- 1 Ǹjò̩-Kóo is the name proposed in Olaogun (2016) for a group of relatively mutually intelligible (...)
1Before Nkemnji (1995), Rizzi (1997) and Aboh and Pfau (2011), the widely held position was that wh-questions are formed by the movement of the wh-phrase to the left of focus marker in the Spec, FocP within the clausal left periphery, while yes/no questions employ different strategies depending on the language in question. For instance, in English, the derivation of yes/no questions was said to require three different rules : Aux-NP-Subject inversion, affix-hopping rule, and do-support, while in Yòrúba, yes/no questions were assumed to have been derived by adjunction as suggested in Yusuf (1992) when he said :
…the derivation of the yes/no questions in other languages may not be this syntactically exciting. For instance, in Yoruba, it is enough to adjoin a question particle to a declarative sentence. No movement is involved. Neither is any morphophonemic rule employed. Adjoined particles are Ǹjé̩ and s̩é, and a few dialectal or archaic forms are question markers (Q-M ).
2 Similarly, Ìlò̩rì (2010) posits that a yes/no question clause in Igálà and Yorùbá is derived by merging a convergent IP with the question element which regularly occurs clause-finally in Igálà and clause initially and finally in Yoruba while, wh-question clauses are derived by moving wh-operators to the Spec, QstP (Question Phrase) where they are immediately followed by the focus marker ni.
3However, Nkemnji (1995), Aboh and Pfau (2011) challenge this commonly held position, and propose that both yes/no questions and wh-questions are projected by the same functional head Intero, and also that wh-words do not participate in typing wh-clause as interrogative. This paper then investigates the claim in Ǹjò̩-kóo, and presents some empirically and theoretically motivated pieces of compelling evidence that lend credence to the said position.
4The work is carried out within the Minimalist Program (henceforth MP), see Chomsky (1995, 1998, 2000, and 2002). The minimalist framework employs very few basic structure-building syntactic operations. They are : select, merge, and agree.
5 Operation Merge and Agree : It is assumed in this work that the computational system that forms syntactic objects is now composed of operations merge and agree. This is because operation move is assumed to be a subpart of merge and that agree is responsible for feature movement.
6 Operation Select chooses relevant lexical items from the lexicon and puts them into the numeration for further computation. Numeration is a storehouse of lexical items chosen from the lexicon for building larger syntactic objects. The items in the numeration are subsequently combined by operation merge.
7 Merge is a binary operation which combines two elements X and Y to form a larger unit that takes its syntactic category from either X or Y (Collins, 2013). Operation merge is of two types, namely, external and internal merge. External merge takes care of the merger operations of two independent elements which originate from the numeration, while internal merge is concerned with two dependent elements that are already introduced into the derivation.
8 Agree and Move F-based systems are two approaches developed to explain the conditions/requirements for movement of features in the MP. In this paper, the agree-based method is employed. Agree-based system, as opposed to move- F technology holds the assumption that only [+interpretable] features of the lexical elements are specified in the lexicon before they enter the derivation, while elements with [-interpretable] features acquire their features in the course of the derivation. Given the appropriate domain for features matching, agree assigns values to unvalued features so as to satisfy morphological requirements while, at the same time, deleting such [-interpretable] features for LF purposes. Agree is a syntactic operation holding between a probe and a goal where a matching relation exists. Chomsky (2000) opines that for β to move to ἁ, a probe-goal relation must hold between at least one feature of ἁ and a corresponding feature of β. A probe is the highest head in a derivation which searches for a matching goal in its c-commanding domain, while a goal represents a constituent which is attracted by a higher head which serves as a probe (Radford 2009 : 387 &400).
- 2 For other appropriate responses, consider the following dialogue :
9Questions in natural languages can be classified into a number of types. One typological division, for example, is between yes-no questions and wh-questions (Radford 1988 :462). Yes-no questions are so called because they require, among other appropriate responses2, ‘yes’ or ‘no’ answers, whereas wh-questions are those that do not require yes or no answer but question a constituent. Wh-questions are generally used to refer to questions that involve an “independent” (wh-) interrogative word such as ko (what ?), konè̩ (who ?), kòfò̩n (when ?), kòsin (where ?) as exemplified in (2a-d) which are the wh-interrogative counterparts of the simple declarative sentences in (1a-d).
1. (a) Títí bo̩ ùji
Títí drink water
‘Titi drank water’
(b) Igbé̩è̩ji uwan ju ò̩gè̩dè̩
the child eat plantain
‘The child ate plantain’
(c) Igbé̩è̩ji uwan ju ò̩gè̩dè̩ úrá
the child eat plantain yesterday
‘The child ate plantain yesterday’
(d) Adé da àju rí aja
Adé buy yam in market
‘Ade bought yam in the market’
2. (a) ko Títí yè bo̩ ?
What Titi Inter drink
‘What did Titi drink ?’
- 3 It is observed that when ‘konè̩’ (who ?) is used in the language, two phonological processes are n (...)
(b) Kone̩ è̩3 ju ò̩gè̩dè̩ ?
Who INTER eat plantain
‘Who ate plantain ?’
10
(c) Kòfò̩n igbé̩è̩ji uwan yè ju ò̩gè̩dè̩ ?
when the child INTER eat plantain
‘When did the child eat plantain’ ?
(d) Kòsin Adé yè ke da àju ?
where Ade INTER ADV buy yam
‘Where did Ade buy yam ?’
(e) Kòdí Adé yè ke da àju ?
how Ade INTER ADV buy yam
‘How did Ade buy yam ?’
11As seen above, the sentences in both (1) and (2) involve transitive verbs. This does not, however, mean that intransitive verbs cannot participate in the wh-questions in the language. Consider the sentences in (3a-d) and (4a-d) below.
3. (a) Bolu yè̩
Bolu dance
‘Bolu danced’
(b) Òjó wo̩n
Ojo laugh
‘Ojo laughed’
(c) Olú se̩n
Olu sleep
‘Olu slept’
(d) Jo̩ké̩ pà
Joke vomit
‘Joke vomited’
4. (a) Kone̩ è̩ yè̩ ?
Who INTER dance
‘Who danced ?’
(b) Kone̩ è̩ wo̩n ?
Who INTER laugh
‘Who laughed ?
(c) Kone̩ è̩ se̩n ?
Who INTER sleep
‘Who slept ?’
(d) Kone̩ è̩ pà ?
Who INTER vomit
12 ‘Who vomited ?’
13
14The example sentences in (4a-d) are the constituent question counterparts of the declarative sentences in (3a-d)
15Yes/no questions are illustrated in (6a - f).
5. (a) Igbé̩è̩ji uwan ju ò̩gè̩dè̩
the child eat plantain
‘The child ate plantain’
(b) Òjó vè
Òjó go
‘Ojo went’
(c) Títí bo̩ ùji
Títí drink water
‘Titi drank water’
(d) Olú wò̩
Olú cry
‘Olú cried’
(e) Délé da àju
Délé buy yam
‘Dele bought yam’
(f) Na ba bàbá
You greet bàbá
‘You greeted daddy’
- 4 This (EHT) Extra High tone is taken to be Emphatic head in this work assuming that questions and fo (...)
6. (a) Igbé̩è̩ji uwan yè ju ò̩gè̩dè̩ é̩ ?
the child INTER eat plantain EHT4
‘Did the child eat plantain ?’
16
(b) Ojó yè vè é ?
Ojo INTER go EMPH
‘Did Ojo go ?’
(c) Títí yè bo̩ ùji í ?
Titi INTER drink water EMPH
‘Did Titi drink water ?’
(d) Olú yè wó̩ ?
Olu INTER cry
‘Did Olu cry ?’
(e) Olú yè da àju ú ?
Olu INTER buy yam EMPH
‘Did Olu buy yam ?’
(f) Na yè ba bàbá ?
You INTER greet daddy
‘Did you greet daddy ?’
17
18It is observed that both yes/no questions and wh-questions are marked morphologically by the same distinct question morpheme yè which follows the subject DP immediately. However, there is an extra high tone that surfaces in clause final position in yes/no questions which accompanies the question morpheme. The high tone is, however, not usually realized in cases where the utterances independently end in high tone words/morphemes as shown in (7a-b) below
7 (a) Òjó yè vá ?
Ojo INTER come
‘Did Ojo come ?’
(b) Òjó yè ké é ?
Ojo INTER do it
‘Did Ojo do it ?’
19It is noticed that the occurrence of this H tone is syntactically conditioned. Generally, it shows up in yes/no questions as shown in (6) above and focus constructions to show emphasis whenever there is movement of a DP to the clausal left periphery. It is also apparent from these examples that questions do not show a difference in word order with their declarative counterparts.
20The occurrence and non-occurrence of this clause final high tone is not limited to direct yes/no and wh-questions but also occurs in indirect yes/no and wh-questions as exemplified in (8a-f) and (9a-f).
8. (a) Mu bìrè pé ko Bólú yè ju
I ask that what Bolu INTER eat
‘I asked what did Bolu eat’
(b) Mu bìrè pé konè̩ è̩ ju ò̩gè̩dè̩
I ask that who INTER eat plantain
‘I asked who ate plantain’
(c) Mu bìrè pé kòfun Bólú yè ju ò̩gè̩dè̩
I ask that when Bolu INTER eat plantain
‘I asked when did Ade eat plantain’
(d) Mu bìrè pé konè̩ è̩ vá
I ask that who INTER come
‘I asked who came’
(e) Mu bìrè pé kòsin Adé yè ke gwe̩ ewó.
I ask that where Ade INTER ADV collect money
‘I asked where did Ade collect the money’
(f) Mu bìrè pé kàwán uwan S̩é̩gun yè s̩í.
I ask that how many child S̩é̩gun INTER bear
‘I asked how many children did S̩é̩gun give
birth to’
9. (a) Mu bìrè pé Ade yè ju ò̩gè̩dè̩ é̩
I ask that Ade INTER eat plantain EMPH
‘I asked did Ade eat plantain’
(b) Mu bìrè pé Bólú yè ju ú
I ask that Bolu INTER eat EMPH
‘I asked did Bolu eat’
- 5 Úrá has two high tones in citation form but is correctly written as ura here because the language (...)
c) Mu bìrè pé Bólú yà á vó̩dí ura5
I ask that Bolu INTER FUT arrive yesterday+EMPH
‘I asked did Bolu arrive yesterday’
(d) Mu bìrè pé Òjó yè vè é.
I ask that Ojo INTER go EMPH
‘I asked did Ojo go’
(e) Mu bìrè pé igbé̩è̩ji ò̩ran yè hù ú.
I ask that the bird INTER fly EMPH
‘I asked did the bird fly’
(f) Mu bìrè pé Joké̩ yè pà á.
I ask that Joké̩ INTER vomit EMPH
‘I asked did Joke vomit.
21Adopting cartography approach, the paper proposes, following Rizzi (1997) and Aboh (2004) that interrogative force is a specification of the functional head in Intero encoding the feature [Interrogative] that projects between ForceP and FinP as given in (10).
10. Force…>Inter…>Topic…>Focus…>Finiteness.
22Therefore, I will propose in this paper that a simple yes/no question is headed by the Intero head which is morphologically realized as yè. So, a yes/no question like (6c) repeated here as (11) is derived as sketched in (12).
11. Títí yè bo̩ ùji í
Titi INTER drink water EMPH
‘Did Titi drink water ?’
12.
23
24The syntax of yes/no questions involves two probes each of which can provoke displacement operations. The yes/no question in (12) is derived thus : The verb bo̩ is first merged with its DP complement ùji (water) to satisfy the c-selection requirement of the head, while the subject DP Títí is second merged in the spec-VP (in line with the predicate internal subject hypothesis) in order to satisfy the EPP demand of the head. Then, T head is merged with the VP to project T-bar. At this point, the T head becomes the probe which searches its c-command domain for a matching goal to attract to the spec-TP so as to value the unvalued/un-interpretable feature. The subject DP, Títí, being the potential and active goal with an unvalued nominative case, is attracted to the spec-TP and the unvalued case feature is valued and deleted.
25The Emph head is externally merged with the TP to meet its c-selection condition. The whole TP pied-pipes to the spec-EmphP to also fulfill the EPP feature of the Emph head that is morphologically realized as the clause final high tone morpheme.
26The derivation proceeds by externally merging Inter head which is realized as yè to EmphP to project Emph-bar. At this stage, the Inter head becomes the probe and begins to search its c-command domain for an active goal to move to its spec and value the unvalued feature. The only active goal is subject DP Títí because it is the only constituent that is licensed at the spec-InterP. The probe has [Inter-EPP] feature which is valued and deleted by moving the goal Títí to its specifier position. Put in another way, the probe yè has [Inter-EPP] feature which must be valued and deleted by moving the matching goal to its specifier position.
27The focus of this section is to propose a unified analysis for both yes/no and wh-questions. The argument put forward above is that yes/no questions are headed by an interrogative morpheme Intero (which is morphologically realized as yè) that projects an InterP (Interrogative Phrase) and that the head of the InterP is licensed by the movement/raising of the subject of the finite clause to Spec, InterP. Given this assumption, the question that is begging for an urgent answer now is whether the analysis proposed for yes/no questions can be straightforwardly extended to wh-questions. Definitely, I will also propose and defend the assumption that wh-questions have the same structure as yes/no questions. Precisely, I posit that wh-questions are also InterP headed by an Intero, and that the “feature” WH, in Nkemnji’s words, is more like a scope marker that serves to delimit the constituent that is questioned.
28As said earlier, wh-questions and yes/no questions share a number of significant morphosyntactic properties. Just like yes/no questions, wh-questions involve a question particle yè (distinct from wh-phrase) that merges in Inter. For this reason, I argue that wh-questions are also headed by Intero head. Accordingly, I will analyze wh-questions in the same way as yes/no questions. As I intend to argue in the next section, the movement of wh-phrase is not meant to license interrogatives in wh-questions. Rather, the licensing of interrogatives takes place in overt syntax by way of the movement of the subject of the TP to the spec of InerP. Thus, a wh-interrogative like (13) is derived as sketched in (14).
13. Ko Títí yè ju
What Titi INTER eat
‘What did Titi eat ?’
29
30
14.
31The syntax of a wh-question involves three probes, namely, inter, foc and emph, each of which has the potential to trigger displacement operations. The derivations go as follows : the verb ju ‘to eat’ merges with content word ko ‘what’ to satisfy its c-selection condition while the DP subject Títí is second merged in the spec-VP (in line with the VP Internal Subject Hypothesis) to satisfy the EPP demand of the head. Then T head is merged with the VP to project T-bar. At this point, the T head becomes the probe which searches its c-command domain for a matching goal to attract to the spec-TP so as to value the unvalued/un-interpretable feature. The subject DP, Títí, being the potential and active goal with an unvalued nominative case is attracted to the spec-TP and the unvalued case feature is valued and deleted.
32The Emph head is externally merged with the TP to meet its c-selection condition. The whole TP pied-pipes to the spec-EmphP to also fulfill the EPP feature of the Emph head that is morphologically realized as the clause final high tone morpheme.
33There is a theoretical reason why the pied-piping of the TP to the spec-EmphP is necessary. Assuming phase impenetrability condition, if the probe yè is merged with the EmphP and later attracts the goal Títí to the spec-InterP, the phase domain of the Emph head will undergo transfer and will no longer be accessible for any syntactic operation. Therefore, this will block the movement of the content word to the spec, FocP and cause the derivation to crash. Similarly, if the ko-word is moved before pied-piping, it will also block the movement of the subject DP Títí to the spec-InterP, bringing about a non-convergent derivation. This being the case, the whole TP is first pied-piped to the spec-EmphP, with other syntactic operations then following.
34After pied-piping, yè is externally merged with the EmphP to project Inter-bar. The Inter head now becomes the probe which searches its c-commanding domain for a goal to raise to its spec. The subject DP, being the only active goal with [+Inter], then moves to the spec-InterP to value and delete the [Inter-EPP] feature of the probe head.
35The derivation proceeds by externally merging null foc head to InterP to project Foc-bar. The foc head now becomes the probe being the highest head which begins to search its c-command domain for a matching goal. According to Chomsky, for a movement operation to take place, there must be an agreement relation between the probe and the goal. That is, the probe must have unvalued/un-interpretable feature while the corresponding feature must be valued or [+interpretable] on the goal so that, assuming feature valuation, the feature on the goal can be copied on to the probe. Given this, the probe has [focus-EPP] feature while the goal ko has [+foc] feature, the requirement for movement to take place is fulfilled. Therefore, the goal moves to the spec-FocP to value and delete the un-interpretable [foc-EPP] feature that is not legible at LF, which can cause the derivation not to converge at that interface level.
36The analysis in (14) is plausible in that, apart from the fact that the correct word order is realized, it also corroborates the unified analysis proposed in this paper. The analysis specifically assumes that Inter head yè merges with EmphP in both yes/no and wh- questions.
37The analysis also assumes that wh-questions involve the occurrence of the clause final high tone that occurs in yes/no and focus constructions as a mark of emphasis. This claim can also be theoretically justified. If we do not assume the existence of the clause final high tone for wh-questions, the unified analysis proposed for both yes/no questions and wh-questions would generate some questions. For instance, why is it that the inter head yè in yes/no question selects EmphP but the same yè performing the same function in wh-questions merges with TP ? But if we assume that a wh-question clause also has clause final high tone which is phonetically null, this asymmetry will disappear. Also, since we claim that yè licenses the existence of the clause final high tone morpheme in yes/ no question, and it also occurs in wh-questions, there is no way we can claim otherwise. The explanation one can offer is to say that it also occurs in wh-questions but phonetically null or not morphologically realized.
38The assumption that the complementizer system codes information that would indicate whether a sentence is a question, declarative, relative, etc. is termed Clausal Type (Cheng, 1991), or the specification of Force (Chomsky, 1995). In her (1991 :30) influential thesis titled “On the Typology of Wh-Questions”, Cheng proposes Clausal Typing Hypothesis as follows :
Every clause needs to be typed. In the case of typing a wh-question, either a wh-particle in CO is used or else fronting of a wh-word to the Spec of CO is used, thereby typing a clause through CO by Spec-head agreement.
39On this basis, Cheng categorizes all the languages of the world into two : Wh-in situ languages and Wh-movement languages. She claims that in Wh-in situ languages, wh-particles are used to type a clause as interrogative, while in Wh-movement languages, wh-questions are typed by the movement of wh-word/phrase to the Spec, CP because such languages lack the kind of wh-particles (which she called typing particles) found in wh-in situ languages. In short, Cheng believes that the two syntactic phenomena are in complementary distribution. However, cross-linguistic evidence shows that there are wh-movement languages with typing particles as demonstrated by the Igbo examples below.
15a Ì̩ hù̩rù̩ ònye ?
You see-rV(past) who
‘Who did you see ?’
b Ònyei kà í̩ hù̩rù̩ ti
Who that you see-rV(past)
‘Who did you see ?’
(Uwalaka, 1991 :186)
40Example (15a) instantiates a case where the wh-phrase remains in situ, while that of (15b) exemplifies an option where the wh-phrase is displaced to the sentence initial position.
41There are two main proposals on how sentences are clause typed as interrogative in language, namely, (1) by using typing particles in wh-in situ languages or by using syntactic wh-movement in non-wh in situ languages (Cheng, 1991), and (2) by employing abstract syntactic X-bar category-InterP (Nkemji, 1995 and Aboh & Pfau, 2011). Nkemji (1995) and Aboh & Pfau (2011) opine that, cross-linguistically, sentences are typed as interrogatives by Inter head, contrary to Cheng’s proposal.
42Aboh and Pfau disagree with the assumption that wh-questions are derived by moving wh-phrase to the specifier of a focus projection cross-linguistically. They further claim that it is linguistically paradoxical to imagine that two functional heads with different properties (Inter, Foc) end up encoding the same discourse information (i.e interrogative force). This being so, they concluded that yes/no operators (or particles) and wh-operators activate different articulations within the C-system, InterP and FocP, respectively. And that cross-linguistically, wh-phrases do not target the same position as evident in French. Therefore, wh-phrase does not participate in clause typing. It is just like a scope marker that serves to delimit the constituent that is questioned. Similarly, Ilori (2017) says that content word question operators are not interrogative heads but some kind of nominal words that interpret the focus of the interrogative force.
43The paper argues in favour of the claim that sentences are clause typed as interrogative by InterP. This is because, in Ǹjò̩-Kóo, there is the presence of question particles as well as wh-movement, contrary to Cheng’s claim that the two syntactic phenomena are in complementary distribution. Consider the sentences below.
16. (a) Títí bo̩ ùji
Titi drink water
‘Titi drank water’
(b) Ko Títí yè bo̩ ?
what Titi INTER drink
‘What did Titi drink ?
44Ǹjò̩-kóo is a wh-movement language because, when the DP object ùji in (16a) is questioned in (16b), the wh-word ko with which the object is questioned is preposed to the beginning of the wh-question clause as given in the tree diagram in (17).
17.
45At this point, two questions beg for answers. If we claim that the movement of wh-word to the Spec, FocP is not for typing purposes in (17), why does it not stay in situ ? And what clause -types the sentence as interrogative ?
46As said earlier, yè is the particle that encodes interrogative force. So, the licensing of interrogatives takes place in overt syntax by movement of the subject DP of the TP to the Spec of InterP, given the Earliness Principle which states that linguistic operations must apply as early in a derivation as possible. Thus, following Nkemnji (1995) and Aboh & Pfau (2011), the movement of wh-phrase takes place at LF for focusing so as to delimit the constituent being questioned or for interpretive purposes as reinforced by the information structure evidence in the next section.
47 Information structure means the way speakers package messages and send them to the hearers (Olaogun, 2016, 2017). The traditional grammarians who claim that the wh-words clause type wh-clauses as interrogative have never bothered to ask why the answer to every wh-question /content word question is always a focus construction as given below.
48
18 (a) Olú ju àju
Olu eat yam
Olu ate yam
(b) Ko Ø Olú yè ju ?
What FOC Olu INTER eat
‘ What did Olu eat ?
(c) Àju úwò̩n Olú ju ú
yam FOC Olu eat EMPH
‘ Olu eat YAM’.
49(18a) is a declarative statement ; (18b) is used to question the object DP in (18a), while (18c) is a response to the question. In the language, the focus marker is úwò̩n and its occurrence immediately after àju in (18c) indicates that àju is focused. Although the focus marker úwò̩n is phonetically null in wh-questions in Ǹjò̩-kóo, the fact that the response to it in (18c) is a focused construction is an attestation to the fact that the question clause has got something to do with focusing. The conclusion therefore is that, because the wh-phrases are focused in Ǹjò̩-kóo, their answers must also be focus constructions. The preceding observation holds because not all languages permit this kind of syntactic operation. For instance, Aboh (2007) demonstrated the dichotomy between focused and non-focused wh-phrases/words. The answer to the non-focused wh-words is never a focus construction as demonstrated in the Oromo data in (19), culled directly from Yiman (1988 :370) :
19. (a) Eennu duf-e ?
Who come-3sg-past ?
‘Who came ?’
(b) Túlluu (duf-e)
Tulluu come-3sg-past
‘Tulluu (came)
(c) Eennu-tu duf-e
Who-Foc come-3sg-past
‘Who is it that came ?’
50Tu is the focus marker in the language but its non-occurrence in the answer to the wh-question in (19a) is an attestation to the fact that wh-phrases may not be focused in the language.
51Besides, there is empirical evidence indicating that wh-phrases in wh-questions or content word questions are not meant to clause-type expressions as interrogative. Thus, in English, there are expressions with wh-phrases that do not express interrogative force as exemplified in (20) taken from Aboh (2010) with modifications.
20. (a) We met the man whom you sanctioned last year.
(b) The committee decided over who will represent the University at the meeting
(c) The boy who bought a car last week is dead.
52The fact that English wh-phrases occur in both declaratives and interrogatives as shown above further strengthens the argument that wh-phrases are not meant for typing clauses as interrogatives.
53Furthermore, many wh-movement languages do not always have (overt) wh-phrases in wh-questions. For instance, according to Aboh (2010), Li and Thompson (1975) in their discussion of word order in Mandarin Chinese, report that Mandarin Chinese exhibits wh-questions without wh-phrases as shown in the example below.
21. Yaoshi ne ?
Key INTER
‘What about the key(s) ?’
54The example in (21) lacks wh-words but is interpreted as a wh-question. All that it contains is a noun phrase closely followed by a question particle : ne. Aboh reports that Li and Thompson (1981 :305-306) claim that this declarative typing particle can also be employed to clause type various questions, including truncated questions that include only a noun phrase. What this suggests is that we can realize wh-questions in Mandarin Chinese without involving genuine wh-phrases such as shei (who ?), sheme (what ?), and duo (how ?), as shown in the examples below.
22. (a) Hufei Chi-le sheme (ne) ?
Hufei eat Asp what QwH
‘What did Hufei eat ?’
(b) Shei mai-le sheme (ne) ?
Who buy-Asp what QwH
‘Who bought what ?’
55 Similarly, Yoruba displays wh-questions without containing canonical wh-phrases such as kí̩ (what ?), ta (who ?), mélòó (how many ?), as exemplified below.
23. (a) Fúnmi dà ?
Funmi INTER
‘ Where is Funmi ?’
(b) Bàtà ye̩n ńkó̩ ?
Shoe that INTER
‘Where is that pair of shoes ?’
(c) Victor ńkó̩ ?
Victor INTER
‘Where is Victor ?’
‘How/What about Victor ?’
56As demonstrated above, the questions in (23a-c) which involve the use of question particles dà and ńkó̩ lack wh-words but are all interpreted as wh-questions. This is another piece of evidence that supports the fact that wh-words do not express interrogative force. The genuine wh-words in Yoruba are kí (what), ta (who), mélòó (how many), as exemplified in the questions below.
24. (a) Kí ni Olú rà ?
What FOC Olu buy
‘What did Olu buy ?’
(b) Ta ní Olú rí
who FOC Olu see
‘Who did Olu see ?’
57Finally, questions do have question particles, regardless of whether the language involves wh-phrases and/or wh-movement. For example, languages such as Lele and Ǹjò̩-kóo, which are wh-movement languages, have question particles as well as wh-phrases in the same sentences as given in (25a-b) and (26a-e) respectively.
25 (a) Wey ba é gà ?
Who FOC go INTER
‘Who went away’ ?
(b) Me ba gol dí gà ?
what Foc see 3sg INTER
‘What did you see ?’
(Aboh & Pfau 2011)
26. (a) ko Títí yè bo̩ ?
What Titi INTER drink
‘What did Titi drink ?’
(b) Konè̩ yè ju ò̩gè̩dè̩ ?
who INTER eat plantain
‘Who ate plantain ?’
(c) Kòfò̩n igbé̩è̩ji uwan yè ju ò̩gè̩dè̩ ?
when the child INTER eat plantain`
‘When did the child eat plantain’ ?
(d) Kòsin Ade yè da àjú ?
where Ade INTER buy yam
‘Where did Ade buy yam ?’
(e) Kòdí Adé yè da àju ?
how Ade INTER buy yam
‘How did Ade buy yam ?’
58The only plausible conclusion we can draw from this is that the particle is silent (i.e phonetically null) in English but overt in Mandarin Chinese, Ǹjò̩-kóo, and Lele.
59 The thrust of this paper has been to provide a unified analysis for yes/no questions and wh-questions, and also to prove that the movement of wh-word in wh-question clause is not for clause typing but, rather, for interpretive purposes. As opposed to Cheng (1991) who observed that wh-movement languages do lack typing particles that are found in wh-in-situ languages, language internal and cross-linguistic data show that wh-questions (in addition to wh-word) have the same interrogative markers as yes/no questions. This makes it empirically possible for us to propose a unified analysis for both question types as well as dissociating movement of wh-phrase from clause typing. This being the case, the analysis assumed a common head structure for both yes/no question and wh-questions thereby challenging the traditional analysis which claims that yes/no and wh-questions involve different derivations.