dotawo nubian studies vol 2 e-issue

8/18/2019 Dotawo Nubian Studies Vol 2 E-Issue

1/310

i

Dotawo ▶A Journal of Nubian Studies

2015 #2


2/310


3/310

Dotawo ▶A Journal of Nubian Studies

2015 #2

Edited byAngelika JakobiGiovanni RuffiniVincent W.J. van Gerven Oei


4/310

Dotawo ▶ A Journal of Nubian Studies

Editors-in-Chie Giovanni Ruffini Vincent W.J. van Gerven Oei

Editorial BoardAbdelrahman Ali Mohamed Julie Anderson Anna Boozer

Intisar Elzein Soghayroun Angelika Jakobi Anne M. Jennings Robin Seignobos Ahmed Sokarno Abdel-Haz Jay Spaulding Alexandros Tsakos Kerstin Weber-Thum Petra Weschenfelder

Design Vincent W.J. van Gerven OeiTypeset in 10/12 Skolar , with Adobe Arabic, Antinoou,Lucida Sans Unicode, and Sophia Nubian.

Cover image Courtesy of Sam Gerszonowicz,Nubian Image Archive

https://www.ickr.com/photos/132357696@N07/

Editorial correspondenceGiovanni Ruffini ([email protected])Vincent W.J. van Gerven Oei ([email protected])

: http://digitalcommons.faireld.edu/djns/

For submission guidelines please see our website.

-13 978–069245843–3-10 0692458433 2373-2571 (online)

Dotawo: A Journal o Nubian Studiesis published once a year byDigitalCommons@Faireld & punctum books, Brooklyn, , undera Creative Commons Attribution–NonCommerical–NoDerivs 3.0Unported License.

: http://www.punctumbooks.com© 2015 by the editors and authors.


5/310

Dotawo ▶

1. A medieval Nubian kingdom controlling the central Nile Valley,best known from Old Nubian documents excavated at Qasr Ibrim

and other sites in Lower Nubia.2. An open-access journal of Nubian studies, providing a cross-disciplinary platform for historians, linguists, anthropologists,archaeologists, and other scholars interested in all periods andaspects of Nubian civilization.

1. , ϩ . - . ϣ

, , ϭϭ ,

.2. , ϣ ϣ .

, . , , ϭϭ ϭϭ . , , ,

, . ,, , ϭϭ .

, , .*

1. Ammiki Nuba-n sirki Tungula-n Bahar aal poccika anda kannim,ne poccika an ammikin Nuba-n kitaaba an Kasr Ibrimiro poonisshi Nuba aro-n ammiki ir kar əəl koran əəllooyanero poccikareəəl oddnooyim.

2. Ele ne Nuba poccikan muɟallayane, aal poccika yaa əərngaanyatn,taariikiro, aallo, elekon poon ammik(i) ir ayin ir kanniyam pirro,poon ammik(i) aallo, elek(i) aallo poccikaa yaa əərngaanyatn.**

* Translation into Nobiin courtesy of Mohamed K. Khalil.** Translation into Midob Nubian courtesy of Ishag A. Hassan.


6/310


7/310

Dotawo ▶ A Journal of Nubian Studies2015 #2 Vincent W.J. van Gerven Oei

Old Nubian Relative Clauses 9 Mohamed K. KhalilThe Verbal Plural Marker in Nobiin (Nile Nubian) 59

Angelika Jakobi and El-Shae El-GuzuuliRelative Clauses in Andaandi (Nile Nubian) 73

El-Shae El-GuzuuliThe Uses and Orthography of the Verb “Say”in Andaandi (Nile Nubian) 91

Ahmed Sokarno Abdel-HazFocus Constructions in Kunuz Nubian 109

Abeer BashirAddress and Reference Terms in Midob (Darfur Nubian) 133

Waleed Alsharee The Consonant System of Abu Jinuk (Kordofan Nubian) 155

Gumma Ibrahim Gul anPossessor Ascension in Taglennaa (Kordofan Nubian) 171

Ali Ibrahim and Angelika JakobiAttributive Modiers in Taglennaa (Kordofan Nubian) 189

Thomas Kuku Alaki and Russell NortonKadaru-Kurtala Phonemes 215

Khaleel IsmailTabaq Kinship Terms 231

Khali a Jabreldar Khali aAn Initial Report on Tabaq Knowledge and Prociency 245

Angelika Jakobi and Ahmed HamdanNumber Marking on Karko Nouns 271Grzegorz Ochała and Giovanni RuffiniNubische Berichtigungsliste (1) 291


8/310


9/310

From the Editors

We are proud to present the second volume of Dotawo: A Journal oNubian Studies. This journal offers a multi-disciplinary, diachronic

view of all aspects of Nubian civilization. It brings to Nubian stud-ies a new approach to scholarly knowledge: an open-access collabo-ration with DigitalCommons@Faireld, an institutional repositoryof Faireld University in Connecticut, , and publishing housepunctum books.

The rst two volumes ofDotawo have their origins in a Nubianlanguage panel organized by Angelika Jakobi within theNilo-Saha-ran Linguistics Colloquium held at the University of Cologne, May 22to 24, 2013. Since many invited participants from Sudan were un-able to get visas due to the shutdown of the German Embassy inKhartoum at that time, the Fritz Thyssen Foundation funded theorganization of a second venue of specialists on modern Nubianlanguages. This so-called “Nubian Panel 2” was hosted by the Insti-tute of African & Asian Studies at the University of Khartoum onSeptember 18 and 19, 2013. This volume publishes the proceedings ofthat that panel. We wish to extend our thanks both to the Fritz Thys-sen Foundation and to Professor Abdelrahim Hamid Mugaddam, thethen director of the Institute of African & Asian Studies, for theirgenerous support.

Future volumes will address three more themes: 1) Nubian wom-en; 2) Nubian place names; 3) and know-how and techniques inancient Sudan. The calls for papers for the rst two volumes maybe found on the back of this volume. The third volume is alreadyin preparation with the assistance of Marc Maillot of the Sectionfrançaise de la direction des antiquités du Soudan ( ), Depart-ment of Archeology. We welcome proposals for additional themedvolumes, and invite individual submissions on any topic relevant toNubian studies.


10/310


11/310

9Old Nubian Relative ClausesVincent W.J. van Gerven Oei*

. Introduction

In this article, I venture to offer an in-depth analysis of the struc-ture of Old Nubian relative clauses (henceforth, s), in an attemptto reorganize and consolidate the observations made in Gerald M.Browne’sOld Nubian Grammar,1 Helmut Satzinger’s earlier article“Relativsatz und Thematisierung im Altnubischen,”2 and MarianneBechhaus-Gerst’sThe (Hi)story o Nobiin.3 Satzinger’s article, ourrst extensive source for the grammatical analysis of Old Nubian

s, was written in response to a series of grammatical observationsby Browne inStudies in Old Nubian.4 However, this response wasunfortunately never fully incorporated intoOld Nubian Grammar,5 where Browne discusses s in §§4.4–6.6 In these few dense andsomewhat confusing paragraphs, Browne organizes Old Nubian smainly based on word order, without clearly marking out syntacti-cal relations, thus losing much of the insights of Satzinger’s morestructured approach. Bechhaus-Gerst offers the most recent analy-sis of Old Nubian constructions inThe (Hi)story o Nobiin, in anattempt to integrate the approaches of Satzinger and Browne, butremarks that “a thorough analysis […] would go beyond the scope,”of her study.7 During the preparation of this article we have alsoconsulted comparative material from related contemporary Nile

* I would like to thank Issameddin Awad, Angelika Jakobi, and Giovanni Ruffini for theircomments and suggestins during the various stages of writing this article.

1 B ,Old Nubian Grammar.2 S , “Relativsatz und Thematisierung im Altnubischen.”3 B -G ,The (Hi)story o Nobiin,esp. pp. 207–11. Glossing has been occasionally

adjusted to match the set of abbreviations listed in fn. 16.4 B ,Studies in Old Nubian.5 Browne refers to Satzinger’s approach as a “somewhat different orientation” (B ,Old

Nubian Grammar, p. 83, fn. 99).6 Browne further combines relative constructions with temporal and adverbial subordinateclauses inOld Nubian Grammar, §4.7, which I will not consider here. However, I touch upon

conditional and nal clauses in “A Note on the Old Nubian Morpheme - in Nominal andVerbal Predicates.” See also B -G ,The (Hi)story o Nobiin, pp. 105–10.

7 B -G , ibid., p. 207.

V G O , Vincent W.J., “Old Nubian Relative Clauses.”Dotawo 2 (2015): pp. 9–57.


12/310

10

Van Gerven Oei

Nubian languages8: the dissertation of Isameddin Awad on subordi-nation in Nobiin [a],9 an article by Abdel-Haz Sokarno for Kenzi/Kunuz Nubian data [xnz],10 and recent work by Angelika Jakobi andEl-Shae El-Guzuuli on s in Dongolawi/Andaandi [dgl].11

The present paper, glossing and methodically expanding theanalyses of the examples adduced by Browne and Satzinger, aims toconsolidate the sometimes divergent interpretations offered by Satz-inger, Browne, and Bechhaus-Gerst, in an attempt to integrate OldNubian s in a general syntactic framework and to harmonize theiranalyses with insights from contemporary syntactic theory and com-parative material from closely related Nubian languages. This willhopefully allow us to make ner distinctions between the differenttypes of Old Nubian s and discuss several pertinent overarchingthemes, such as le ward movement and extraposition, which owingto relatively marginal penetration of contemporary syntactical theo-ry in Old Nubian studies have so far received little attention.

. Brief overview

We will start with an overview of attributive s in §3, divided be-tween coreferential (the subject of the is coreferential with theantecedent of the ) in §3.1 and non-coreferential (the subject ofthe is not coreferential with the antecedent of the ) in §3.2.Non-coreferential s are further subdivided into those with overtsubjects (§3.2.1) and those without overt subjects (§3.2.2). As wewill see, the presence or absence of an overt subject inuences themorphology of the verb in the . For both coreferential and non-coreferential s I will also discuss exceptions to the general patternin which s are seemingly postnominal. Section 3.1.1 treats corefer-ential s that have moved le ward and only allow for a restrictivereading. In §3.3.1 we will treat several exceptions with non-corefer-ential s that seem to be generated prenominally, and s of time,place, and manner (§3.3.2). Finally, §3.4 deals with constructions inwhich non-coreferential s show an anaphor coindexed with theantecedent. Free s, those without an overt antecedent, are treatedin §4 according to their grammatical function in the main clause,starting with subject clauses (§4.1) and object clauses (§4.2), whichalso include different types of complement clauses (§4.2.1–2). Sec-tion 4.3 deals with free s in other, secondary positions. A specicsection (§5) is devoted to s in combination with the so-called pred-

8 See R ,Le méroïtique et sa amille linguistique, p. 165, for a brief discussion. Languageabbreviations follow the 639–3 standard employed byEthnologue.

9 A ,The Characteristic Features o Non-Kernel Sentences in Nobiin. I have adjusted some ofAwad’s Nobiin orthography based on suggestions by Angelika Jakobi.

10 A -H , “Nubian Relative Clauses.”11 J & E -G , “Relative Clauses in Andaandi.”


13/310

11

Old Nubian Relative Clauses

icative suffix - , including nominal predicates (§5.1), complex ver-bal predicates, also called periphrastic constructions (§5.2), s invocative or appellative contexts (§5.3), and adjunctive/appositionalclauses, which usually feature only a bare - suffix without any ad-

ditional tense morphology (§5.4). Sections 5.5–7 deal with a series ofmore complex syntactical constructions, including the topicaliza-tion through - of the antecedent of an (§5.5), s in the scope ofquantiers (§5.6), and nally quantier raising through - (§5.7).Section 6 treats with two different types of extraposition, in which(part of) the appears to have moved to the right of the clause,motivated by the heaviness of the . We have distinguished twotypes of extraposition, depending on whether the non-coreferential

shows verbal agreement (§6.1) or not (§6.2). Finally, §7 deals withpreterite tense morphology in s, which appears to be distributedaccording to whether the is coreferential or not. An concludingoverview is given schematically in §8.

. Attributive relative clauses

Attributive s are full clauses showing tense morphology, and mayfeature an overt subject and be introduced by a relative pronoun.An is embedded with the main clause, connected through an an-tecedent that has a syntactical function both in the embedded andin the main clause. If the subject of the coincides with its ante-cedent, we speak of a coreferential attributive . If this is not thecase, the attributive is called non-coreferential. The distinctionbetween coreferential and non-coreferential attributive s in OldNubian is reected in the syntax, whence Satzinger labels coref-erential attributive s as “Type A,” and non-coreferential ones as“Type B.” Bechhaus-Gerst broadly follows Satzinger’s categoriza-tion, whereas Browne makes no descriptive distinction between thetwo types, lumping both under the heading “adjectival conversion.”12

Old Nubian is an language, like Japanese, Turkish, Dutch, orthe other Nile Nubian languages.13 This generalization allows us tomake several predictions about its general syntactic structure. Firstwe expect all phrasal heads to align on the right side. This seems tobe generally the case when we inspect verb inection, which con-sistently appears on the right edge, and nominal inection (case anddeterminer). Also note the fact that Old Nubian has postpositionsinstead of prepositions. This generalization implies that any syn-tactical construction that on the surface does not follow this general

principle will be suspected of movement. Within Nubian languages,whose structure is generally very stable and predictable,14 the12 B ,Old Nubian Grammar, §4.6.13 Ibid., §4.9.1.14 Cf. A , “Noun Phrase Construction in Nubian Languages.”


14/310

12

Van Gerven Oei

neutral or default position for attributive s, whether coreferen-tial or non-coreferential, seems to be postnominal. As we will seebelow, the majority of Old Nubian examples in the extant literatureshows this order, which is corroborated by contemporary Nile Nu-

bian languages.15

man [rel buru ir-iin doll-ee] ii. girl 2 - love- 1

tan-jutti-li3 . -niece- 2. .316“The girl whom you love is his niece”

tod [rel een-gi jom-e-l] nog-s-uboy woman- hit- - go- -3“The boy that hit the woman le ”

ogij [rel in kaa=r aag-il] man . house= live- .am-beena-n1 . -uncle- .3“The man who lives in this house is my uncle” / “The man living inthis house is my uncle”

Ex. 1a shows a non-coreferential attributive in Nobiin. The ante-cedent man buru does not correspond to the subject of the , whichis the genitive-marked subjectiriin.17 The entire subject is, as wouldbe expected in an language, marked at its le edge by rst thecomplementizer -eeand then the nominative case markerii. Ex. 1b

15 List of sigla: : A ,The Characteristic Features o Non-Kernel Sentences in Nobiin; :B -G ,The (Hi)story o Nobiin; M.: V G O & E -G ,The Miracleo Saint Mina; : A -H , “Nubian Relative Clauses”; : B ,Old NubianGrammar; P. QI 1: P & B ,Old Nubian Texts rom Qasr Ibrim ; P. QI2: B ,Old Nubian Texts rom Qasr Ibrim; P. QI 4: R ,The Bishop, the Eparch, and the King; : J & E -G , “Relative Clauses in Andaandi”; : S , “Relativsatz undThematisierung im Altnubischen.” Other sigla follow , §0.3.

16 List of abbreviations: 1, 2, 3 – rst, second, third person; – accusative; – affirmative;app –apposition; – coordinator/conjunction; – causative; – clitic; –comitative; – complementizer;comp– complement clause; – completive;

– consuetudinal; – copula; – dative; – demonstrative pronoun; – determiner; – directive; – distal; – emphatic clitic; – exclusive; – nal; – focus; – future; – genitive; – human; – imperative; –inchoative; – inclusive; – interrogative pronoun; – imperfective; – juncturevowel; – locative; – negative; – nominative; – nominalizer; – nounphrase; – passive; – plural; – pluractional; – predicative/predicate; – present; – past; 1 – preterite 1; 2 – preterite 2; – participle; – possessive;

– proximal; – question marker; – direct speech marker; – reexive; – relative pronoun;rel– relative clause; – same-subject converb;t – trace; – transitive; – vetitive; – vocative.

17 (Nearly) all subjects in non-coreferential relative clauses in Nile Nubian are marked withthe genitive case, see §3.2.

1a 497

Nobiin

1b 497

Kenzi

1c 2

Andaandi


15/310

13


shows a coreferential attributive in Kenzi, with the markedby the complementizer-l. In both examples, the follows the an-tecedent. In both Kenzi and Nobiin, the marker (-eein ex. 1a;-l inex. 1b) changes based on whether the is coreferential or not. As

Old Nubian s do not feature any distinctive marking on the rightedge, such morphological variation has not been observed.18In strictly languages such as Turkish and Japanese, s, like

all other modiers, are prenominal.19 In order to account for s thatseem to be postnominal in languages, such as Dutch and Nubianlanguages, the so-called head-raising analysis of s posits that an-tecedents originate within the and move le ward to a positionpreceding the ,20 leaving a gap or trace in the , marked byt.21

[rel tod eengi jomel

] →todi

[rel

ti eengi jomel

]

The movement illustrated in ex. 2 itself is subjected to certain con-straints, as we will nd in §3.4. For the remainder of this paper wewill mostly assume this movement, and for reasons of simplicitynot indicate it in the examples unless necessary. In certain Nubianlanguages, including Old Nubian, s can also appear prenominally.The motivation here is always semantic. Whereas Abdel-Haz doesnot provide any other type of constituent order, Awad provides uswith examples of attributive s that precede the antecedent.

[rel ir-iin doll-ee]-n buru ii2 - love- 1- girl

tan-juti-li3 . -niece- 2. .3“The girl you love is his niece”

The postnominal in ex. 1a differs in two aspects from the pre-nominal in ex. 3. First, the is marked by the genitive case, sug-gesting that it has moved into the position where normally the gen-itive-marked possessor would appear; second, Awad indicates thatburu can no longer be preceded by the demonstrativeman,suppos-edly because all possessed nouns are by denition determinate. Wewill see in §3.1.1 that also in Old Nubian, le ward movement of a is accompanied by different morphology in the and is motivatedby semantics. Awad also presents headless s as a third possibility,which we will discuss in §4 as free s.18 However, there seems to be a specic distribution of the two preterite tense morphemes in

attributive relative clauses. See §7.19 Cf. P ,Describing Morphosyntax, p. 327.20 See, for example, K ,The Antisymmetry o Syntax,pp. 86ff.21 Following Comrie’s classication, Old Nubian s are of the gap type. See C ,Language

Universals and Linguistic Typology, p. 151.

2= ex. 1b

3 499

Nobiin


16/310

14

Van Gerven Oei

. Core erential attributive relative clausesCoreferential attributive s without an object basically correspondto adjectival constructions with a participle,22 such as in English“the singing man,” in the sentence “the singing man walks on the

street,” which may alternatively be rendered as “the man that singswalks on the street.” The subject of the participle “singing” corre-sponds with the subject of the main verb “walks,” i.e., “man.” In OldNubian, these constructions can only be formed by means of a par-ticipial form consisting of at least a verbal root, tense/aspect suffix,and the determiner -( /) ,23 which, however, is dropped before overtcase marking. Coreferential attributive s generally appear a erthe noun, and, as a rule, number, case marking, and any other typeof right edge suffix (conjunctions, focus, etc.) only appear on theright edge of the entire noun phrase that contains the .

We nd the following general pattern for coreferential attribu-tive s:

[ … Antec- [rel … Verb-Tense/Asp]]-Det/Num/Case

is-lo pi-na [ ioudaios-gou-n ourou-ouinter- exist- .2/3 Jew- - king-[rel ounn-outak-o]]-lbear- - 1-“Where is the born king of the Jews?”

The , formed by the single embedded verb - “born,”follows the antecedent noun phrase “king ofthe Jews,” which is also its antecedent.24 Note that the antecedent

- ends in what I usually refer to as a juncturevowel, whereas Satzinger calls it an “Appositiv”) and Browne the“annective,” “an anaptyctic juncture vowel (‘Bindevokal’) insertedbetween two words that closely cohere.”25 The same juncture, orepenthetic vowel may be observed in adjectival constructions andshould not be confused with a case ending such as accusative or gen-itive, as it is purely a noun phrase-internal phenomenon.26

22 Cf. S , “Relativsatz und Thematisierung im Altnubischen,” p. 186. Browne andSatzinger refer to participles a “verbids.”

23 See V G O , “The Old Nubian Memorial for King George,” pp. 256–62. The precisedistribution between the vowels and when following a consonant is still uncertain. InAndaandi the different vowels indicate perfective and imperfective aspect, and this may alsobe the case in Old Nubian. See J & E -G , “Relative Clauses in Andaandi,” p. 91.

24 I have le nominative case marking, which is - in Old Nubian, unglossed throughout.25 S , “Relativsatz und Thematisierung im Altnubischen,” p. 186 et passim; B ,

Old Nubian Grammar, §3.6.5.26 The same juncture vowel appears sometimes on personal pronouns, e.g. ex. 32. Its precise

distribution, which seems to be of a morphosyntactic nature, has not yet been adequatelydescribed.

4

5L. 113.5–6

§4.6a


17/310

15


eu-a-sin ein-na-sin [ till-ou[rel añ]]-infear- - be- .2/3 . - god- live-ei-la deiŋar-a touri

hand- fall. - ?“(For) it is to fear to fall into the hands of the God who lives.”

In this example, in which the nal word remains unaccountedfor,27 the under the antecedent consists of a single verb

-, without overt tense marking (and therefore by default presenttense) and no determiner - due to the presence of the genitive casemarking at the end of the noun phrase -, which is attribu-tive to : “into the hand(s) of the living God.”

[ ?] [ei [rel man tauk-lo doull-aŋ-ad]]-il-gou-lman . time- exist- - - - -meijr-a-gou en-d-immanadisobey. - - be- - .3“The men who will come into being in that time(?) will bedisobedient”

Satzinger is correct to interpret the in the not as a relativepronoun, but rather as a deictic element referring to the emendednoun - “time.” He suggests, contra Browne,28 that overt com-plementizers only appear in non-coreferential attributive clauses,which seems to be conrmed by our survey of the extant Old Nubianmaterial. In ex. 7 we again nd all nominal inectional material onthe right edge of the noun. The double determiner before and a erthe plural suffix is a common occurrence (see also exx. 8, 34, 35, 49,72, 74).29 Also note the truncated predicative plural - , where wewould, according to Browne, expect - .30 Perhaps it was droppedbecause of the initial - of the copula. In the lines following thisexample, the verb is repeated several times with differ-ent adjectival predicates marked by predicative -, except K. 23.4

- “ungrateful,” ending with the privative adjective marker- , which seems to be directly connected to the verb, and the ir-regular predicative plural of K. 23.8–9 “liars.”

The case marking in the embedded phrase is not always complete:

27 In B ,Old Nubian Dictionary, p. 59, Browne gives the Greek gloss φοβερὸν τὸ ἐμπεσεῖνfor […] . It is possible that we are dealing here with a complementizer.Angelika Jakobi (p.c.) suggests that it is an unknown form of the verb - “to enter,”contributing to the meaning “fall into.”

28 Cf. B ,Old Nubian Grammar,§4.6.29 See also V G O , “The Old Nubian Memorial for King George,” p. 260.30 B ,Old Nubian Grammar, §3.5.2.

6K. 33.5–7

1

7K. 22.14–23.2

13


18/310

16

Van Gerven Oei

ϣ ·

[aggelos-ou kolot-ou[rel potot-i kolotangel- seven- trumpet- seven

kon-l-o]]-l-gou-ll-on tek-k-onohave- - 1- - - - 3 - -medd-il-ŋ-is-ana potot-ka ouš-enouaready- - - 2-3 . trumpet- sound- .2/3“And the seven angels who had seven trumpets readied themselvesto sound the trumpet”

The attributive in this example clearly shows how the juncturevowels basically appear inside the noun phrase as a placeholderwhere we would otherwise expect number and case marking toappear, which, however, no matter how heavy the noun phrase, al-ways appears only at the right edge of the . The

- has two peculiar features. First, the accusative case marking- that we would expect on has been dropped (cf. ex. 20, be-low). Perhaps this is a scribal error because the author conated itwith the initial kappa of the verb, or perhaps we should interpretthe absence of the accusative case with the verb “to have” as an in-termediate stage between - as an active verb with an object in theaccusative case, and - as an adjectival suffix, which is well attestedelsewhere.31 The second curiosity is the appearance of the determin-er - behind the verbal root - and before the preterite 1 suffix - (Browne marks it with a “sic”). Although a determiner in this posi-tion has been regularly attested in case of, for example, modal suf-xes (e.g. - - - in the above example), its appearance directlypreceding a tense suffix is exceedingly rare.

. . Lefward movement: semantic restrictionWhereas in exx. 5–8, the followed the antecedent, it may alsoprecede it, as in Nobiin ex. 3. There are a few examples present inthe Old Nubian corpus that show this inverse order; as is clear fromthe morphology, we are dealing here with a type of le ward move-ment that seems to be semantically motivated, and that is restrictedto coreferential attributive s. This is also suggested by the Nobiinexx. 1 and 3, where ex. 3 has a restricted meaning.32

[ … [rel … Verb-Tense/Asp]i-Det [Antec ti]]-Det/Num/Case

Let us inspect this rst example:31 B ,Old Nubian Grammar, §3.4.2. According to Jakobi (p.c.), in Andaandi the accusative

case marker may sometimes be dropped in the context of the verbko.32 Isameddin Awad, p.c.

8P. QI 1 8.ii.8–10

12

9


19/310

17


[[rel till-ik ounn-o]i-l [ maria -n ti]]-n eigon-gille

God- bear- 1- Mary- icon- goudal-a ki-a

run?- come-“Running toward the icon of Mary Theotokos”

Here the , which supposedly translates the commonepithet “Theotokos,” has as its antecedent and subject - , whichitself is attributive to -, marked with the genitive - . Based onour observations in the previous section, we would expect a con-struction like * . In this case, however, the entire

has moved upward and supposedly adjoined to the determinerphrase. The question is what would motivate such movement. Firstof all, we may observe that in the current conguration, - isin the scope of instead of the other way around.

therefore restricts the meaning of -; we are not deal-ing here with a Mary who happened to give birth to God, but ratherwith the God-bearing Mary. The le ward movement of the maytherefore be semantically driven. Satzinger suggests that the posi-tion of these s le of the noun is comparable to adjectival con-structions such as “Saint Mina,” although the juncturevowel - cannot appear on verbal forms. This however seems to becontradicted by the existence of a separate class of examples suchas \ / “the sins that I said” (ex. 20) and

“Christ whom you sent” (ex. 21), which clearlyshow a juncture vowel instead of a determiner (see §3.3.1 below).The movement observed in ex. 10 should therefore have an analysisdistinct from adjectives or s ending in a juncture vowel.33

The same type of movement may be observed in the following,slightly more complicated example:

:

eiskel-ad-j-amsō [ ein gad-kiññ-eibeseech- - - - .1 . esh-without-[[rel ŋook kon-j]i-il [ añel ti]] kemsō]-n-gou-ka glory. have- - living.being four- - -ouns-illa en-kouannōalove- be- .3“Let us beseech these four incorporeal, glorious animals that they

may abide in love”

33 One example given by Browne seems to resist analysis: SC 18.5 [ ] .

10M. 11.3–4

34, §4.6d

11P. QI 2 16.vii.2–6

35


20/310

18

Van Gerven Oei

If we disregard the initial material in the phrase -, and focus our attention on the phrasal element

, we nd a pattern similar to ex. 10. hasmoved le ward and adjoined to the determiner phrase. As in the

previous example, the motivation may have been semantic in na-ture, restricting the meaning of : the “glory-having creatures”instead of the non-restrictive “the creatures that have glory.” We willhave to leave the grammatical analysis of , , and -aside for the moment, but let me briey say that relative pronounsalways appear in the topmost position of the determiner phrase,and numerals such as - always appear phrase nal. Note fur-ther that the suffix - in here should be interpretedas the transitivizer - with regressive assimilation preceding thepluractional suffix - . If it had been the future suffix - , it wouldhave followed the pluractional suffix.

. Non-core erential attributive relative clausesNon-coreferential attributive s are s in which the subject of the

does not coincide with the antecedent. In Old Nubian, we can dis-tinguish two subcategories, namely those in which the subject of the

is overtly expressed (Satzinger’s “Type B1”), and those in whichit is not (Satzinger’s “Type B2”). Whereas these s behave similarlywhen in situ, we will discover that under extraposition this differ-ence becomes morphologically explicit (§6). Also differently fromcoreferential attributive s, non-coreferential ones may feature arelative pronoun, either “this” or “that.”34 Another differenceis that non-coreferential attributive s cannot move to the le ofthe antecedent, as described in §3.1.1.

. . Non-core erential with an overt subject in In case the subject of the is overtly expressed, it nearly always ap-pears in the genitive case, whereas the verb usually shows no agree-ment, therefore appearing similar to the embedded verbal formsfound in coreferential attributive s.

[ … Antec- [rel (Rel) [Subj- Verb-Tense/Asp]]]-Det/Num/Case

[ koumpou[rel ein [ tan-na kip-s]]]-il egg 3 - eat- 2-doumal doutrap añ-r-aŋ-asuddenly fowl live- - -“The egg that he had eaten suddenly coming to life as a fowl”

34 See P ,Describing Morphosyntax, p. 333. I have been unable to nd any semantic orsyntactic constraint on their distribution.

12

13M. 12.2–3

§4.6a


21/310

19


We nd here a non-coreferential attributive with a relativepronoun in the topmost position. The subject of the

-, , is marked with the genitive case, and the entire clauseis marked on the right edge with a determiner -. Note also that the

juncture vowel that we expect a er has been dropped a er- .

[ ] [ ] anktan-gou-ka ouskar-isna ter-in aei-lauaconcern. - - place. - 2.2/3 . 3 - heart-within[ oñ-in park-ou[rel ēn [ eir-in aou-s]]]-lō

tear- valley- 2 - make- 2-“He placed concerns within their heart in the valley of tears thatyou made.” (

Ps. 83:5–6)

Apart from the slightly erroneous translation of Psalm 83:5–6,which inter alia seems to omit a rendering of ἀναβάσεις and mis-interprets αὐτοῦ as , the clause itself is grammatical and theattributive construction straightforward. The with overt subject - “that you made” has as its antecedent [ ]

, ending in a juncture vowel. The embedded verb - shows noagreement marking because of the overt subject. Finally the locativemarker - is attached to the entire noun phrase on the right edge.Satzinger suggests, pace Browne, that ought to be analyzedas *aous-il-lō, with some type of regressive assimilation. However,none of the extant forms in the Old Nubian corpus suggest that thisanalysis is correct, nor that it is necessary; the -(i/e)l that Browneand Satzinger assume as the marker of the participle or “verbid”only appears in a nominative context as a determiner, and forms nointrinsic part of any “participial” morphology.

. . Non-core erential without an overt subject in In case the subject of the is not overtly expressed, we nd agree-ment marking on the main verb of the . The following patternemerges:

[ … Antec- [rel (Rel) [Verb-Tense/Asp/Agr]]]-Det/Num/Case

el-on [ mēstēr-ou[rel eik-ka ekid-rou]]-ka ou-kanow- mystery- 2 - ask- .1/2 - 1 -

pill-igr-a deñ-j-esoshine- - give.1- - .2“And now reveal us the mystery that we ask you about”

15

16St. 5.4–7

5 §4.6a

14P. QI 1 2.ii.8–10

8 §4.6a


22/310

20

Van Gerven Oei

The verb of the - shows agreement marking, as the subject “we” is not overtly expressed, and the indirect object of the di-transitive -, , is marked with the accusative case. Theverb phrase in the main clause consists of the verb

- “to shine,” which, together with the causative suffix is usuallytranslated by “to reveal.” The verb - here functions as a benefac-tive or applicative, adding the semantic role of the indirect object

“to us.”35

alesin ouetr-il keik-a pes-entruly anyone- blaspheme- say- .2/3[ proskol [rel kisse-llo ken-j-ran]]-gou-ka offering church- place- - .3 - -“If anyone blasphemes and says about the offerings that they placein the church”

Like ex. 16, we nd here a non-coreferential attributive , with theantecedent , without a juncture vowel, perhaps because weare dealing here with a loanword from Greek,36 or because of thephonologically unstable nature of -. The - hasas its subject a non-overt third person plural, as can be determinedfrom the morphology on the verb. The object of the verb - isthe antecedent .

[…]

ŋas-j-ara-sin ou-ka till-il tar-inchoose- - 1. - 1 - God- 3 -seu-ae-gar-ainherit- . - -[ iakōb-in gajjour[rel ēn [ ous-s-in]]]-dekel-ka Jacob- beauty - 2-2/3 - -“God chose us to cause (us) to be his inheritors of the beauty of Jacob which he loved” (Ps. 46:5)

This example is syntactically rather complex, as it seems that thescribe attempted to imitate the Greek word order of Psalm 46:5 inthis bilingual fragment: < >

[…] (P. QI 2 13.ii.23–26, the Sep-tuagint reads as follows: ἐξε έξατο ἡμῖν τὴν κ ηρονομίαν αὐτοῦ, τὴν

καλονὴν ᾿Ιακώβ, ἣν ἠγάπησεν). The interpretation of the sentence35 For an overview of the benefactive/applicative in Old Nubian and Nobiin, ee B -

G ,The (Hi)story o Nobiin,pp. 142–7.36 Cf. B ,Old Nubian Dictionary, p. 152.

17K. 20.15–17

, p. 83, n. 100

18P. QI 2 13.ii.24–28

7


23/310

21


depends on whether the attributive […] is interpreted as the object of or of . Browneopts for neither, translating with the rather puzzling “God choseus, making (us) his heirs and the beauty of Jacob, which he loved,”

somehow attempting to coordinate with the , ignoringthe accusative case marker. If is interpreted as the direct objectof , the entire attributive […] canonly be read as an object to the verbal root - “inherit,” with pos-sibly the nominalizing predicative plural suffix -,37 causative, andpredicative suffix. The interpretation of the attributive construc-tion itself is straightforward. The - is dependent on theantecedent , here without a juncture vowel becauseof the presence of the clause-initial relative pronoun .

In a few rare cases, we nd that both the genitive subject and per-son morphology on the verb may be overt in the :

· elon-de-eeion dekk-igir-men-dre-lo [ ŋeei-ou[rel einnow- - conceal- - - .1 . - thing- [ oun-na ai-ka ekid-rou]]]-k-on 1 . /2 - 1 - ask- .1/2 - -“And now also I will not conceal the thing that you asked me”

The , indicated by the relative pronoun has as its antecedent, ending in a juncture vowel, and the entire attributive is

the object of the verb . What is curious about the is that it features both an overt subject in the genitive case andthe verb -, with overt person morphology. Satzinger suggestsan interpretation of this double occurrence of overt subject andagreement marking by supposing a disambiguation strategy, mis-takenly assuming that is exclusively the genitive-marked pro-noun for the second person plural, which is not the case. So neitherthe overt subject, nor the overt person marking disambiguates theother; it is the indirect object that makes a reading “the thingthat we ( ) asked myself” rather implausible.

. More on the lefThere are several examples of non-coreferential attributive clausesappearing in a position that precedes the antecedent. In these caseswe are dealing with two patterns. In the rst pattern, s seem tohave been generated in situ, and are marked with the juncture vow-

el that we normally nd on the antecedent when it precedes the ,whereas the antecedent is marked with the customary number and

37 B ,Old Nubian Grammar,§3.5.2e.

19St. 6.4–7

9


24/310

22

Van Gerven Oei

case marking (§3.3.1). In these cases, the seems to appear in theposition of the adjective. The second pattern, which only occurs inthe case of s of time, place, and manner, the is either generatedin the position of the possessor, or is moved there, being marked by

the genitive case (§3.3.2). . . Preceding non-core erential attributive sAs we have seen in §3.1.1, preceding coreferential attributive sare the result of movement driven by semantics. These clauses alsoshow a specic morphological pattern, always being marked by thedeterminer - . There is, however, also a small class of examplesthat feature a non-coreferential preceding its antecedent, whichseems to be constructed in a way similar to phrases such as

, where the adjective precedes the noun.

ϩ · ·

petros-i harm-i kolotit-in tere-gou-ka koñ-j-raPeter- heaven- seventh- key- - have- - .[[rel ai ei-a pes-s]-i ŋape]-gou-ka tok-ar-a 1 say- say- 2- sins- - forgive- 1-“Peter, who has the keys of the seventh heaven, who has forgiventhe sins that I have said”

This complex example from one of Griffith’s graffiti contains severals which we will inspect in more detail below in ex. 62. Note here,

however, the attributive relative construction \ / ,the object of . It seems to be the case that the non-coreferen-tial \ / precedes its antecedent . Moreover, un-like the examples in §3.1.1, it is marked by a juncture vowel - and notby a determiner, and its subject appears in the nominative insteadof in the genitive case. Another example shows a similar pattern:

· eiar-i on[[rel eit-iss]-ou iēsous-i khristos]-ikaknow- ? send- 2- Jesus- Christ-

“And to know Jesus Christ whom you sent” ( Jn. 17:3)

Again this is an example taken from a larger, more complex sen-tence (ex. 79). But as in ex. 20 we may notice the non-coreferentialattributive preceding its antecedent . In

both cases, an explanation as in §3.1.1 seems unlikely, in the sensethat the marking of the s and antecedents follow exactly the samepattern as in exx. 5–8, albeit with and antecedent in reverse or-der. Also the fact that the in ex. 20 shows a nominative subject,

20gr. 4.1–2

38

21SC 9.13–14

§4.6a


25/310

23


suggests that we are dealing here with a different type of construc-tion. However, there is not enough Old Nubian data and an absenceof comparative data from modern Nile Nubian languages to allowfor a full explanation.

. . Non-core erential relative clauses o time, place, and mannerAs already suggested by Browne and Satzinger in their respectivetreatments of s, expressions of place and time follow a differenttemplate, in which the is marked with a genitive case and alwaysprecedes its antecedent.

[ ] { } mari-on[[rel iēsous-in dou-es]-in goul]-lo ki-enMary- Jesus- be- 2- place- come- .2/3“And when Mary came to the place where Jesus stayed”

The attributive relative construction could also be literally translat-ed as “to the place of Jesus’s staying,” and is otherwise grammaticallyunremarkable. The following examples provide further illustration:

ϣ [ ] ·[[rel ten-na dou-es]-in tauk]-a miššan-no-eion 3 - be- 2- time- all- -ouerouel-dal gittas-sanaeach.other- be.like?- 2.3 .“And all the time that they existed they were like each other”

[ ] [ ] < > ·[[rel ierousalim-ka ekkid da-san] tauk]-lo-eion Jerusalem- be.near. be- 2.3 . time- -“And at the time that they were near Jerusalem”

Note that in this example the genitive - has merged with the tense/person marker - .

[ ] [rel [ [rel tar-io joo]]-k-ka kap-es-ou]-n oukour-rō 3 - go- - eat- 2-2 - day-“On the day that you have eaten that which comes from it”38

The same strategy may be observed in Nobiin, where s of place

and time always seem to precede their antecedents, and are markedwith the genitive case. For example:

38 This example contains a free relative clause, see §4.

22P. QI 1 4.11.1–2

45

23P. QI 210.A.ii.9–10

46

24P. QI 2 14.i.1–2

47

25SC 21.3–4

§4.6c


26/310

24

Van Gerven Oei

an-uu ii [rel ir-iin kora-ga1 . -grandfather 2 - football-batar-ee]-n agar aag-iplay- 1- place stay- .3

“My grandfather is staying at the place where you play football”As in the Old Nubian examples above, the antecedentagar followsthe iriin koraga bataree-n, which is marked by a genitive. Note,however, that this pattern, which in Old Nubian is only found incase of s of place and time, seems to have been generalized in laterstages of language development until the attributive construc-tion of + genitive preceding the antecedent became a commonlyaccepted pattern for all restrictive s (cf. Nobiin ex. 3).

In the same context, Awad’s discussion of s of manner in No-biin also allows us to shed light on an otherwise obscure construc-tion in what Browne identies as the Old Nubian version of a homilyattributed to St. John Chrysostom:

· [ ] [ ]·

till-ik aurout-ka ank-imin-esoGod- alone- consider- - .2[[rel gad-aŋ-es-in]-n-a akdatt]-ou aurout-k-ende esh- - 2-3 - - order?- alone- -adinkan-gou-ka ouer-igar-es-ik-enkōboth- - one- - 2- -but“Don’t consider God alone, nor only the order in which he becameesh, but both as made one”

The syntax of this fragment is complicated, as there is only onemain verb, , with object and its apposition .In the second part of the sentence the same verb is implied, with theentire clause [ ] - as object. In the thirdpart the object of the implied verb is . - hereis technically not an extraposed (see §6), but rather an apposi-tion just like , in a construction “considerx as y.” The form

in the second part of the clause, however, seems morepuzzling. If we followed Browne and translate [ ]

- as “construction that became esh” (his Greek retrotranslationhas οἰκονομίαν for [ ] -) we would not expect to nd a corefer-ential attributive to precede with this type of complex morphol-

ogy: no agreement ought to be present in coreferential attributiveclauses (cf. §3.1.1), and the - suffix would remain completely un-explained. If we would apply our observations thus far, we wouldconclude that must be a non-coreferential attributive

26 547

Nobiin

27SC 9.18–21

41 §4.6


27/310

25


, and Browne’s translation of [ ] - as “construction” andsubject of is incorrect. Moreover, it follows none of thepatterns observed in §3.3.1; in that case we would expect somethinglike * gadaŋesi akdatt-. The only remaining option is that we are deal-

ing with some type of of place, time, or manner, which ends in agenitive and precedes its antecedent. This assumption allows us toexplain the agreement suffix - (the is non-coreferential withoutexplicit subject), the subsequent genitive - ( of time, place, ormanner), and its position in front of the antecedent. However, the - would remain unaccounted for; in exx. 22–5 the genitive case alwaysappeared as -( ) . If we observe the following example from Nobiin,however, it becomes clear that we are dealing here with a predica-tive - , which may have been preserved in Nobiin in s of manneras a same-subject converb:

man ideen an-een aaw-ee-n a kir. woman 1 . -mother do- 1- way

jelli-ga aaw-i job- do- .3“The woman does her job in the same way as my mother does”39

We may observe here that the aneen aaween-a preceding the an-tecedent kir is marked by both the genitive anda, here glossed assame-subject converb. However, recall that one of the functions ofthe Old Nubian predicative suffix - is precisely marking verbs withthe same subject. I suggest that the same is the case in ex. 27, and thatconsequently the translation of [ ] - in ex. 26 as “construction”is erroneous. In his Old Nubian Dictionary,Browne rightly observesthat we are probably dealing with some type of nominalized formwith - of the habitual verb -, or its derivative - “to set up,”which already suggests a mode or manner of doing things. The sug-gested Greek retrotranslation οἰκονομίαν suggests something simi-lar. I have therefore opted here tentatively for the neutral transla-tion “order” while suggesting that just like in Nobiin ex. 28 we aredealing with a of manner.

. AnaphorsAccording to the head raising analysis of s, antecedents of an are originally generated within the and subsequently move upto a higher (in the case of Old Nubian, le ward) position. Withoutdelving into the technical details, this movement may explain the

appearance of the juncture vowel that we mentioned before, andthe fact that the neutral position of attributive clauses seems to be

39 Cf. also A ,The Characteristic Features o Non-Kernel Sentences in Nobiin, ex. 633.

28 551

Nobiin


28/310

26

Van Gerven Oei

following the noun, rather than preceding it. It also predicts thatthere are certain syntactic positions within the , organized on ahierarchical scale,40 that do not allow for relativization, such as pos-sessor and postposition phrases. Old Nubian seems to conform this

generalization: subjects and objects can be relativized (also calledextracted), whereas other syntactic functions cannot.41 In case rela-tivization is impossible, we nd an anaphor – also called or “re-sumptive pronoun”42 – which in Old Nubian is always a form of thethird person singular pronoun .

[ ] agend-a-lō [ eiti-ou [rel ŋod-ina tani ŋape-gou-kablessed- - man- lord- 3 . sin- -

ank-ij-men-d]]

-elremember- - - -“Blessed is the man whose sins the Lord will not remember”(Ps. 31:2)

in tannan[ ogiji [rel tenni ossi. man 3 . leg

bud-s-in-tirti]]dislocated- 2-3 - .“This is the man whose leg was dislocated”

Ex. 29a follows the pattern of non-coreferential attributive clausesdiscussed in §3.2.1, but this time we nd an anaphor “his” coin-dexed with the antecedent , as possessor phrases do not allowextraction. The same occurs in Andaandi in ex. 29b, where we ndthe anaphor tenn in a similar position. Anaphors also appear in thecase of oblique phrases such as locatives:

…] · ·

-gad-d-re [ daui [rel ein [ tari-io jō-en]]]-nō- - - .1 . path 3 - go- .2/3 -“I will [guide (vel sim.) you] on the path on which you go” (Ps. 31:8)

in tannan[ beledi [rel ay teri=do ogol=lo. country 1 3 = before=

taa-s-i]]come- 2-1

40 See K & C , “Noun Phrase Accessibility and Universal Grammar.”41 Owing to a lack of examples, it is unclear whether indirect objects can be extracted, thatis, can become an antecedent to an in which its syntactical function would be that of anindirect object, e.g. “the person John wanted to give a present to.”

42 J & E -G , “Relative Clauses in Andaandi,” p. 97.

29aP. QI 1 1.i.7–8

4

29b 14a

Andaandi

30aP. QI 1 1.ii.4–5

§4.6a

30b 17b

Andaandi


29/310

27


“This is the country to which I came before”

Just like in ex. 29a, cannot be extracted from a postposi-tional phrase in the , which is the reason why we nd the loc-

ative-marked anaphor . Browne’s suggestion that is here apostponed “regular” demonstrative pronoun is erroneous; demon-strative pronouns in Old Nubian never follow nominal heads. Thesame construction can be found in Andaandi, ex. 30b. A similar ana-phor construction is found in the next example, which however can-not be faithfully rendered in English:

[ ] [ ][koeiri [rel ein [ el tari-io jōo]]]-k-katree now 3 - go- -kap-a-tamē-aeat- - .2 -“‘Do not eat that which now comes from the tree’ (lit. ‘do not eat thetree which now comes from it’)”

Finally, anaphors also appear in contexts i

n which raising is impossible due to intervening constituents, suchas adjectival phrases:

douk-imme [ ēn eŋganei ŋookko-lpay.homage- .1 2 . brotherhood glorious-[rel till-ilokō tari-ou ei-tak]]-il-ka God-by 3 - bless- - -“I pay homage to your glorious brotherhood, blessed by God”

In this example, the adjective intervenes between the ante-cedent and the -, hence the appear-ance of the anaphor .

The strategies of exx. 31 and 32 may also appear together:

̣ ̣ ̣ : ̣ ̣ ̣ [ ] ̣̣ ·ed-d-rō [ kaji-ka [rel ein [ dig]]]-elnd- - .2 colt- bound-[rel eil tadi-ou[ ei-gou-la ouen-na tadi-dō ak-imis-s]]-ika now 3 - man- - one- 3 -upon sit- - 2-

“You will nd a colt that is bound, one upon which no one amongmen has yet sat” ( Mk. 11:2)

31SC 22.6–9

§3.9.15

32P. QI 4 110.v.1–2

33P. QI 2 14.i.9–11

43


30/310

28

Van Gerven Oei

In this example, ̣ is coindexed with both ̣ ̣and ̣ ̣ inthe dependent on it, in the rst case because of the intervening

̣ , and in the second case because ̣ ̣ is an inaccessiblepostpositional phrase. Note also that the accusative case ending on

̣ is unexpected. Owing to a lack of comparative evidence I amunable to suggest an explanation, but see the commentary to ex. 89for further discussion.

. Free relative clauses

Free or headless s have no overt antecedent, but are otherwisesyntactically similar to the coreferential and non-coreferential ex-amples discussed in §1. Whereas Satzinger does not discuss themseparately, and subsumes them under his main types “A” and “B,”Browne treats them separately inOld Nubian Grammar,§4.7.1. Head-less s may further be subdivided into subject clauses (§4.1); objectclauses (§4.2), which include different types of complement clauses(§4.2.1–2); and free s in other, oblique positions (§4.3). Unlike non-coreferential attributive clauses, free s are never introduced by arelative pronoun.

. Subject clausesSubject clauses are s that in their entirety, i.e., without anteced-ent, form the subject of a sentence.

agend-a-gou-lō [ [rel eir-in ŋog-la dou]]-l-gou-lblessed- - - 2 - house- exist- - -“Blessed are those who stay in your house”

Satzinger includes this example as a “Type A” , interpreting as an antecedent, but when compared to ex. 29 it be-

comes clear that we are dealing here with a subject clause, that is,a free without antecedent. The entire - is thesubject of the adjectival predicate , which agrees innumber with the subject. The same holds for the following examplequoted by Satzinger as an attributive , which in fact appears to bea coordinated subject clause:

·

34P. QI 1 2.ii.3–4

10


31/310

29


ein eis-sana [ [rel ŋarmit-in eigon-ka. be- 2.3 . beast- image-

douk]]-il-gou-lworship- - -

[ [rel tan taŋs-in eid-k-on ten koñ-in3 . name- sign- - 3 . face-tinnatt-iddō ett-o]]-l-gou-lfront. -upon receive- 1- - -“Those have been the ones who worship the image of the beast, whoreceived the sign of his name upon their forehead.” (Rev. 14:9)

Just like the previous example, the interpretation here is straight-forward. The subjects of are two subject clauses coordinatedby the suffix - in the noun phrase . Both subjectclauses are fully marked with a plural suffix and the double deter-miner construction we have seen previously.

A nal example is slightly puzzling, in the sense that we are notstrictly dealing with a subject sentence, but rather with a sentencethat as a whole seems to have been nominalized:

[ ] [ [rel einnin-ka oud-gille pes-ij-eri]]-l . . - 2 - say- - .1 -“(The fact that) I say these things to you” ( Jn. 16:33)

To take here as the antecedent of the [ ] makes no sense, as it would not be marked with the accusative case(but cf. in ex. 33); it can be nothing but the direct object of

. However, this leaves the occurrence of the determiner -lunexplained. Perhaps the scribe was attempting to render the per-fect tense found in the Greek (ταῦτα ε ά ηκα ὑμῖν) by nominal-izing the entire sentence. Again, lack of comparative evidence pre-vents further speculation.

. Object clausesObject clauses follow the same pattern as subject clauses, but are ob-viously marked by the accusative case - , sometimes preceded bythe determiner - . As yet it remains unclear in which contexts thedeterminer may precede the accusative suffix.

35P. QI 1 9.i.16–20

11 §4.6a

36L. 106.4–5

§4.5.1b


32/310

30

Van Gerven Oei

·[ [rel till-in taŋs-il-do auou-j-ou]]-ka God- name- -in do- - .2 -kourran-lo auou-j-anasō

joyful- do- - .3“Do the things you do in God’s name joyfully”

The free is here the object of .Note that the implicit antecedent of the is plural, as signied bythe pluractional marker - on both the embedded verb and the main verb .

ϩ

eihi[

[

rel ou-n eiar-i doll-is]]

-il-ka ouk-kalo 2 - know- ? want- 2- - 2 -eair-il-gas-se-sinknow- - - 2.1 . -“Lo, I have informed you about what you wanted to know.”

Observe that in ex. 38 we nd two instances of a determiner that arerelatively rare, the rst preceding the accusative case in (cf.ex. 25, 31 [ ] and ex. 32 ) and the second directly fol-lowing the verbal root in (cf. ex. 8 ). As forthe curious form , with either a phonologically reduced predica-tive suffix or a juncture vowel, cf. ex. 21 .

· · ·[ [rel pappajja ŋal-da eid-is]]-ka eit-ere-sin

Pappajja son- send- 2- take- .1 . -dippi-lavillage-“I take what was sent with the son (of) Pappajja to the village”

Ruffini’s translation is different here, analyzing as a serialverb consisting of - “to take” and - “to take, bring” with the ac-cusative case, rendering it with “receipt.” However, it seems to methat an analysis as a free , just like ex. 38, is more likely. I also take

- to be a rare instance of the verb - “to send?,” but obviously “totake” remains a valid option as well.

. . Complement clauses with verbs o reporting and desiringObject clauses are a subcategory of complement clauses which mayappear with verbs of reporting and desiring such as “believe,” “say,”“wish,” “think,” “write,” and so on. Grammatically speaking, Old

37K. 22.10–13

§4.6a

38SE A.i.11–13

2

39P. QI 4 124.r.2–3


33/310

31


Nubian complement clauses do not differ from regular object claus-es, or non-coreferential s in general. The subject of the comple-ment clause, if overtly expressed, will appear in the genitive case,and the entire clause will be marked by the accusative case. As with

regular non-coreferential attributive s, there is a complementarydistribution between overt subject and the presence of agreementmorphology.

] [ ] pisteu-eis-ana [comp ein ai-k eitr-es]-kabelieve- 2-3 2 . 1 - send. - 2-“They believed that you sent me”

Here we nd a complement clause dependent on the verb] [ ] , with a genitive subject and an embedded verb solely

marked for tense. There is no overt agreement marking because thesubject is explicit. The object clause is marked withthe accusative case as an object of the main verb ] [ ] . Cf.also L. 107.4–5 [ ] (sic) “So that theworld knows that you sent me.”

doll-immo[comp ou-kawish- .1/2 1 . -eiar-il-gad-j-ad-en]-kaknow- - - - - .2/3 -“We wish that you will inform (lit. cause to know) us”

In this example we nd explicit person marking on the embeddedverb owing to the absence of explicit subject. The object clause

is marked with the accusative case as the object ofthe main verb .

If there is an explicit subject in the complement clause, it ismarked by the accusative case.

(42)P. QI 4 91.r.6–7· ̣

ai-on [comp eik-ka tar-io-kon sipitor ann-eika1 - 2 - 3 - - foundation 1 . -tir-men]-ka eid-do ounn-ir-a-magive.2/3- - 2 -from love- - -

“And I want from you to not give from it to you and to my founda-tion.”

40L. 107.11–12

§4.7.1.2a

41St. 6.12–7.2

§4.7.1.2b

42P. QI 4 91.r.6–7


34/310

32

Van Gerven Oei

The subject of the complement clause is here marked by theaccusative.43 Ruffini translates slightly differently here, choosing torender as “for you.” This example shows several curious fea-tures, such as the absence of tense and agreement morphology in

the verb of the complement clause .44

Observe also the so-called copulative suffix - on the main verb,45 with the element - that otherwise appears in emphatic environments such as the affir-mative and vetitive. Finally, note the inverse order of possessed andpossessor in , with the genitive - following .

· ̣[ ] ̣ ·ŋai-sin [comp kau eik]-ka eik-ka pes-a tir-r-awho- naked be- 2 - say- give.2/3- -“Who told you that you were naked?” (

Gen. 3:11)

In ex. 43 we nd a double object construction, with indirect object and complement clause . The double object construc-

tion is supported by the applicative verb -, usually translated with“to give.”

For comparison, observe the following two examples from Nobiin,respectively with an intransitive and transitive verb in the comple-ment clause, which feature a similar construction with genitive-marked subject and accusative marker on the complement clause:

[comptar-iin kir-ee] ka dolli-ri 3 - come- 1 wish/love- .1“I wish him to come”

[comptar-iin kaba-ka kab-ee] ka rgi-ri 3 - food- eat- 1 want- .1“I want him to eat the food”

. . Complement clauses with verbs o abilityOther types of complement clauses have been attested in combi-nation with variants of the verb - “to be able,” which, just as theverbs of reporting in §2.2.1, takes a complement clause marked bythe accusative case. The two following examples show such a com-plement clause construction embedded under -.

43 See alsoP. QI 1 4.3–4elon eimme tillika[…] tiddekka.44 The morphology of negative contexts is still not completely understood for Old Nubian. Butcf. Nobiin exampletar entiga kab-i “He eats the date,” with tense/agreement suffix, andtar

entiga kam-muunwith a negative portmanteau suffix (A ,The Characteristic Features oNon-Kernel Sentences in Nobiin, §3.1.1.1, table 31). Perhaps the same is the case with .

45 B , Old Nubian Grammar,§3.10.

43SC 22.5–7

44 365

Nobiin

45 378

Nobiin


35/310

33


\\ ̣̣ ̣ ̣ [ ] iskel-itt-il-dekel-ka [ [rel [comp diar-iō tak-kapray- - - - death- 3 -auoul-os]-ik eir]]-il-gille

save- - be.able- -“…and prayer ( ) to whom is able to save him (away/completely)from death.”

The complement clause [ ] is the object of theverb - and therefore marked with the accusative -. The entirefree [ ] is then marked with a de-terminer and directive suffix: “to whom is able to save him (away)from death.” According to Bechhaus-Gerst, the suffix - in -should be interpreted as a “movement away from a […] deicticcenter,”46 which can also carry the meaning of fully completing acertain action, in casu the saving.

· · /ir-men-ta [ [rel ein douŋ-in mañan tri-kabe.able- - . blind- eye. -pikk-igar-o]]-l [comp ein-ketal di-minn-a-gar]-ikaawaken- - 1. . -also die- - - -“Is the one who opened the eyes of this blind man not also able toraise this one from the dead?” ( Jn 11:37)

The main verb here shows the compounding of the two neg-ative suffixes - and - , which is only attested with the verb- “tobe able” and - “to die.”47 Again note the absence of person morphol-ogy in the negative verb, as observed in ex. 42 . Assuming,with Browne, that we are dealing here with the introduction to aquestion “Is (he) not able to…?,” the clause is acomplement to , marked by the accusative case - . The subject of is formed by the subject clause […] “the one who opened the eyes of this blind man.” should not beinterpreted as a relative pronoun, as these never appear in free s.

Again we nd a similar construction in Nobiin:

[compay sirig-ka kay-inan]-ga esk-ir-i 1 boat- make- - be.able- -1“I am able to make a boat”

46 B -G ,The (Hi)story o Nobiin, 156. The directionality of the perfective suffix iscontested by J & E -G , “Semantic Change and Heterosemy of Dongolawied,”pp. 128–9.

47 B ,Old Nubian Grammar, §3.9.20. See also B ,Old Nubian Dictionary, p. 163, andS , “Einige Probleme der Morphologie des Altnubischen,” p. 395.

46L. 105.12–13

14

47P. QI 1 4.ii.13–15

3

48 353

Nobiin


36/310

34

Van Gerven Oei

. Other ree relative clausesExcept for subject and object positions, free s may also appear inother positions in the sentence.

·ouel tir-l [ [rel allilouia-ka psall]]-il-gou-l-dalanyone- Alleluia- sing.psalms- - - -psal-ēmen-ensing.psalms- - .2/3[ till-ou[rel tak-ka au-o]]-l-ka aeir-a-lo

God- 3 - make- 1- - insult- -“If anyone does not sing with those who sing Alleluia, he insults theGod who made him”

The free here is 48 “with those who singAlleluia,” with the antecedent “those” implied.

· · · · · · · ·[ [rel genkit-ka on-ke]]-l appa

goodness. - love- - for[ [rel till-in ount-il-dal pel]]-in ouer-a-sin

God- love- - be- one- -enn-a-sinbe- -“For who loves goodness is one of those who are with the love ofGod”

This example contains two free s, the rst · as subject of the sentence and the second · · as a genitivebelonging to . Satzinger suggests an “unexpected” interpre-tation of the suffix - in as third person singular morpheme,whereas the fact that the precedes its antecedent and the pres-ence of the verb -, which besides “to be” can also mean “to comeout,” suggest here a genitive case.

We are now able to analyze a complicated sentence such as thecomplete verse ofHeb. 6:7, which contains a number of embedded

s:

[ ] [ (?)] · < > [ ]

[ ] ·

48 Browne writes .

49K. 29.8–11, p. 208

50P. QI 2 18.iv.2–6

26

51P. QI 1 1.7.1.12–15

40


37/310

35


[ iskt-ou[rel [app ŋedian-no[ arou [rel tad-dō jōn-aearth- o en- rain 3 -upon strike-

soukk-o]]-l-ka sirip-a et-a tor-a] descend- 1- - drink- take- enter-

[ des dess-ou[rel el-takk-ou]]-l-ka crop green- obtain- - 1- -[ [rel ein [ tak-ka torpak-k]]]-il-gou-ka

3 - harvest- - - -peei-a os-a tij-j]]-il-deproduce- come.out- give.2/3- - -till-illo-jōa taoue-ka et-ara-goue-loGod- -through blessing- receive- 1. - . -“And the earth which, drinking up the rain that o en strikes uponit, produces obtained green crops for those who harvest it, receivedblessing(s?) from God” (Heb. 6:7)

The subject of the verb at the end of the sentence, - is the entireclause […] “and the earth which produces….”The ending in the verbal complex - has as its object “obtained green crops,” with a slightlycurious verb -. I have opted here to interpret the mor-pheme - as a variant of the preterite 1 morpheme, both becauseit is a coreferential attributive (cf. §7 below) and an interpreta-tion as an agreement suffix would make no sense. The indirect object, supported by the applicative verb -, is the free < >

[ ] “for those who harvest it.” All of this is preceded bya lengthy apposition […] [ (?)] “drinking upthe rain that o en strikes upon it,” which in turn contains a corefer-ential attributive with antecedent [ . Note that both and retain the determiner before the accusative case.

Relative clauses and predicative –

As Satzinger points out, in case the antecedent of an is markedwith the predicative, both the and its antecedent will feature thepredicative suffix. The predicative - therefore behaves differentlyfrom genuine case markings such as the accusative - and geni-tive - , which only appear at the right edge of the noun phrase.The precise syntactical structure underlying the assignment of thepredicative case, however, still lacks clarication. What follows willtherefore be a description of the different types of s that we have


38/310

36

Van Gerven Oei

encountered in the previous sections in the different contexts inwhich a predicative suffix appears.49

. Nominal predicates

The predicative suffix - rst of all indicates the main verbal ornominal predicate of a sentence. In case the nominal predicate isaccompanied by a , it is also marked with the predicative.

̣ \\ [ ] \\

[ [rel katapetasma ouou-ka-lo pal-a kil]]-l-on veil second- - come.out- come- -[ ŋaueir-a [rel ŋiss-igou-na ŋiss-a-a-lo

tent- holy- - holy- - -ok-tak-n]]-acall- - .2/3 -“And that which comes a er the second veil is the tabernacle, whichis called the Holy of Holies” (Heb. 9:3)

The subject of the nominal predicate is a subject clause ̣ , and it is accompanied by the

attributive [ ] . As we expect, both thenominal predicate and the are marked by the predicative suffix.

. Complex verbal predicatesWhat Browne calls “periphrastic” constructions are in fact not es-sentially different from the complement clauses we have discussedabove in §4.2.1–2, namely a full sentence embedded under a verb.In most instances of such constructions,50 the main verb is a copulawithout any overt tense marking, whereas the is marked with thepredicative, like a regular nominal or verbal predicate. The embed-ded verb can either appear with or without agreement morphology.I will just give a number of representative examples.

·tar-on mir-a ag-ende3 - be.barren- remain-[[ ounn-ar]-a men-n]-a-lō bear- 1- - .2/3 - -“And she, remaining barren, did not bear”

49 I follow here the observations made in “A Note on the Old Nubian Morpheme - in Nominaland Verbal Predicates.”

50 I include here only what Browne refers to as “predicative” and “indicative” periphrasticconstructions (Browne,Old Nubian Grammar, §§3.9.14–15).

52L. 111.7–9

§4.6a

53M. 1.8–11

§3.9.14


39/310

37


· tikkin-non imjirk-lo paj-anasōnevertheless- disobedience- cease- .3imjirk-il [[ deiar-ka ŋeg-r]-a

disobedience- death- produce- -en-en]-nejounbe- .2/3 -because“But nevertheless cease disobedience, because disobedience pro-duces death”

̣ ·[[ armis-tak-k]-a-eion ein-d-n]-a

judge- - - - be- - .2/3 -

eig-il-de tia -il-dekel-lore- - sulphur- - -“And he will be judged in re and sulphur” (Rev. 14:10)

In this example, the copula contains the modal suffix -, giv-ing the entire verbal complex a future sense. Note also the progres-sive assimilation of the present (or neutral) tense marker - a er- in .

As we have observed above, agreement morphology appears inthe embedded verb at the moment the subject of the embedded verbis not coreferential with the subject of the copula:

· ·ale-sin [[ ounn-r-e] en-en]-nontruly- bear- -1 . be- .2/3 -khristianos-aŋ-ad-immeChristian-i - - .1 .“If I give birth, I will become a Christian”

[ -jouri-ka mindi mindi-lo pes-in]-a

-about- individually- say- .2/3 -doum-men-ess-inexist- - 2-2/3“… about … it is impossible to speak individually” (Heb. 9:5)

Browne analyzes the verb here asdoum-men-es-sin, with the em-phatic marker - . However, the positioning of a verb marked with

this suffix at the end of sentence is unexpected, and with the begin-ning of the sentence missing no denitive grammatical analysis canbe given.

54K. 24.4–7

§3.9.14

55P. QI 1 9.i.22–ii.2

§3.9.14

56M. 6.8–10

§3.9.15

57L. 112.1–2

§4.6c


40/310

38

Van Gerven Oei

Complex verbal predicates can of course in turn be embedded,such as in the following example:

ϣ [\\] [

dogd-ri-gou-l mašalosk-lo tar-a ierousalm-iomagus- - - east- come- Jerusalem-ki-s-an-a [rel [ pes-r]-a-goue ein]-ilcome- 2-3 - say- - - - be-“The magi, coming from the east, arrived in Jerusalem, and said: …”

The [ ] is attributive to in spite of its dis-tance. And although it is nominalized with the determiner -, theinternal structure of the clause remains the same: the copula -preceded by an marked by the predicative suffix, as well as num-ber agreement with its antecedent.

. Vocative or appellative contextsThe predicative - appears also in vocative or appellative environ-ments, where someone is called or called upon.

[ [rel on-tak-r]]-a-goue-ke love- - - - . -2“You, beloved”

tan-nasō [ [rel apo-nacome- .2 1 . .father-tarou-as-s]]-a-goue-kebless- - 2- - . -2“Come, you, whom my father blessed”

Both examples contain a free which is marked by a predicative -,with plural predicative suffix - , and in both cases a special voca-tive suffix that refers to the second person plural “you,” -. Other-wise the seems to behave grammatically like the ones discussedabove, such as marking the subject with a genitive case.

Other examples are grammatically a bit more complex:

58L. 113.3–5 §3.9.14

59M. 1.5; St. 2.9

et passim 23

60St. 29.10–12

24


41/310

39


[ŋokkor-a [rel khristos-in marturos-ou ŋiss-oumiracle- Christ- martyr- holy-mēna-na au-s]-a-lō

Mina- do- 2- -“(It is) a miracle performed by Mina, the holy martyr of Christ”

We are dealing here with regular non-coreferential attributive in which both the antecedent and […] - aremarked with the predicative - . The shows no agreement mark-ing, as the subject is overtly expressed, with a genitive suffix. Theexample in question is the rst sentence of the text known as theMiracle of St Mina (or Menas) and as such functions as a type of an-nouncement or caption for the entire story.

This vocative-like usage of the predicative suffix also appears inother contexts, for example in one of Griffith’s graffiti. I give here anextensive fragment:

ϩ · · · · ·

petros-i[rel harm-i kolotit-in tere-gou-ka koñ-j-r]-aPeter- heaven- seventh- key- - have- - -[rel [[rel ai ei-a pes-s]-i ŋape]-gou-ka tok-ar]-a 1 say- say- 2- sins- - forgive- 1-din-esō an-na añ-en oukour-rōgive.1- .2 1 - life- day-ai-ka matta-ka kipr-a git-tamisō1 - affliction- eat. - - .2“Peter, who has the keys of the seventh heaven, who has forgiventhe sins that I have spoken, give, do not cause my affliction to con-sume me in the days of my life”51

Peter, the one who is called upon here by the author of the graffito,is qualied by two coreferential attributive s, both ending in thepredicative suffix - . The rst coreferential attributive shows thepattern we have seen before, whereas the second one contains anadditional non-coreferential attributive , with theantecedent (see ex. 20 for discussion). This lengthy ap-pellation to Peter is then followed by an imperative and a ve-titive .

51 B , “Griffith’s Old Nubian Graffito 4,” p. 19, translates “O Peter, you who have the keysof the 7 heavens, forgive me for the sins that I uttered. Cause me not to eat tribulation in thetime of my life,” interpreting as an imperative and as the object of · instead of .

61M. 1.1–3

§4.6a 21

62gr. 4.1–3

38= ex. 20


42/310

40

Van Gerven Oei

. Adjunctive/appositional clausesSatzinger gives a few other examples in which the predicative - ap-pears. These, however, are not proper s, i.e. with a fully developedclausal structure including tense and subject and/or agreement

morphology, but rather embedded verb phrases without any higherprojection, without possible subject, and without ever featuring arelative pronoun. Needless to say, all of them are coreferential.

· ·pap-a [app ŋal-la doull]-a ŋa [app pap-la doull]-afather- son- exist- son. father- exist-“Father being in the son, son being in the father”

ourou-a [app til-ka ounn]-aking- God- love-“God-loving king”

e [app kipt-a [ khristos-ka ounn]]-a-goue-keoh people- Christ- love- - . -2“O Christ-loving people”

. Topicalization o the antecedentThere are a few recorded cases of s with an antecedent marked bythe predicative suffix that seems to have moved to a more le wardposition. In all of these cases we are dealing with a topicalized ante-cedent, as signaled by the appearance of the “emphatic” suffix -52 and the predicative suffix. However, as predicative morphology isnot always present in the itself (as in exx. 52–62), it may be thecase that this assignment only takes place a er movement of the an-tecedent. In all examples below, the movement is not visible on thesurface. The intuition that le ward movement is involved derivesfrom 1) the fact that - does not always appear in these contextsand that its presence must be linked to a specic syntactic positionin the sentence, and 2) the abundant presence of - in contexts ofquantier raising, which in other languages explicitly features le -ward movement (see §5.6). This type of-topicalization is allowedfrom both coreferential and non-coreferential attributive clauses.

52 There is no room here for a full discussion of the - morpheme, which has been analyzedby S , “Relativsatz und Thematisierung im Altnubischen,” pp. 195ff; B ,OldNubian Grammar, §3.10; B , Old Nubian Dictionary: Appendices, pp. 28–37, esp. p. 31“Predicative + - as Antecedent”; and Bechhaus-Gerst,The (Hi)story o Nobiin, pp. 103–4.

63St. 17.5–8

19

64P. QI 1 10.C.ii.7

20

65P. QI 2 16.vi.11–vii.2

22


43/310

41


· ·stauros-il adñike-r-a-lo till-a-sini cross- life.giving- - - God- -[ti [rel gad-lo tad-dō oll-o]-l-lojō-a

esh- 3 -upon hang- 1- -because-“The cross is life-giving, because of God who hung upon it in theesh”

In this example, has apparently moved to the le , and hasbeen marked by the emphatic suffix - . Note that the

is marked with the predicative -.

· · ·

ŋod-a till-a ou-naLord- God- 1 . -ou-n aeil-gou-na-eion esoggi-der-a1 . - heart- - - release- -ai-a-goue-sini ŋape-lo toukm-aheart- - . - sin- stink-[ti [rel kor-ka et-o]-l-gou-na-eion iatoros-a wound- receive- 1- - - - doctor-“God, our Lord, and deliverance of our hearts, and doctor of wound-ed hearts, stinking in sin”

This series of appeals to God contains one ,with the antecedent , which clearly has moved out of geni-tive construction into a higher position in the phrase, where it hasbeen marked with the emphatic marker - , and the predicativemarker - . More clear than ex. 66, this example is perhaps evidenceof -topicalization as both the predicative - and - are assigned inthe target position, whereas the is not marked by the predicative,but rather with the genitive, as attributive to .

· ·

kristianos-igou-n eilēu-gou-l pes-ran oulgr-aChristian- - woman. - - talk- .3 hear-tōek-a-goue-sini [ti [rel ŋiss-ou mēna-namiracle- - . - holy- Mina-mareōt-in kisse-la au-j]]-il-gou-kaMareotis- church- do- - - -“Hearing the women of the Christians talk about the miracles thatSaint Mina performed in the church of Mareotis”

66St. 27.7–10

§4.6a 49

67St. 4.7–5.1

50

68M. 2.14–17

51


44/310

42

Van Gerven Oei

This is another example in which the antecedent hasmoved le ward and become marked with the predicative and em-phatic markers. Again the is not marked with the predicative -,but with the accusative.

· ·

eigarigra-lo ouel-ende[comp dir-j]-ika eir-imen-ta-lothus- one- count- - be.able- - -ŋisse-goue-sini [ti [rel aggelos-naholy. - . - angel-aul-os-ij-is-in]]-na dirti-kasave- - - 2-2/3 ?- number-“So no one is able to count the number of holy ones whom the angelsaved”

This nal example rst shows a complement clause depen-dent on the verb of ability (see §4.2.2), followed by theantecedent of the . As in ex. 67,the antecedent has moved out of a possessor phrase, here depen-dent on the object of , ·. Note that the analysis of - in

is uncertain. Because the subject of the isexplicit, we normally do not expect to nd agreement morphology.The only other option is that we are dealing here with an assimilatedform of the determiner - , which, however, never seems to appearbefore the genitive case.

. Relative clauses within the scope o quantiersA special instance of attributive s marked with the predicativesuffix are those whose antecedents are the quantiers ϣ “all”and - “any” which mark constituents within their scope withthe predicative - . Apart from featuring this additional marker, the

s have also moved into the scope of the quantier and are there-fore structurally similar to coreferential s with a restricted read-ing as discussed in §3.1.1.53

ϣ [[rel tan ŋog-la dou-ar]-a miššan]-gou-ketalle-eion 3 . house- live- 1- all- -also-meir-a-goue-lō eis-s-anabe.barren. - - . - be- 2-3 .

“And also all who lived in her house were barren”

53 Le ward movement of the has not been indicated in the examples below.

69P. QI 1 10.A.i.12–15

66

70M. 2.5–7

16


45/310

43


This example features a coreferential attributive embedded un-der the antecedent ϣ - and therefore marked with the predica-tive. Although Satzinger follows Browne in claiming that becauseof this predicative marker before ϣ - “es steht Prädikativ statt

Verbid, und Indikativ statt Subjunktiv,”54

such a statement is mean-ingless in a descriptive context in which we attend to Old Nubianmorphology, under the assumption that it is more an agglutinativerather than a synthetic language.55 Note also the complex verbalpredicate with copula and predicative-marked - (cf. §5.2).

[ ]ϣ ̣̣ \[\] [ ][ ] ̣

[[rel ein tak-ka tij-j-is-n]-a miššan]-ka 2 . 3 - give.2/3- - 2-2/3 - all-añj-i ellen ketallen-ka tek-ka tij-j-ikonnoalife- eternal- 3 - give.2/3- - .2/3“So that all that you have given him you give them eternal life”( Jn. 17:2)

The grammatical analysis of this sentence, with no less than fouraccusative marked constituents, does not appear straightforward,and it is helpful to look at the Textus Receptus from Jn. 17:2, whichBrowne identies as the verse that is translated here. In Greek weread ἵνα πᾶν ὃ δέδωκας αὐτῷ δώσῃ αὐτοῖς ζωὴν αἰώνιον, and at rstsight it seems that the Old Nubian follows the Greek original ratherfaithfully, perhaps to the detriment of its own grammatical coher-ence. The purposive construction ἵνα […] δώσῃ “in order to, so thatyou give” is translated in Old Nubian by the (emended) nal verbform [ ] [ ], whose indirect object αὐτοῖς is rendered withthe accusative and direct object ζωὴν αἰώνιον with

. The translation of the Greek phrase πᾶν ὃ δέδωκαςαὐτῷ, with πᾶν “everything, all” in the accusative case and renderedin Old Nubian [ ]ϣ, is somewhat less intui-tive. Note both the genitive subject and agreement marking on

are explicit, to avoid any ambiguity. The King James Bibletranslation of the Textus Receptus gives the rather uent “that heshould give eternal life to as many as thou hast given him,” whereasBrowne translates the Old Nubian with “in order that, as for all youhave given him, he may give eternal life to them,” with an inexpli-cable “as for.” Yet in both translations, [ ]ϣand πᾶν ὃ δέδωκας αὐτῷ δώσῃ are appositions to and αὐτοῖς

54 S , “Relativsatz und Thematisierung im Altnubischen,” p. 18955 V G O , “Remarks toward a Revised Grammar of Old Nubian,” pp. 174–80.

71L. 106.14–16

17


46/310

44

Van Gerven Oei

respectively. Also note that the plural object marker - in refers to the antecedent [ ]ϣ -.

. Quantier raising

Satzinger points out that it is “remarkable” that the constructionwith - as discussed in §5.3 o en appears in the context of the quan-tiers ϣ “all” and “everyone.” This fact, however, seems tosupport our initial intuition that a constituent marked by predicativeand - has moved out of its original position to a higher position (cf.§5.4). This type of movement in the scope of quantiers in commonlyreferred to as “quantier raising,” a result of the interaction betweensemantics and syntax levels of representation. I repeat here the ex-amples given by Satzinger, which are otherwise grammatically akinto the examples adduced in the previous sections.

ϣ · ·

[ eit-a miššan]-a-sini [ti [rel till-il taŋs-il-doman- all- - God- (sic!) name- -in

auou]]-l-gou-l aiouil-en till-il-orodo- - - be.grudging?- .2/3 God- -fromouel-ka el-men-d-inna-loone- obtain- - - .2/3 . -“All men who act in God’s name, when grudging(?), will obtainnothing from God”

ϣ { } · ·

[añir-a miššan]-a-sini [ti [rel ai-agille pisteu]]-l-onliving.being- all- - 1 - believe- -di-men-ta-lo ellen-gou-lo kiskil-lodie- - - eternity- - until-“And all living beings who believe in me do not die until eternity”( Jn. 11:27)

·

jimmil-a-goue-sin [ti [rel ten aeil ouatto-lo ieveryone- - . 3 . heart entire-stauros-lagille pisteu-o]]-l-gou-ll-on tauō-lo ŋonj-ilcross- believe- 1- - - - under- stand-

dou-d-innaexist- - .2/3 .“And everyone who believes in the cross with their entire heart willstand under (it)”

72K. 22.4–9

53

73P. QI 1 4.i.12–13

54

74St. 3.3–8

57 §4.6a


47/310

45


In all three examples we nd that the consituent including the quan-tier has moved to the rst position in the clause, leaving behind the

. Note that we have observed in §5.5 that in a neutral environments always precede a quantier. In exx. 72–4, however, we nd that

they all have moved and have been marked by the predicative mark-er - and the emphatic marker - .Whereas exx. 72–4 all showed subjects containing a quantier,

the following examples all feature an object containing a quantier.In each case the quantier has moved up to the le edge of the con-stitutent.

· [ ] ϣ · {·} [ ]

·e an pidt-a petros-i onoh 1 . friend- Peter- our-ou an eŋŋae-goue-ke2 - 1 . brother. . - . -2oul-lo eier-a [ ŋeei-a miššan]-a-goue-sini2 - know. - thing- all- - . -[ti [rel ioudaios-ri mamiskaei-gou-na ai-agille

Jew- unjust. - - 1 -au-eij-s-an]]-gou-n-kado- - 2-3 - -?-“Oh Peter, my friend, and you, my brothers, you know all the thingsthat the unjust Jews did to me”

Note here the presence of both an overt subject {·} and agreement morphology in the embedded verb -

(cf. ex

dotawo nubian studies vol 2 e-issue

Documents