds), Linguistics in the Netherlands 1993 . Amsterdam: Benjamins, 61-68.
Hoekstra, E. & C. Smits. 1997. Vervoegde voegwoorden in de Nederlandse dialecten. In E. Hoekstra & C. Smits (eds), Vervoegde voegwoorden . Amsterdam: P.J.Meertens-Instituut, 6-29.
Hol, A.R. 1955. Het meervoud van het praesens in onze oostelijke dialecten. Taal en Tongval 7, 160-175.
Klatter, J. 1933. Dialectstudie en syntaxis. Onze Taaltuin 2, 73-84.
Marle, J. van. 1996
Meer, G. van der. 1991. The 'conjugation' of subclause introducers: Frisian -st . NOWELE 17, 63-84.
Schutter, G. de. 1997. Incorporatie-in-C in de Vlaamse en de Brabantse dialecten. In E. Hoekstra & C. Smits (eds), Vervoegde voegwoorden . Amsterdam: P.J.Meertens-Instituut, 31-49.
Vanacker, F. 1949. Over enkele meervoudsvormen van voegwoorden. Taal en Tongval 1, 32-45, 77-93, 108-112.
Visser, M. de & A. Goeman. 1979. Voegwoord, relatief partikel en persoonsvorm in een dialect. Een geval van schijnbare kongruentie bij voegwoorden in de 1e persoon enkelvoud. Taal en Tongval 31, 222-241.
Vries, W. de. 1940. Congruerende voegwoorden. Tijdschrift voor Nederlandse Taal- en Letterkunde 59, 78-79.

Eric Hoekstra & Caroline Smits
Meertens Institute
Po Box 94264
1090 GG Amsterdam
e-mail: ehoekstra@fa.knaw.nl
Caroline.Smits@Meertens.knaw.nl
 
  -----------------------------130621191427728 Content-Disposition: form-data; name="file_description"