• Dutch1
  • Frisian
  • Saterfrisian
  • Afrikaans
Show all
Interrogative main clause with verb second (wh-interrogative)
quickinfo

Specific interrogative, or wh-interrogative, main clauses in Afrikaans are formed by placing an interrogative expression in the clause initial position, followed by the (first) verb in second position, and the remainder of the clause from the middle field onwards. If one construes interrogative formation as a syntactic derivation, then the element that is being asked about is moved from whatever position it would otherwise occupy in the clause to the clause-initial position. Any argument, complementive or adverbial can in principle be targeted for interrogation, as illustrated by the selection of examples in (1), with syntactic function of the interrogated element, combined with animacy, determining the form of the interrogative (or wh) expression itself. Similar to Dutch and German, and unlike English, specific interrogative formation in Afrikaans involves no additional auxiliary verb, and neither does general (or Yes/No) interrogative formation.

1
a. Wie is daar?
[(CI) wie] [(V2) is] [(MF) daar]
who be.PRS there
Who is there?
[WH-subject]
b. Wie/wat sien jy?
[(CI) wie/wat] [(V2) sien] [(MF) jy]
whom/what see.PRS you.SG.SUB
Who/what do you see?
[WH-direct object]
c. Waar kruip hulle weg?
[(CI) waar] [(V2) kruip] [(MF) hulle] [(VF) weg]
where crawl.PRS they away.PTCL
where do they hide?
[WH-adverbial]

When the interrogative element is part of a preposition phrase, it is conventional for Afrikaans to place the entire phrase, including the preposition, in the clause-initial position, as shown in (2a). This is known as pied-piping. The combination of preposition + wh-word results in a compound with an alternative form for the interrogative wat what, viz waar. Certain prepositions, like met with, also changes its form in these compounds, e.g. met wie > waarmee. However, the data attest to an alternative formulation where the preposition remains stranded in its original position in the clause and only the interrogative is moved to the initial position, as shown in (2b). Ponelis (1979:178-179) attributes this phenomenon, known as preposition stranding, to the influence of English.

2
a. Waarmee word die baard geskeer?
[(CI) waarmee] [(V2) word] [(MF) die baard] [(VF) geskeer]
what.with be.AUX.PASS.PRS the beard shave.PASS
With what is the beard shaved?
b. Wat word die baard mee geskeer?
[(CI) wat] [(V2) word] [(MF) die baard mee] [(VF) geskeer]
what be.AUX.PASS.PRS the beard with shave.PASS
What is the beard shaved with?

Wh-interrogatives are positioned in the clause-initial position of main clauses, even if the wh-element is part of an embedded clause, a phenomenon known as long-distance wh-extraction. While generative analyses tend to focus on the formal properties and constraints on extraction, functional analyses based on large corpora have yielded the insight that long distance wh-extraction is a construction with a relatively limited application, and largely applies to main clauses with a second-person pronoun, singular rather than plural, and the verb dink to think, or a small set of other mental or cognitive verbs (Van Rooy and Kruger 2015). A typical and less typical case of wh-extraction is shown by (3a) and (3b) respectively, while (3c) is an invented example (after Verhagen (2005)) that seems unlikely to be acceptable in Afrikaans.

3
a. Wat dink jy gaan jy leer tydens hierdie aktiwiteit?
[(CI) wat] [(V2) dink] [(MF) jy] [(PV) gaan jy ___ leer tydens hierdie aktiwiteit]
what think.PRS you.SG.SUB go.LINK you.SG.SUB learn.INF during this activity
What do you think you are going to learn during this activity?
TK
b. Watter een sou jy wou gehad het moes uitgegooi word?
[(CI) watter een] [(V2) sou] [(MF) jy] [(VF) wou gehad het] [(PV) ___ moes uitgegooi word]
which one will.AUX.MOD.PRT you.SG.SUB want.to.AUX.MOD have.PST have.AUX must.AUX.MOD.PRT out.throw.PASS be.AUX.PASS.PRS
Which one would you have wanted to have been thrown out?
TK
c. *Wat het die minister ná die perskonferensie aan die joernaliste erken (dat) die Eskomwerkers het gesaboteer?
[(CI) wat] [(V2) het] [(MF) die minster ná die perskonferensie aan die joernaliste] [(VF) erken] [(PV) dat die Eskomwerkers het ___ gesaboteer]
what have.AUX the minister after the press.conference to the journalists acknowledge.PST that.COMP the Eskom.workers have.AUX sabotage.PST
What did the minister acknowledge to the journalists after the press conference (that) the workers sabotaged?

While cases with more than one interrogative element in the same sentence are theoretically possible, as shown by the hypothetical example in (4), they are unattested in the available corpora of Afrikaans that I had occasion to analyse. Presumably, as in English, the first wh-element, as one scans the sentence from left to right, will be extracted to the initial position, while the other one will remain in situ.

4
Wie het jou wat vertel?
[(CI) [(SUB) wie]] [(V2) het] [(MF) [(IO) jou] [(DO) wat]] [(VF) vertel]
who have.AUX you what tell.PST
Who told you what?
readmore
[+]Syntactic function of clause-initial interrogative

The interrogative pronouns of Afrikaans encompass a range of options from a structural perspective. The interrogatives wat what and wie who can be used for an entire noun phrase, and are typically used as subject or direct object of the clause, as illustrated by the examples in (5). Irrespective of the syntactic role of the interrogative pronoun, it is always placed in the clause-initial position, and other nominal arguments follow the verb-second in the middle field. The interrogative pronouns of Afrikaans is invariable in form, irrespective of whether they function as subject or object. The formal contrast between these two pronouns relates to human versus non-human referents: wie for human referents and wat for non-human referents, with some domestic animals potentially forming a grey area in-between.

5
a. Wie melk die koeie?
who milk.PRS the cows
Who milks the cows?
TK
[Subject interrogative]
b. Wat soek hulle dan?
what search.PRS they then
What are they looking for?
TK
[Direct object interrogative]

The indirect object of a clause is represented by the interrogative form vir wie for who, rather than just wie who, as illustrated by (6a). The form without the preposition seems unlikely, as illustrated in (6b), while preposition stranding, shown in (6c), is a more likely alternative to the initial PP containing the interrogative pronoun. It is also clear from the gloss and translation that this particular type of preposition stranding is closely modelled on the template of English.

6
a. Vir wie koop jy dié klas van beitels?
for who buy.PRS you.SG.SUB this class of chisels
Who do you buy these quality chisels for?
TK
b. ?Wie koop jy dié klas van beitels?
who buy.PRS you.SG.SUB this class of chisels
To mean: Who do you buy these quality chisels for?
c. Wie koop jy dié klas van beitels voor?
who buy.PRS you.SG.SUB this class of chisels for
Who do you buy these quality chisels for?

The interrogative watter which/what and its informal variant watse are not used as bare noun phrases on their own, but function as interrogative determiners to a head noun, in which case the entire NP is placed in the clause-initial position. Afrikaans also allows a genitive construction, with the interrogative pronoun wie who, combined with the genitive particle se, to form wie se, that can be used in the same syntactic position as watter and watse. These options are illustrated by the examples in (7).

7
a. Watter aaklige ding het haar nou getref?
what terrible thing have.AUX her.OBJ now hit.PST
What terrible thing struck her now?
TK
b. Watter mens ken sy kind as hy gebore word?
which human know.PRS his child when he born.PASS be.AUX.PASS.PRS
Which person understands his child at birth?
TK
c. Watse ammunisie gebruik jy?
what ammunition use.PRS you.SG.SUB
What ammunition do you use?
TK
d. Wie se skuld was dit?
who PTCL.GEN fault be.PRS it
Whose fault was this?
TK

The final set of interrogative pronouns substitute for adverbials or prepositional objects. When an explicit preposition is available, it combines with wat to form a [waar+PREP] unit, e.g. waarop upon which, waaronder whereunder, or waarvoor for which. However, there are also other adverbial interrogatives that are not directly dependent on a prepositional form, such as wanneer when, hoekom why and hoe how. Irrespective of the kind of adverbial role, or even a prepositional object role, the interrogated element is still placed in the clause-initial position, with the verb-second following, and then the arguments and other, non-interrogative adverbials, as exemplified in (8).

8
a. Wanneer het sy laas geëet?
when have.AUX she last eat.PST
When last did she eat?
TK
b. Waar bêre sy dit?
where store.PRS she it
Where does she store it?
TK
c. Waarop kom dit dan neer?
what.on come.PRS it then down
Upon what does it come down to?
TK
[+]Stranding and pied-piping

When the interrogative forms part of a preposition phrase, the conventional syntactic patterns in Afrikaans are to fuse the preposition with the waar form of the interrogative pronoun wat what, or in the case of the interrogative pronoun wie who for human referents, to form a regular preposition phrase consisting of [PREP wie]. The latter of these options is a typical case of pied-piping, where the entire phrase is placed in the clause-initial position, along with the interrogative, while the fusion variant represents a different type of pied-piping with the same effect, that the entire preposition phrase containing the interrogative occurs in clause-initial position. These two options are illustrated by the examples in (9).

9
a. Waaroor sou hulle praat?
what.about will.AUX.MOD.PRT they talk.INF
About what would they talk?
TK
b. Oor wie gaan ons praat sodat albei kante daarmee kan identifiseer?
about who go.LINK we talk so.that both sides there.with can.AUX.MOD identify.INF
About who(m) are we going to talk in order for both sides to be able to identify?
TK

Afrikaans does allow the stranding of the preposition, with two known variants. The first variant, which seems to be perfectly acceptable, from a grammatical but also a prescriptive point of view, is illustrated in example (10). The form waar where, rather than wat, occurs in the clause-initial position, while the associated preposition occurs in situ in the remainder of the clause. However, this variant seems to be restricted to combination with heen towards and vandaan whence/from hence, which may be used in the main clause as if they are directional adverbs, rather than prepositions, but then, both forms allow fusion with the form waar in the clause-initial position. The acceptable form of preposition splitting is illustrated by the examples in (10). Both variants are attested freely in the available corpora, although the split variant seems to be the more frequent one in the case of waarvandaan, while the fused variant is more frequent in the case of waarheen.

10
a. Waar kom hierdie skielike seksisme vandaan?
where come.PRS this sudden sexism hence
Where does this sudden sexism come from?
TK
a.' Waarvandaan kom die gebroke werklikheid waarin ons leef?
whence come the broken reality which.in we live.PRS
Where does the broken reality we live in, come from?
TK
b. Waar het al ons mag heen gegaan?
where have.AUX all our power toward go.PST
Where did all our power go to?
TK
b.' Waarheen het die tyd gevlieg?
where.to have.AUX the time fly.PST
Where did the time go?
TK

The other form of preposition splitting from the interrogative is widely regarded as non-standard in the prescriptive sources, but is attested, especially in spoken language. In this form of splitting, the preposition remains in situ, and the interrogative wat what, or some other form, but not waar, occupies the first position of the clause on its own. Two, examples, taken from Ponelis (1979:179) are reproduced in (11), with their standard language variants in the primed examples.

11
a. Wat lê jou skoene hier so rond voor?
what lie.PRS your shoes here so around for
What do your shoes lie around for like this?
a.' Waarvoor lê jou skoene hier so rond?
what.for lie.PRS your shoes here so around
Why do your shoes lie around like this?
b. Watse goedjies dink jy vanmiddag aan?
what.PTCL.GEN stuff.DIM.PL think.PRS you.SG.SUB this.afternoon on
What stuff are you thinking about this afternoon?
b.' Aan watse goedjies dink jy vanmiddag?
on what.PTCL.GEN stuff.DIM.PL think.PRS you.SG.SUB this.afternoon
About what are you thinking this afternoon?
[+]Long distance wh-extraction

The clause-initial position of a clause can be filled by an interrogative pronoun (or phrase containing an interrogative pronoun) that is a constituent of an embedded clause, and not only a constituent of the main clause itself. Du Plessis(1977) offers the following example of a sentence where two layers of embedding can be identified, and the PP containing the interrogative pronoun is extracted from the most deeply embedded clause and positioned in the initial position of the main clause:

12
Met wie het jy nou weer gesê het Sarie gedog gaan Jan trou?
[(MC) [(CI) met wie] [(V2) het] [(MF) jy nou weer gesê] [(PV) [(CC) het Sarie gedog [(CC) gaan Jan ____ trou]]]]
with who have.AUX you.SG.SUB now again say.PST have.AUX Sarie think.PST go.LINK Jan ____ marry.INF
Who did you say Sarie thought Jan was going to marry ____?
(Du Plessis 1977:725)

This construction has attracted attention from generative linguists for a long time, because of its ability to reveal general properties of syntactic movement, especially movement across clause boundaries. Du Plessis (1977) is a contribution to this debate from the perspective of Afrikaans grammar. However, Verhagen (2005) challenges the derivational view of the construction, on the basis of corpus data that point to severe limitations on the actual attested forms of wh-extraction. He observes that only a limited selection of main clauses, variants of WH denk jij dat + complement clauseWH do you think that + complement clause actually occur. Verhagen relies on data from Dutch and incorporates insights from research on English in his alternative view of the construction. Van Rooy and Kruger(2015) examine the phenomenon for Afrikaans, and find that Afrikaans also uses a very narrow range of main clauses with long-distance wh-extraction. The alternative analysis that they propose for Afrikaans is that long-distance wh-extraction is not a productive and very general (or highly schematic) rule of Afrikaans grammar, but rather a specific template for an wh-phrase in the initial position, followed by the expression dink jy you think and then the remainder of the complement clause, minus the interrogative element. The main clause dink jy functions semantically like a subordinate clause, and has the function of inviting intersubjective agreement between speaker and listener, while the presumed subordinate clause is thematically the most prominent question that is being asked. The ostensible main clause functions like a parathetic insert rather than a main clause from the thematic point of view. Because Afrikaans allows complementiser omission quite freely, the notion that the usual main clause-subordinate clause relation is inverted, is even more obvious for Afrikaans than for Dutch. According to Van Rooy and Kruger (2015:59), the basic pattern in Afrikaans is [WH + dink jy + Clause] WH + think you + Clause.

The corpus analysis of the Taalkommissiekorpus conducted by Van Rooy and Kruger (2015) confirms that the prototypical examples of long distance wh-extraction in Afrikaans are the following ones, where variants on dink to think are other mental and communication verbs that can also invite the addressee to participate in intersubjective coordination, or the sharing of perspectives, with the speaker:

13
a. Waarin dink jy lê die oplossing om die leemte wat die adolessent mag ervaar uit te skakel of te beperk?
In what do you think lies the solution to eliminate or restrict the gap experienced by the adolescent?
TK
b. Wat sou jy sê is jou sterkste eienskappe as atleet?
What would you say are your strongest attributes as athlete?
TK
c. Wat reken jy sal die resultaat van 'n belastingverhoging op voedsel wees?
What do you reckon will be the result of a tax increase on food?
TK
References
  • Du Plessis, H1977Wh Movement in Afrikaans.Linguistic Inquiry8723-726
  • Du Plessis, H1977Wh Movement in Afrikaans.Linguistic Inquiry8723-726
  • Du Plessis, H1977Wh Movement in Afrikaans.Linguistic Inquiry8723-726
  • Ponelis, F.A1979Afrikaanse sintaksis.Van Schaik
  • Ponelis, F.A1979Afrikaanse sintaksis.Van Schaik
  • Van Rooy, Bertus & Kruger, Haidee2015The case for an emergentist approachStellenbosch Papers in Linguistics Plus4841-67
  • Van Rooy, Bertus & Kruger, Haidee2015The case for an emergentist approachStellenbosch Papers in Linguistics Plus4841-67
  • Van Rooy, Bertus & Kruger, Haidee2015The case for an emergentist approachStellenbosch Papers in Linguistics Plus4841-67
  • Van Rooy, Bertus & Kruger, Haidee2015The case for an emergentist approachStellenbosch Papers in Linguistics Plus4841-67
  • Verhagen, Arie2005Constructions of intersubjectivity: discourse, syntax, and cognitionOxford/New YorkOxford University Press
  • Verhagen, Arie2005Constructions of intersubjectivity: discourse, syntax, and cognitionOxford/New YorkOxford University Press
printreport errorcite