Printer Friendly

The syntax of yes-no questions in Embosi.

1. Introduction

This paper provides a syntactic analysis of yes-no questions in Embosi. It is widespread in the generative framework that each sentence has a given spellout according to its different phases of derivation. It will result from this survey that at the S-structure both the statements and the yes-no questions will be so isomorphic that there difference is signalled out by their individual PF. This similarity between yes-no questions and statements stands as a violation of the endocentric principle will postulate projection from commonness. In addition, this description of yes-no questions will demonstrate two idiosyncratic features of Embosi which rejects either the insertion of an auxiliary to support the question formation or the auxiliary raising in case of T to C movement. And finally, it is argued in the Minimalism Program that movements are instances of features valuation. Yet, the non movement observed in embosi raises the question of features valuation in that features are not checked nor satisfied as the related item remains in situ.

2. Statements and yes-no questions

This section is just a starting point that introduces our analysis. It aims at showing the similarity between the affirmative and the yes-no questions in embosi.

(1) a. n[??] o-dzaa [??]k{/} you.SG you-eat.PRES.2SG cassava 'You eat cassava'

b. bini le-pera osunga nga you.PL you-can.PRES.2PL help me 'You can help me'

c. ba a-dii o lek[??]li them they-be.PAST.3PL at school 'They were at school'

d. wa a-si-dzwa him/her s/he-ASP-go. pres.3SG 'S/he has already gone'

f. nyak[??] a-konga oyaa wo grandparent s/he-FUT come there 'Grand parent will come here'

We do not provide a literary translation because these sentences have two readings in accordance with the way they are used. That is to say, if these sentences are not pronounced, no one can predict their category. From a surface look, they appear like ideal declarative sentences. There is nothing, indeed, which points out some speculations on their interrogative features. In fact, there is no syntactic clue which raises some Q item in the structure. In the chomskyan grammar, it is admitted that Q has the following [EPP, Tns, Wh]. And the WH features attracts the lower item that bears its feature and triggers its movement so that Wh features are valuated. One can wonder why the implementation of Move Wh is unsuccessful in embosi. If Chomsky (1995) consider C as strong head capable of triggering T to move to C, Ndongo Ibara dissertation offers another point of view which considers C as a weak head in embosi as it fails to attract some of elements bearing its features.

In broad terms, there is no syntactic marker for yes-no questions in embosi which means that they are affirmative. In fact, yes-no questions and affirmative differ only from their intonation patterns. As a matter of fact, yes-no questions make use oh raising intonation, while affirmative sentences use falling intonation. The application of this argumentation couches out to the following interpretations for (1).

(2) a. n[??] o-dzaa [??]k[??] you.SG. you-eat.PRES.2SG cassava Affirmative: You eat cassava. Interrogative: Do you eat cassava?

b. bini le-pera osunga nga you.PL. you-can.PRES.2PL help me Affirmative: You can help me. Interrogative: Can you help me?

c. ba a-dii o lek[??]li them they-be.PAST.3PL at school Affirmative: They are at school. Interrogative: Are they at school?

d. wa a-si-dzwa him/her s/he-ASP-go.PRES.3SG Affirmative: He/she has gone. Interrogative: Has he/she gone?

f. nyak[??] a-konga oyaa wo grandparent s/he-FUT come there Affirmative: Grandparent will come there. Interrogative: Will the grandparent come there?

The syntactic representation of (2a) will look like the following:


As yes-no questions has null spellout of Comp, it can be asked about the erasure of the features [Tns, Wh, EPP] which are deleted after the movement of the item that bears these features. It should be argued that the Tns feature is weak in embosi as it fails to attract the lower T that follows it, hence it remains in situ. As no movement is overt nor covert, it means that none of Comp features are erased; in the light of Head Strength Parameter and the consideration of all CPs as interrogative (Torrego and Pesetsky 2001), it is evident that some of embosi Comp are weak heads and TPs with a null spellout of Comp.

One of the arguments that leads to the consideration of yes-no questions as CPs are issued from the reported speech of yes-no question where an interrogative word 'whether/if is used. It will appear from the next examples that even the presence of an interrogative words does not suffice to trigger any movement of T element. Consider:

(4) a. wa a-duu: 'bini le-pera osunga nga' him/her s/he-ask you you-FUT help me 'S/he said "You can help me"'

a. 1 Wa a-duu waa bini le-pera osunga nga him/her s/he-ask whether you you-fut help me 'S/he asked whether you could help me'

b. edii wa a-dua a-s[??]l[??] Ngala? when-be him/her s/he-said you-know Ngala 'When s/he said "do you know Ngala?"'

b.1 edii wa adua waa is[??]l[??] Ngala, Nga is[??]ri [??]h n[??] kaa when-be him/her he/she-said whether I-know Ngala, I said yes you not 'When he asked whether I know Ngala, I said I do, you said you don't.'

C. no o-yaa [??]ioo kaa you you-come or not 'you will come or not'

c.1 i-di opera osere waa i-yaa [??]oo i-di oyaa ko I-be can say whether I-come or I-be come not 'I cannot say whether I will come or not'

As (4) indicates it, the yes-no questions can be introduced by an overt interrogative word in bold. This argument subtends that although yes-no questions have no Q operator, their reported form resorts to the insertion of an overt Q item. As a matter of fact, we can slightly be reformulated our previous postulate to the extent that embosi yes-no questions are CPs with a null Q item. The main characterisation of these CPs is the incapacity of C features to constrain the movement of the lower item bearing its features so as its features are erased. As a result, the presence of the overt Q item does not suffice to trigger the movement of T to C since these types of sentences have a genetic property of weak head C.


African languages offer a real wealth of linguistic investigations as it is not an easy task to advocate formalism amongst these languages. What is proved perfect in one language can be proved pointless in another. In fact, an analysis based upon a Nigerian language Ao offered by Oye Taiwo (2003) stands as a counterpoise to smbosi examples. It results from this study that Ao has four yes-no questions markers namely NZe, se, pa...rin and parin. The first two markers are sentence initial words and can be used interchangeably. The following are examples (2) taken from Oye Taiwo (2003: 41)

(6) a. Nze/Se Taye o ze udon? QUES.MKR Taye agr eat meat 'Did Taye eat meat?'

b. Nze/Se aza a gbo in? QUES.MKR dog fut bark you 'Will the dog bark at you?'

As for pa...rin and parin, it is argued that the latter is a sentence final word whereas the former appears in a discontinuous way; that is, the first particlepa is immediately placed after the NP subject, while the second particle rin appears as the complement of the verb. In other words, we have the structures NP -pa and VP rin as follows.

(7) a. Wo ya parin? you.SG come QUE.MKR 'Is it that you come?'

b. Wo pa ya rin? you.SG QUES.MKR come QUE.MKR 'Did you come?' (Oye Taiwo, e.g. 10: 44)

The next section is devoted to the analysis of yes no questions together with their answers. 3. Yes-No questions and answers

As has been said so far, there is no difference in structure proper to yes-no questions. Consider:

(8) a. n[??] o-dii ote ba? you you-be see them 'Have you seen them?'

[??]h, i-dii i-te ba yes, I-be I-see them 'Yes, I have seen them' (Yes, I have)

Ehehe, i-dii ota ba ko no, I-be see them not 'No, I have not seen them' (No, I have not)

b. bin[??] andzoro ma-kya bwa? you bodies they-make pain 'Are you ill?' sh 'Yes, I am' kaa 'No, I am not'

c. Okombi la-h[??] koyo? Okombi he-speak koyo 'Does Okombi speak Koyo' la-ho 'he speaks' he-speak 'Yes, he does' a-li oh[??]] y[??] 'he does not speak' he-be speak not 'No, he does not'

If there were no yes-no markers in the previous sections, it is evident from (8) that there are yes-no answers devices in embosi. They are eh 'yes' with a long open vowel and ehehe ou ka (so, yo, kale, te, ko, kaa) 'not'. Of interest is the fact that yes-no answers are initial sentence words. In addition, there are three choices which govern the use of these yes-no answers markers.

(9) i-Yes-no markers followed by a whole clause ii- Yes-no markers alone without a clause, and finally iii- A finite predicate in case of affirmative

In all evidence, 9(i) represents the ideal prototype for yes-no answers in embosi. This is the citation form. And for the economy of representation, this citation form can be subject to some reductions to the extent that there is only the yes-no marker or the clause. Again, all these operations take place at the S-structure; they do not affect something on the underlying structure. If it were asked to compare embosi yes-no answers with another language, it should be claimed that there is similarity with the French ones. Compare:

(10) a. Tu vas a l'ecole? No odzwa lek[??]li? Are you going to school?

Oui, j'y vais [??]h, i-dzwa 'Yes, I go.

b. Elle est a la maison? Wa a-di o ndai? Him/her s/he-be at home Is she at home ?

Non, elle n'est pas. kaa, a-di y[??] Not, s/he-be not No, she is not'

Next, it appears that the Neg marker appears in both sentence initial and final position. An explanation to this fact is found in the category of the Neg marker. In fact, there are two kinds of Neg item namely ehehe and kaa. The first Neg item is mainly used in yes-no questions, while the others are used in negative sentences and yes-no questions. The use of ehehe at the end of a sentence leads to deviancy. All this means that ehehe is an initial Neg marker for yes-no questions only. The second category of Neg items can be initial and final without bringing out no ungrammaticality. Basing upon this argumentation, it should be borne in mind that all Neg markers can apply to the first two choices elaborated so far except ehehe.

Moreover, the third principle that accounts for the distribution of yes-no answers put forward in (9) need to be revisited. In this effect, it should include non finite predicate in case of negative sentences. In fact, the negation in embosi implies the use of Neg item preceded by a discontinuous morpheme ' o' which is prefixed to the verb that is negated. That discontinuous morpheme renders the verb infinite as there is no agreement between the external verb argument and its predicate. This is illustrated from (10c) in its negative answers. As a result, 9 (iii) will now be read as ' finite or non finite predicate in case of negation'.

In the following lines, we want to analyse yes-no questions in relation to agreement and disagreement where too, either, neither and so are used in the expressions. Let's consider these:

(11) a. nga la-linga swe me I-like fish 'I like fish'

Nga he Me too 'Me too'

b. wa a-[??][??] Emb[??]si him/her s/he-speak embosi 'S/he speaks embosi'

b.i la mwana ya wa and child of him/her 'so does his/her child'

b.ii- ndzani nga ka but me not 'But, I don't'

c. n[??] o-baara ka wa la-yaya you you-think that him/her s/he-come 'You think that s/he will come'

c.i nga i-ta mbi la-yaya me I-see indeed s/he-come 'I think so'

c.ii nga [??]i li-baara bungu me too I-think so 'So do I'

c.iii nga i-ta mbi a-li oyaya yo me I-see indeed s/he-be come not 'I don't think so'

d. Nyanga a-di odzwa ko Nyanga he-be go not 'Nyanga does not go'

d.i nga [??]i me too 'Me too'

d.ii la wa [??]i and him/her too 'So does her/him'

It results from (11) that the agreement expressions are made of a complement pronoun which is coupled with an adverb. One of the particularities of this kind of structure lies on its infinite boundary regarding the type of the predicate involved. It transpires from 11(b) that the adverb and the complement pronoun are not used.

One argument in support of this structure derives from the fact that, in addition to the use of the adverb and the complement pronoun; we can resort to some other structures where there occurs a linking word followed by an NP item. It is worth admitting that such a structure is a reduced form where the adverb has been, for the economy of speech, reduced at PF level. 11 d(ii) is a key illustration of this structure.

The agreement structure will be made of the gloss X too where the X variable can be noun or a complement pronoun. There are two interesting remarks to be echoed in 11 (c). Firstly, there is another adverb namely mbi 'indeed, really' which is used to express agreement. The difference between mbi and [??]i is related to their individual distribution; the former requires a long clause whereas the latter does not necessarily need it. In addition, the former occurs as the complement of the verb, while the latter occurs as the complement of a noun. Otherwise mbi appears in the structure VP--, while [??]i appears as in NP--.

Secondly, it appears in 11 c(ii) that two agreement markers are present in the same clause. If we proceed by eliminating one of the two markers, something will matter undoubtedly. In case that the first agreement marker is dropped, the sentence will be read as 'I think like that' which is quite close to 'I think so'.

But the dropping of the second one will read as follows 'I am also thinking', which is not similar to 'I think so'. What makes the second reading different is certainly the presence of the predicate because this adverb seldom requires a verb in its neighbourhood. As a result, the presence of the second agreement marker is demanding as it reinforces the idea and precludes any other possible readings.

As a final remark on agreement, it should be noted that the occurrence of the adverb is very compulsory as its absence leads to some other readings.

As for disagreement, it is obvious that there is no difference with the agreement process. The same expressions which follow a negative sentence in order to express disagreement are similar to ones that follow an affirmative sentence in keeping with the structure involving pronoun and adverb. The difference can be pointed out when dealing with double markers involving the adverbs 'mbi' and 'bungu'. The use of these adverbs will lead to the presence of the Neg markers which occur in clause final position. Therefore, when it happens that there is a disagreement expression that follows an agreement one, there should be a change, that is, the adverb following the pronoun will be substituted by any Neg marker. Equally interesting is the fact that it is not surprising to have the adverb 'pi' in such constructions as follows:

(12) Andz[??]l[??] a-dzaa bea la iyele Andzele she-like food at morning 'Andz[??]l[??] usually eats all the mornings'

i. li nga ka but me not 'But I don't'

ii. la bana [??]i y[??] and children too not 'And children don't'

On a purely structural way, the agreement and the disagreement will look like the following:

(13) Complement pronoun + adverb

The next section is an analysis of the interrogative sentences in order to determine whether all CPs are weak heads and hence explaining the in situ position of its C items in their lower position as the complement of VP.

4. The structure of interrogative Wh

In our previous analysis, it appears that Comp fails to attract the lower lexical item which bears its features. In the following, it will come out that this argument will be partly rewritten.

(14) a. kombo la n[??] nde? name of you what 'What is your name?'

b. nde a-boml Ngwa[??]i who s/he-kill Ngouabi 'Who kills Ngouabi?'

c No o-beri ba mbi you(sg) you-beat them how 'How do you beat them?'

d ye mb[??]ng[??] yamba nde? this money of what 'Whose money is this?'

e. bini le-yee [??]a tina nde? you.PL you-come here reason what 'Why do you come her?'

f. nyek[??] a-wuru pe? grandparent s/he-come where 'Where does the grandparent come from?'

g. mwana a-bor-im-a [??]k[??] nde? child s/he-pasv-born day what 'When was the child born?'

The following stand as counterexamples of yes-no questions presented so far in that they show real question markers. In Ndongo Ibara (2009), it has been argued that the interrogative pronouns are different from the relative pronoun hence the latter are [-Wh]. That is to say, the relative pronouns are bound morphemes which are affixed to the prefix of the verb and cannot be separated from them. In addition, the relative pronouns which refer to [+human] and [-human] properties namely 'who' and 'which' are different from the interrogative pronouns 'who' and 'which' in embosi as it is illustrated below:

(15) a. mwasiye-bori a-di bola a nga wife who-give.birth s/he-be sister of me 'the wife who gave birth is my sister'

b. mwasiye-bori e-dii nde wife who-give.birth it-be who 'who is the wife who gives birth?'

c. mwere mo-bwe o-di ombange tree which-fall it-be mango.tree 'The tree which fell is a mango.tree'

d. mwere nde o-bwe tree which it-fall 'Which tree fall?'

Furthermore, the fact that the relative pronouns do not share the Wh properties raises the question on the clause features. In Pesetsky and Torrego (2001), Agbayani (2000), Radford (2004) among many other literary works, it has been claimed that all clauses are CPs with C bearing the features [Wh, EPP, Tns]. These features are the causes of the movement of the lower entries that bear these features because the movement is the only key that erased these features. Yet, embosi relative clause will be taken as a violation of this argumentation as the embedded process takes place at morphological level. As a matter of fact, it should be admitted that the relative pronouns are clauses lacking interrogative properties, but they have the features [EPP,Tns]. Of interest is the fact that the separation of Comp from the VP is only possible at morphological level, elsewhere the two nodes are overlapped and merged into a nutshell.

In (14), embosi instances cases of real CPs as interrogative expressions. From a lexical point of view, there is a lexical poverty of words which account for different interrogative words. This is one of the reason why the lexemes who, what, how, when, whose and why are basically coined from a unitary word nde. The difference between 'who' and 'what' can be explained in terms of the word position within a sentence. As a result, the word nde will be read as 'what' when it is in sentence final position, while it will be 'who' as a sentence initial position. Then, how, when, whose and why are formed via a lexicalisation through the use of other words. Glossary, they are manner what 'how', of who 'whose', reason what 'why', time expressions and what 'when'.

At syntactic level, the interrogative expressions have two sentences positions. There are two insightful arguments at stake here. Firstly, there are interrogative expressions which are only final sentence position words namely nde 'what', mbi 'how', and pe 'where'. We hold the view that theses questions markers never undertake any movement to appear as Spec-Comp since they are in situ Q items. In keeping with the Head Strength Parameter (a parameter whose settings determine whether a given head is strong as it triggers the movement of its lower item to attach to its features or weak as it cannot instigate the movement of correspondent item), it will be asserted that these particular C are weak heads because they are incapable of triggering the closest C items which is lower in derivation to them. They have indeed all the Comp features, but they fail to attract their correspondent items in order to valuate their features and erase them through movements.

We can formulate a further argument in support of the non raising of these interrogative expressions in relation to their lexical content. This argument is worthy for mbi and pe. In fact, mbi can de read as 'like' and pe can raised confusion with another word pe 'there'. It is certainly owing to this matter that these interrogative words are preferable to remain in situ as verb complements than undertaking raising into the Spec-Comp. The following illustrative examples contrast the two positions of mbi and pe with different readings.

(16) a. Ngakala a-di mbi Ngan[??]ng[??] Ngakala he-be like Nganongo 'Ngakala is like Nganongo'

b. pe nare la i[??][??] there seems with problem 'there should be a problem there'

c. mbi la bwa indeed with pain 'it is hard indeed'

Secondly, the other interrogative markers are initially final sentence words and they have been raised to Spec-Comp under different positions. The subject Q item nde shows a case of Spec to C movement. That is to say, nde derives from the position where the subject NP appears in a finite clause. The arrow is used below to indicate this movement.


So when the move is applied, the derivation will allow the construction of the following tree diagram as the last stage of derivation for the transfer of info to other levels of interpretations.


This derivation process will also be successful to the other compound forms of nde. The specificity of these compound groups is that nde cannot be moved alone leaving its co-concurrent word. If this happens, the immediate result will be deviancy and ungrammaticality as follows:

*(19) i. nde bana a-dzwe bo tina what children they-eat sleep reason ii. nde bini le-bomi mboo ndenge what you you-kill buffalo manner

iii. nde iboro ba-dzwa mboa oko what parents they-go village day

iv. nde ndai yamba what house of?

The above examples are ill-formed because they violate the Condition on Extraction Domains and the Functional Head Constraint because the whole constituent has not been moved but only a string of it. In addition, once the head nde is moved alone it leads to deviancy because the head cannot substitute to a specifier. And since the two words standing for a question marker cannot be separated, we will refer to Amfani's (1996) Broken Determiner Analysis. Amfani argues that since the two independent lexical items which are qualified as determiner share a unified semantic sense which requires their unification to express such a fact. Putting things in a quite different way, it means that if the two words are not together, they will not refer to the same reading.

In the case just discussed, it appeared that the separation of the question marker from the other units that constitute the question domains has lead to ungrammaticality. In this view, the collocation of nde with its accompanied words will be taken as a single set at PF level. The output of this argumentation will bring it about that the two syntactic words which form the question marker occur adjacently. Consequently the (19) examples will now become grammatical as in:

(20)i. tina nde bana o-dzwe b[??] reason what children they-go sleep 'Why do children go to sleep?'

ii. ndenge nde bini le-bomi mb[??]? manner what you you-kill buffalo 'How do you kill the buffalo?'

iii. [??]k[??]nde iboro ba-dzwa mboa ? day what parents they-go village 'When do parents go to the village?'

iv. yamba nde ndai ye of what house this 'Whose house is this?'

In fact, the Broken Question Marker Analysis which is born from the Broken Determiner Analysis will lead to the elaboration of the C structure in the following fashion [X-nde] where the variable X will refer to any words that go with nde to form a question marker. In fact, the X variable is nothing but an N item. Owing to that analysis and once all the derivation phases are over, the following tree for 20(ii) is derived where strikethrough and arrows are used to denote the position where the different words have been extracted from in the early position.


5. Conclusion

In conclusion, this paper has demonstrated that yes-no questions are similar to affirmative sentences in embosi. This similarity has been substantiated by the weakness of the C to attract the closest lexical item that bears its features. It has then been admitted that yes-no questions are CPs with an exception that their Q operator is not overt at syntactic level. As such, the Q operator has a null spellout.

Besides, the yes-no answers are characterised by the choice of adverbs and the Neg item. They are indeed clause initial words. They can occur as mid or final position words in cases of disagreement or agreement. The presence of adverbs in these answers is mandatory.

Finally, the analysis of interrogative expression proves that there is a difference between a relative and an interrogative pronoun in Embosi. The former is phonologically and morphological weak as it is bound to the verb prefix, whilst the latter is phonologically and morphologically free. Hence, we claim that relative clauses are exceptional CPs which lack [Wh] features of Comp. From a syntactic point of view, there appears to stand as in situ question markers and those which can undergo movement toward the Spec-Comp position. As the long list of question markers points out cases of a couple of words made of an X variable and nde, we have referred to Amfani's Broken Determiner Analysis to account for the syntactic representation of these kinds of question markers.


AMFANI, A.H. 1996. Aspect of Agreement Relation in Hausa Clause Structure. PhD Thesis, University of Ibadan, Nigeria. 1996.

AGBAYANI, B. 2000. Wh-subjects in English and the vacuous Movement Hypothesis. In Linguistic Inquiry 31, Cambridge, Massachusetts: MIT Press, 2000, pp: 703-713.

CHOMSKY, Noam.1995. The Minimalism Program. Cambridge, Mass.:MIT Press, 1995.

CHOMSKY, Noam. 2000. Minimalist inquiries: the framework. In Martin, R., Michaels, D., and Uriagereka, J. (eds.), Step by step. Cambridge, Massachusetts: MIT Press, 2000, pp. 89-155.

CHOMSKY, Noam. 2001. Beyond Explanatory Adequacy. ms, MIT, 2001.

NDONGO IBARA, Yvon Pierre. 2009. A Comparative Study of Complements in Embosi and English. PhD, Universite Marien Ngouabi, Congo. 2009.

PESETSKY, D. and TORREGO, E. 2001. T to C movement: causes and consequences. In M. KENSTOWICZ (Eds) Ken Hale: A Life in Language. Cambridge, Mass.: MIT Press, 2001, pp: 355-426.

RADFORD, Andrew. 2004. Minimalism Syntax: Exploring the Structure of English Cambridge: CUP, 2004.

TAIWO Oye. 2003. Yes/No Questions in Ao. In SKASE Journal of Theoretical Linguistics, Vol 3,no3, 2003, pp: 40-58.


Departement de Langues Vivantes Etrangeres

Universite Marien NGOUABI

Brazzaville, Congo
COPYRIGHT 2011 The Slovak Association for the Study of English
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ibara, Yvon-Pierre Ndongo
Publication:SKASE Journal of Theoretical Linguistics
Article Type:Report
Geographic Code:6CONG
Date:Dec 1, 2011
Previous Article:Human picturing of birds--the case study of zoosemic developments in the category female domesticated birds.
Next Article:Genitive-dative syncretism in the Balkan Sprachbund: an invitation to discussion.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters