The syntax of relativization

(1)

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

de Vries, M.

Publication date

2002

Link to publication

Citation for published version (APA):

de Vries, M. (2002). The syntax of relativization. LOT.

General rights

It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)

and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open

content license (like Creative Commons).

Disclaimer/Complaints regulations

If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please

let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material

inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter

to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You

will be contacted as soon as possible.

(2)

Appendixx II Typological data

Thiss appendix summarizes the syntactic characteristics of relative constructions in a

largee sample of languages around the world.

Thee following tables are included:

Tablee 1: Genetic and geographical distribution of the languages in the sample.

Tablee 2: Characteristics of relative clauses.

Figuree 1: Relative clauses around the world.

Alll other tables are derived from table 2. Tables 3-7 list all main types of relative

clauses;; 8-14 list relative pronouns and particles; 16-20 concern the position of the

determiner;; 21-24 list unexpected relative strategies with respect to the basic word

orderss of the languages involved; 25 lists relative strategies in ergative languages; 26

listss languages with more than one main strategy.

Tablee 3: Circumnominal relatives.

Tablee 4: Correlatives.

Tablee 5: Prenominal relatives.

Tablee 6: Participial relatives.

Tablee 7: Postnominal relatives.

Tablee 8: Relative pronouns.

Tablee 9: Resumptive pronouns and clitics.

Tablee 10: Relative complementizers.

Tablee 11: Relative markers.

Tablee 12: Relative affixes.

Tablee 13: Strange relative particles.

Tablee 14: Nominalizmg/attributive particles.

Tablee 15: Zero relativization.

Tablee 16: The position of the determiner in postnominal relative constructions.

Tablee 17: The position of the determiner in prenominal relative constructions.

Tablee 18: The position of the determiner in circumnominal relative constructions.

Tablee 19: Split determiners.

Tablee 20: Determiners non-adjacent to N.

Tablee 21: Circumnominal relatives in non-SOV languages.

Tablee 22: Correlatives in non-SOV languages.

Tablee 23: Prenominal relatives in non-SOV languages.

Tablee 24: Postnominal relatives in non-SVO languages.

Tablee 25: Relative clauses in ergative languages.

(3)

Thee sample contains all languages on which I found a description of relative clauses

inn the sources mentioned below. As is clear from table 1, there are examples from

manyy different language families. I am convinced that the typology is nearly

complete,, in the sense that if more data is added, this will probably not increase the

numberr of patterns found so far (although several details can be added).

II have tried to standardize the relevant information on relative clauses. Hence

thee description here frequently diverges from those in the sources consulted. This is

necessary,, because the terminology and the degree of detail in the notation differ

greatlyy with author. The objective of the tables below is to facilitate a comparison

betweenn relative clause systems in different languages, and to reveal the patterns

madee possible by the language faculty.

Sincee I have included all information easily accessible to me, the tables are

biasedd towards Indo-European languages. Hence it is not possible to draw statistical

conclusionss from this sample without further processing. For instructions on this

matterr I refer to Bakker & Hengeveld (2001) and the references there. To me,

informationn on possible patterns is more important than knowledge of the frequency

off these patterns.

Finally,, notice that an interpretation of secondary sources as presented here

runss the risk of errors. At this point I wish to express the hope that future work will

reveall possible mistakes here, add new data, and, most importantly, follow the

standardizedd typology argued for, and of course the corresponding notational

system. .

Thee first table contains genetic and geographical information on the languages in

question.. The columns phylum > stock > family > branch > group > language

containn a genetic classification of the pertinent languages. It is mainly based on

Grimes'ss (1992) Ethnologue: Languages of the World and Moseley & Asher's

(1994)) Atlas of the World's Languages. The spelling is according to the Ethnologue.

(Thee Ethnologue's index contains many possible alternative names. I have added

synonymss only if the name used here poses difficulties for finding a language in the

sourcess mentioned.) Where the sources diverge on the genetic classification, the

compromisee is my responsibility. The Ethnologue''s classification is often more

fine-grainedd than the Atlases. Since it concerns geographical notions mainly, and

confusionn is unlikely, I follow the Atlas in many cases. Finally, I think the genetic

hierarchyy must be seen as relative, not absolute. An absolute classification in terms

off phyla, families, etc. is quite arbitrary. It seems that the Atlas tries to do so, but it

iss inconsistent in different chapters.

Thee columns code and place contain the Ethnologue's unique 3-letter code of

thee language, and the principal country where the language is spoken.

(4)

TYPOLOGICALL DATA

367 7

Tablee 1. GeneticGenetic and geographical distribution of the languages in the s

phylumm > stock > family > branch > group >

Afro--Asiatic c

Chadic c Eastt 1 Kwang-Kera

Westt | Hausa Cushitic c Eastern n Egyptian n Semitic c Central l Ethic--semitic c South h Southh Canaanite North h Southh | Transversal Algic c Algonquian n Altaic c Arawakan n Australian n Austro--Asiatic c Ojibwa a Japanese e Korean n Mongolian n Turkic c Maipuran n Eastern n Oghuz z Campa a Murrinh-Patha a Pama--Nyungan n Arandic c Artuya a Urtwa a Djirbalic c Kalaa Lagaw Ya South-Westt | Ngarga Mon-Khmer r Khmer r Vietnamese e language e Kera a Hausa a Komso o [=konso] ] Oromo o r=GaU(my)al l Saho-Afar r Coptic c Egyptian n (ancient) ) Akkadian n [KMd d Babylomc] ] Arabic c (classical) ) Arabic c (modern n standard) ) Arabic c (Tunisian) ) Hebrew w Geez z Tigré é Amharic c Yurok k Ojibwa a (Eastern, , Northern, , Western) ) Japanese e Korean n Mongolian n Turkish h Ashéninca a Murrinh--Patha a Gaididj j [=Kaititjl l Arremte e (Eastern) ) [=Mpamtwe e Arende] ] Djirbal l Kala a Lagaww Ya [=Mabuiag] ] Warlpiri i Khmer r [=Cunbodian] ] Vietnamese e code e [KER] ] [HUA1 1 [KXC] ] [GAZ] ] [AFR], , [SSY] ] [COP] ] [ARA] ] [ABV] ] [AEB] ] [HBR] ] [GEE] ] [TIE] ] [AMH] ] [YURI I [OJG], , [OJB], , [OJI] ] [JPN] ] [KKN] ] [KHK] ] [TRK] ] [CPU] ] [MWF] ] [GBB] ] [AER] ] [DBL] ] [MWP] ] [WBP] ] [KMR] ] [VIE] ]

ample. ample.

place e Chad d Nigeria a Ethiopia a Ethiopia a Ethiopia a (old) ) Egypt t (ancient) ) Egypt t (ancient) ) Meso--potamia a Saudi i Arabia a Saudi i Arabia a Tunisia a Israel l (ancient) ) Ethiopia a Ethiopia a Ethiopia a USA A Canada a Japan n Korea a Mongolia a Turkey y Peru u Australia a Australia a Australia a Australia a Australia a Australia a Cambodia a Viett Nam

(5)

phylumm > stock > family > branch > group > Austro--nesian n Malaio-- Poly--nesian n Azteco--Tanoan n Central l Eastern n Wes--tern n Oceanic c Borneo o Micronesian n New w Georgia a Northh & Central l Vanuatu u Poly--nesian n Barito o Rovi--ana a East t Vanu--atu u Nuc--lear-» » East t Tongic c East t Chamorro-Palauan n Meso-- Philip--pine e Cen--tral l Bikol l East t Mindanao o Natan n Sundic c Kiowa-Tanoan n Uto--Aztecan n Boran-Witotoan n Javanese e Malayic c Sumatra a Local l Malay y Malayic c -Dayak k Para--Malay y Batak, , Southern n Kiowa-Towa a Tewa-Tiwa a Aztecan n Northern n Numic c Sonoran n Corachol l Pimic c Tara--cahitian n Papago--Pima a Cahitan n Tara--humaran n Tepiman n Takic c Cupan n Boran n language e Ponapean n Kiribati i f=Gflbertese] ] Roviana a Ambae e (East) ) [=Aoban] ] Maori i Tongan n Malagasy y Palauan n Bicolano o (central) ) [=Bikol] ] Tagalog g Kalagan n Ivatan n Javanese e Indonesian n Malay y Iban n [=Seaa Dayak] Minang--Kabau u Batakk Toba Kiowa a Tewa a (Arizona) ) Nahuatl l Hopi i Sboshoni i Utee / Paiute Huichol l Papago--Pima a Yaqui i Tarahumaran n (Central, , Northern, , Southwest) ) Tepecano o Cahuilla a Luiseno o Bora a code e [PNF] ] [GLB] ] [RUG] ] [OMB] ] [MBF] ] [TOV] ] [MEX] ] rPLUl l [BKL] ] [TGL] ] [KQE] ]

fiwi i

[JAN N [DC] ] [MLI] ] [JBA] ] [MPU] ] [BBC] ] fKIOl l [TEW] ] [NAI] ] [HOP] ] [SHH1 1 [UTE] ] [HCH] ] [PAP] ] rYAOl l [TAR], , [THH], , [TWR] ] [TEP] ] [CHL] ] [LUI] ] [BOA] ] place e Micro--nesia a Kiribati i Solomon n Islands s Vanuatu u New w Zealand d Tonga a Mada--gascar r Belau u Philippines s Philippines s Philippines s Philippines s Indonesia a (Java) ) Indonesia a Malaysia a Indonesia a (Kaliman--tan) ) Indonesia a (Sumatra) ) Indonesia a (Sumatra) ) USA A USA A Mexico o USA A USA A USA A Mexico o USA A Mexico o Mexico o Mexico o USA A USA A Peru u

(6)

TYPOLOGICALL DATA

369 9

Dravidian n Southern n Central l Eskimo-Aleut t Hokan n Indo--European n Eskimo o Yuman n Tamil-Kannada a Telugu-Kui i lnuit t Delta-Californian n Riverr Yuman Uplandd Yuman Albanian n Anatolic c Balto-Slavic c Celtic c Germanic c Greek k Slavic c East t South h West t Brythonic c Goidelic c North h West t language e Kannada a [=Kmaresian] ] Tamil l Telugu u Green--landic c Dieguefio o Mohave e Yavapai i Albanian n r=Toskl l Hitnte e Russian n Russian n (Medieval) ) Ukrainian n Bulgarian n Macedonian n Serbo--Croatian n Czech h Polish h Slovenian n Welsh h [=Kymtic] ] Gaelic c [=Irish] ] Danish h Icelandic c Norwegian n Swedish h Dutch h English h Frisian n (Northern) ) German n Schwyzer--dütsch h (Zurich) ) [=Züriiüüts, , Swisss German (Zurich)] ] Greek k (modern) ) Greek k (ancient) ) [=Homeric, , Attic] ] code e [KJV] ] EKJV1 1 [TCW1 1 [ESG] ] P H I I [MOV] ] fYUFl l [ALN] ] [RUS] ] [UKR] ] [BLG] ] [MKJ] ] [SRC] ] [CZC] ] rPOL] ] [SLV] ] [WLS] ] [GLI] ] PNS] ]

ncEi i

[NRR] ] [SWD] ] P U T ] ] [ENG] ] [FFR] ] [GER1 1 [GSW] ] [GRK] ] [GKO] ] place e India a India a India a Green--land d Mexico o USA A USA A Albania a (ancient) ) Turkey y Russia a (Medieval) ) Russia a Ukraine e Bulgaria a Macedonia a Yugasloviaa / Croatia a Czecho--slovakia a Poland d Slovenia a United d Kingdom m Ireland d Denmark k Iceland d Norway y Sweden n Nether--lands s UK,, USA Germany y Germany y Switzer--land d Greece e Greece e

(7)

phylumm > stock > family > branch > group > Indo--European n Indo-Iranian n Romance e Khoisan n Maya a Na-Dene e Niger-Congo o Indo--Aryan n Central l Eastern n Southern n Western n Iranian n Daco-Romance e Gallo-Romance e Ibero-Romance e Italo-Romance e Latino-Faliscan n Khoe/Central l Cholan-Tzeltalan n Kanjobalan-Chujean n Kanjobalan n Quichean n Yukatecan n Athapaskan--Eyak k Atbapaskan n Adamawa-Ubangi i Atlantic c Benue--Congo o Apachean n Mbum m Northern n Bantu u Southern n Eastern n Senegal-Guinea a Senegambian n Central l East t Central l West t Shona a Kimbundu u Kongo o Mbana a language e Hindi i Bengali i Sinhala a Gujarati i Marathi i Sanskrit t (Vedic) ) Avestic c Farsi i [== Persian] Rumanian n French h Ligurian n (Genoese) ) Catalan n Portuguese e Spanish h Italian n Latin n Nama a [=Hottentot] ] Tzeltal l Jacaltec c Kekchi i Yucatecan n Navaho o Mbum m Bainouk k Fulfulde e (Adamawa) ) [=Fula(ni)] ] Wolof f Shona a Mbundu u (Loanda) ) Kongo o [=Dzamba] ] Hungana a [-KiHungana] ] code e [HND1 1 [BNG] ] [SNH1 1 [GJR] ] [MRT] ] [SKT] ] [PES] ] [RUM] ] [FRN] ] [LIJ] ] [CLN] ] [POR] ] [SPN] ] [ITN] ] [LTN] ] [NAQ] ] [TZH] ] [JAI] ] [KEK] ] fYUA] ] [NAV] ] [MDD] ] [BCZ] ] [FUB] ] [WOL] ] [SHD] ] [MLO] ] [KON] ] [HUM] ] India a Bangla--desh h Srii Lanka India a India a (ancient) ) India a (ancient) ) Persia a Iran n Romania a France e Italy y Spain n Portugal, , Brazil l Spain n Italy y (ancient) ) Mediter--ranean, , Vatican n State e Namibia a Mexico o Guate--mala a Guate--mala a Mexico o USA A Cameroon n Senegal l Cameroon n Senegal l Zimbabwe e Angola a Kongo o Kongo o

(8)

TYPOLOGICALL DATA

_{371 1}

Niger-Congo o Nilo-Saharan n North h Caucasian n Benue--Congo o Bantu u North--East t North--west t Defoid d Edoid d Ganda a Rwanda--Rundi i Sabaki i Zinza a Mbere e Yoruboid d South h Igboid d Kainji i Kauru u Dogon n

Gur r Oti-Volta a Northwest t

Southwest t Ijoid d Kru u Kwa a Mande e East t Gbe e Tano o Western n Easternn Sudanic North h Nilotic c East t West t Kalenjin n Northwest t Oto-Manguean n Papua a Zapotecan n Sepik-Ramu u Zapotec c Sepik k Sepik k Hill l Sanio o language e Ganda a [=Luganda] ] Rwanda a [=Kinya--Rwanda] ] Swahili i Haya a Mbama a [=Bamba, , Bemba] ] Yoruba a Urhobo o Igbo o Kinuku u [=Kinung'an] ] Dogon n (DonnoSo) ) Dogon n (Togoo K3) Moore e Dagbani i Ijo o f=Kdokumal l Godié é [=Koyo] ] Ewé é Akan n Bambara a Maninka a Mandinka a Vai i Lango o Lango o Kupsabiny y [=Sebei, , Nandi] ] Abkhaz z Hurric c Sumerian n 'Zapoteco' ' Hewa a code e [LAP] ] [RUA] ] [SWA] ] fHAYl l [MBM] ] [YOR1 1 [URH] ] RGR1 1 [KKD] ] [DOG] ] [MHM] ] P A G ] ] [UC] ] [GOD] ] [EWE] ] [TWS] ] [BRA] ] [MNI] ] [MNK] ] [VAI] ] [LNO] ] [LAJ1 1 [KPZ] ] [ABK] ] [...] ] [HAM] ] Uganda a Rwanda a Tanzania a Tanzania a Gabon n Nigeria a Nigeria a Nigeria a Nigeria a Mali i Burkina a Faso o Ghana a Nigeria a Ivory y Coast t Ghana a Ghana a Mali i Senegal l Guinea a Liberia a Sudan n Uganda a Uganda a Georgia a (ancient) ) Meso--potamia a (ancient) ) Meso--potamia a Mexico o Papua a New w Guinea a

(9)

phylumm > stock > family > branch > group > Quechu--Aymaran n Sino-Tibetan n Siouan n Quechuan n Central l Peripheral l Chinese e Tibeto--Burnian n Bunnese-Lolo o Burrrush h Lolo o Naga-Kuki-Chin n Tibetan n

Mississippii Valley | Dakota Missourii Valley Southh Caucasian Tai-Kadai i Trans--Neww Guinea Tupi i Uralic c Kam-Taii Tai Eastern n

Neww Guinea Highlands East-Central l

Tupi-Guarani i Finno-Ugric c Finnic c Ugric c Finno-Cheremisic c Yuki i (Isolated) ) (Unclassified) ) language e Ancash h Cuzco o Huanca a Huayllaa / Jauja Huanuco o Huallaga a Ayacucho o Imbabura a Chinese e (Mandarin) ) Burmese e Lahu u Lushai i Tibetan n Lakota a Crow w Georgian n Thai i Alekano o [=Gahuku] ] Guarani i Finnish h Hungarian n Man n Erzya a [=Mordvin] ] Wappo o Basque e American n Sign n Language e code e [QED, , QAN..] ] [QUZ] ] [QHU/ / 0HJ1 1 [QUB] ]

rouYi i

? ? [CHN] ] [BMS] ] [LAH] ] [LSH1 1 [TIC] ] rDHG] ] [CRO] ] [GEO] ] [THJ1 1 [GAH] ] [GUG] ] [FIN] ] [HNG] ] [MALI I [MYV] ] [WAO] ] [BSO] ] [ASE] ] _pjace e Peru u Peru u Peru u Peru u Peru u Equador r China a Myanmar r [=Burma] ] China a India a Tibet t USA A USA A Georgia a Thailand d Papua a New w Guinea a Paraguay y Finland d Hungary y Russia a Russia a USA A Spain n USA A

(10)

T Y P O L O G I C A LL D A T A

373 3

Tablee 2 - the main table - contains all relevant information on relative clauses in the

languagess under consideration. There are nine columns, where I use the following

abbreviations: :

(2-4)(2-4) A characterization of the language.

S,0,VV The main constituent order: a permutation of Subject, Object, Verb.

acc/ergg Accusative or ergative system.

Casee Indicates possible Case markings:

noo visible Case;

ss suffixal Case on N;

s++ suffixal Case on N which is doubled on a restrictive relative clause;

S(+)) suffixal Case on N or a free relative;

pp prefixal Case on N;

ff non-suffixal Case following the relevant constituent.

11 non-suffixal Case preceding the relevant constituent.

(5)(5) RC type. The relative clause main type.

pree prenominal relative;

preparr prenominal participial relative;

1

postt postnominal relative;

postpa,.. postnominal participial relative;

cirr circumnominal relative;

corr correlative.

(6)(6) Det. The matrix clause determiner ((in)definiteness).

normallyy no determiner;

Dll D first (i.e. for post D-N-RC, for pre D-RC-N);

DmDm D middle (i.e. for post N-D-RC, for pre RC-D-N);

Dff D final (i.e. for post N-RC-D, for pre RC-N-D, for cir RC-D);

cDD Correlative Demonstrative or personal pronoun in the matrix clause.

Manyy languages can use a partial participial strategy, e.g. the winning athlete = the athlete who

wins.wins. This is not indicated in the tables. A complete participial strategy has at least the possibility of

(11)

(?)(?) Gap. Occupation of the gap in the relative clause.

zeroo gap;

RPP relative pronoun (first position, <}>, (abstract) subCase):

RPwhh a relative pronoun in wh (question) format;

RPdd a relative pronoun in d (demonstrative) format;

RPspp a relative pronoun in specialized format;

GDD the gap contains a resumptive demonstrative/personal pronoun;

2

GAA the gap contains (a trace of) a resumptive affix/clitic (notice that if the gap

iss a prepositional/genitive object then the affix is on P or N);

NN the gap contains (a copy of) the head noun N.

(8)(8) C/REL. Relative particles, etc. (no gap occupation).

noo relative particle;

RCC relative complementizer (C position (normally first), no (|>, no Case):

RCSRR relative subordinator;

RCNRR nominalizing relative complementizer;

RC

A

TT attributive relative complementizer;

RCspp specialized relative complementizer;

RMM relative marker (first position, <J>, if Case then mainCase):

RMCLL relative classifier marker;

3

RAA relative affix on V (specialized, sometimes § and/or subCase):

RA(Agr)) relative agreement affix (replaces Agr on V);

RA(T)) relative temporal affix (replaces T on V);

RA(NR)) nominalizing affix:

RA(NR

T

)) nominalizing affix that replaces a temporal affix;

RA(NR

add

)) additional nominalizing affix;

RA(AT)) attributive affix;

RA(SR)) subordinating affix;

RA(CL)) relative classifier affix;

RA(add)) other additional relative affix;

RR unclassified relative particle.

N.B.. Relative pronouns and particles are discussed in Chapter 5. See especially

Ch5§44 for more details on the classification used.

2 2 3 3

Resumptivee strategies for functions very low on the syntactic function hierarchy (cf. Ch2§4) such as

genitivegenitive are not indicated.

Iff my analysis in Ch5§3.3.1 is correct, classifier markers are (remnants of) relative pronouns, contraryy to appearances.

(12)

T Y P O L O G I C A LL DATA

375 5

(9)(9) The Source of information.

CC Comrie(1981)

Cuu Culy(1990)

DD Downing (1978)

GG Givón(1984)

KK Keenan(1985)

KCC Keenan & Comrie (1977)

LL Lehmann(1984)

PP Peranteau et al. (1972)

4

SS Smits (1988)

Iff the source is between brackets, it only mentions the facts without (extensive)

illustration.. Lehmann, Smits and Peranteau et al. are often much more detailed than

thee others. Note that there are cross-references between the cited authors; they

sometimess base themselves on the same primary sources, too.

GeneralGeneral annotations:

(...)) conditional or optional

11 first position

22 second position

ff final position

ss suffix

pp prefix

?? presumption, but not certain

++ in combination with

possiblyy separated

addd additional

CLL classifier

[blank][blank] no (clear) information from the sources consulted

Thee number of languages is 172.

Thee number of strategies described is 223.

Thee volume edited by Peranteau et al. contains the following contributions: H. Berman (Hittite, Yurok),, D. Adams (Ancient Greek), J. Ehrenkranz & E. Hirschland (Latin), Z. Goiab (Russian, Czech,, Ukrainian), V. Friedman (Slovenian, Serbo-Croatian, Macedonian, Bulgarian), A Loetscher (German),, J. Sadock (Danish), J. Morgan (Albanian, English), D. Perlmutter (French, (Arabic, Japanese,, Turkish)), F. Karlsson (Finnish), R. de Rijk (Basque), H. Aronson (Georgean), C. Killean (Arabic),, G. Gragg (Sumerian, Geez, Amharic, Oromo), E. Keenan (Malagasy, (Kalagan, Ivatan, Batakk Toba, Javanese, Malay, Ganda, Shona)), T. Givón (Kongo, Mbama, Kinuku, Swahili, (Hebrew,, Bambara, Amharic)), C. Masica (Fundi, Bengali, Gujarati, Telugu), J. McCawley (Japanese),, Y. Tagashira (Korean), J. Heath (Huichol, Tarahumara, Papago-Pima, Tepecano, Hopi, Tubatulabal,, Luiseno, Shoshoni), J. Rosenthal (Nahuatl).

(13)

Tablee 2. Characteristics of relative clauses.

language e Abkhaz z Akkadian n Akan n Albanian n Alekano o Ambaee (East) American n Sign n Language e —— % Amharic c Ancash h Quechua a Arabic c (classical) ) Arabic c (Tunisian) ) Arrernte e (Eastern) ) Ashéninca a Avestic c Ayacucho o Quechua a Bainouk k Bambara a Basque e Batakk Toba

s,o,v v

sov v

SVO O SVO O SVO O SVO O SOV V

sov v

vso o

sov v

vso o

sov? ?

sov v

SVO O

sov v

sovv /

(SVO) ) VOS S ace/ / erg g erg g ace e ace e erg g ace e ace e ace e erg? ? ace e ace e erg g Ca a se e --s --s s s s s . . V i i s+ + V) ) s s s s s+ + --s --s s+ + s s .. ? --V) ) --RC C type e pre e post t post t post t pre e post t cir r post t pre e prepar r cir r post t post t post t cir r post t cor r post t cir r pre e post t cor r post t pre e post t Det t Dm, , .. ? Dm, , --Dnis s --Dlp p __ ? --cD D --cD D --Dfs s Df f gap p --(GA.) ) GD D RPwh h --G D ? ? N N --N --N (GA.) ) (RP) ) N N --RP+N N RP P N N --NN + RPwh h RPwh h --(GDW W C/REL L RAtAgrX, , RCAT T RCSR R --RCSR R --RA(SRadd)p p RA(NRT)S S RA(NRT)S S RM M (RC) ) RAS S RA(add)s s RMCL L --RACNIU,^ ^ R-CSR R Source e L L L L (D) ) P,(D) ) (KC) ) (KC) ) Cu u (Cu) )

W D ) )

L,(CuJC) ) Cu,(LJC) ) (DX) ) L,P,(KC) ) (D) ) (Cu) ) G G (L) ) Cu u (Cu) ) (L) ) Cu,D,G,L L ,(C,K,P) ) (GX) ) L,P,(C,D, , K,KC) ) KC,P P

Theree is also a tone change which indicates subordination. Additionall clitic doubles are possible in both variants.

Americann Sign Language has also a kind of relative marker, but it is not clear to me how to classify it. .

Thee definite suffix belonging to the head noun is placed on the preceding RC verb. Case marking is verball only.

Thee relative pronoun (in cor and post) and the head noun (cor) are in situ, as in Maninka, Mandinka andd Vai. Almost all authors assume that Bambara has circumnominal relatives, too. These claims are basedd on Bird (1968). However, Culy (1990) clearly and extensively shows that this is a mistake. Thee relative affix is identical to the complementizer used in indirect questions. Hence it could be simplyy RCsiuf, as suggested in P(p. 117).

(14)

TYPOLOGICALL DATA 377 language e Bengali i Bicolano o Bora a Bulgarian n Burmese e Cahuilla a Catalan11 1 Chinese e (Mandarin) ) Coptic c Crow w Cuzco o Quechua12 2 Czech h Dagbani13 3 Danish h Diegueno14 4 Djirbal l Dogon n (Donnoo So) Dogon n (TogoKa) ) Dutch h Egyptian n (ancient) )

s,o,v v

SOV V

vso o

SOV V

svo o

SOV V SOV V

svo o

svoo /

(SOV) ) VSO O SOV V

svo o

SOV V

osv v

SOV V SOV V SOV V VSO O ace/ / erg g ace e ace e ace e ace e ace e (ace) ) ace e ace e (ace) ) ace e erg g ace e (ace) ) Ca a se e s s

P P

s s s s s s --. 9 9 s+ + s s s s --s+ + s s s+ + s s --RC C type e cor r post t post t post t post t pre e

JP?sL._ _

post„ar r post t pre e post t cir r post t cir r pre e post t cir r post t post t cir r cor r

JP9?t__. .

pre? ? post„„ „ cir r cir r post t post t Det t CD D Dl l _7 7 --_? ? Dl l --Df? ? Df f --Dl l Df+ + --Dl l Dfs+ + cD D Dl l D %+ + Df+ + Dl l gap p +N N RPm m --RP* * --.. ? --RPwb b (GD) ) N N --N --N --(GD2) ) RPwfa a N N --RPd/wh h N N N N GD D N N --N --N N N RPd/wh h --C/REL L --RCjp p RAfCUX X --RCSR R RCNRJ--RACNRT), , RCSR R --R C N --R / / R M ? ? --RC* * RCsR R R Q R R (RCSR) ) --(RA(Agr)p) ) --RA(T)S S --(RM) ) Source e P,(L) ) G G G G P P (L) ) (L) ) S,(KC) ) L,(D,G, , KJCC) ) (L) ) (L) ) Cu u (Cu) ) P, , (C^CJCC) ) CuJ. . (CiO.) ) P,S,(DJ.) ) CnX, CnX, (CJCJCC) ) .(L) ) (Cu) ) (Cu) ) D,L,(iyCC) ) Cu u Cu u S,(KC) ) (L) )

Seee the footnote on French.

Culyy reports Case attraction in both strategies: the RC or external head displays subordinate clause Case,, where it should have been main clause Case.

Thee particle which is always at the second position is classified as RCSR here. This may be correct if theree is always a topic/subject in SpecCP preceding it. Apart from this, there is a particle la followingg the RC. It marks the definiteness of the relative construction, hence Df+. This is strange, becausee normally definiteness is not expressed

(15)

language e English h Erzya a Éwé15 5 Farsi i Finnish h French h Frisian n (Northern) ) Fulfulde e (Adamawa) ) Gaelic c Gaididj j Ganda a Geez z Georgian n German n Godiéé 1B Greek k (modern) ) Greek k (ancient) )

s,o,v v

svo o

sov? ?

svo o

sov v

svo o

sov v

svo o

vso o

svo o

vso o

svo o

sov v

svo o

svoo /

vos s

svo o

ace/ / erg g (ace) ) ace e ace? ? (ace) ) (ace) ) ace e ace e ace e ace e ace e ace e Ca a se e --s --s s ? ? --s+ + --s --s s s s s --s --s s s RC C type e post t cor r post t post t cor r post t preoar r post t post t post t post t cor r err r post t post t post t post t post t post t post t cor r cir r Det t Dl l cD D Df f Dms s (+D1) ) --Dl l Dl l --cD D --D l l Dnis+f f Dl l --cD D gap p RPwh h --RPwb b +N N --(GD) ) RP P --RPvA A --GD D --N --N N N --(GAS) ) RPvA A --RPdAvfa a (GD) ) (GD) ) RPwh h RPwh h +...N N N N C/REL L --(RCSR) ) --RM M RCSR R --RCSR R --RCSR R RCSR? ? RCSR R RCSR R RA(SR/ / Agr^d),, , RM, , --RCSR R --R --R RCSR R --Source e CJ),GJC, , KC,LJ>,S S (L) ) L L

ex, ,

(D,K,KC) ) (DX) (DX) P, , (K,KC,L) ) P,S,(D,K, , KCX) ) (KC) ) (Cu,, KC) ( D l ) ) L L (L) ) P,(K,KC) ) P P P P G,K,L,P, , S,(DJCC) ) L L L,(CuJC, , KC) ) L,P,(C,D) ) (L) )

Thee particle looks like RCjp, but the plural marker is different according to Lehmann, hence it cannott be a complementizer (since complementizers do not intrinsically bear ^features).

L(p.58)) suggests that only subject-participials are possible, like in German and many other languages.. However, from other authors I conclude that Finnish has a real participial relative; see e.g.. P(p.l07,ex.9): an object participial. Still, there are severe restrictions on the use of this strategy. Italiann and French marginally allow for resumptive clitics (GA). Clitics in Rumanian and (most varietiess of) Catalan RCs do not occupy the gap, since i) those languages allow for clitic doubling, ii)) long relativization is impossible, iii) clitics may coocur with relative pronouns. The situation in Spanishh is ambiguous. See Smits (1988:56-60) for discussion.

Thee unclassified relative particle follows the verb. The RC is followed by another element, glossed ass 'specific', which seems to be the second part of a discontinuous determiner; hence Dnv+f, as in Yucatecan. .

(16)

TYPOLOGICALL DATA

_{379 9}

language e Green--landic c Gnarani i Gujarati i Hausa a Haya a Hebrew w Hewa20 0 Hindi i Hittite e Hopi i Huanuco o Huallaga a Quechua a Huanca a Quechua a Huichol l Hungana a 21 1 Hunganan n Hurric c Iban n Icelandic c Igbo o Ijo o Imbabura a Quechua a

s,o,v v

sov? ?

svo o

sov v

svo o

sov v

vsoo /

svo? ?

svo o

sovv /

osv v

svo o

sov v

ace/ / erg g erg g ace e ace e ace e acee / (erg) ) ace e ace e ace e ace e ace e erg g ace e ace e Ca a se e s s --s --s p p s s s s s s s s s+ + s s s+ + --s --s s+ + s s --s+ + s s RC C type e pOStpgr r post t cor r post t post t post t pre? ? cor r post t cor r Jpost... . cir r pre e cir r pre e cir r post t post t post t

je°?L__. .

jpre e cir r cor r post t post t post t pre e cir r pre e Det t --Dl l cD D D lp p --cD D cD D D l l cD D --D l l . ? ? Dl l cD D Dnis s Dfs s --gap p --RPSP P +N N (GDW W (GD) ) (GA*) ) N, , RPsp p +N N RPsp p RPwh h +N N --N --N --N --N N N --(GA.W W RP P --N --N ++ N --N --N --C/REL L RA(T)S S RA(add)s s --RCSR R RCSRjp p --RA(add)s s --RA(addsR)p p RMCL L ( R CSR ) ) RA(add)s s (RA(add)s) ) RCSR R --RA(NR)S S Source e L,(KC) ) D,(L) ) DJP,(L,K) ) KC,(CJD) ) (K) ) K,G,(CJ) ) ,KC,L,P) ) G G KJCC) ) P.CKCJ.) ) DX4» » (CuX) ) Cu u (Cu) ) (L) ) (L) ) L L (CJJJ.) ) L L (KC) )

s s

(L) ) L L Cu,(C^) ) C,(Cu) ) Seee the footnote on Tamil.

Hencee the head noun is used twice: in the main clause and the subordinate clause, with different Cases.. Nmam may be accompanied by or replaced by a demonstrative. Thus the whole construction lookss more like a correlative than a prenominal relative. However, the RC is not left-peripheral (as normall correlatives are), but in situ. This issue remains to be clarified

(17)

language e Indonesian n Italian2 3 3 Ivatan n Jacaltec c Japanese2 5 5 Javanese e Kalagan n Kalaa Lagaw Y a a Kannada a Kekchi i Kera a Khmer r (Central) ) Kinuku u Kiowa a Kiribati i K o m s o o Kongo o Korean n Kupsabiny y Lahu u Lakota a TT 2S Lango o Latin n Ligurian n (Genoese) )

s,o,v v

svo o

vso o

sov v

svo o

vso o

sov v

vsoo /

svo? ?

svo o

vos s

sov v

svo o

sov v

vso o

sov v

svo? ?

svo o

a c e / / erg g (ace) ) erg g ace e ace e ace e ace e erg g ace e ace e ace e Ca a se e --s --s s s s+ + --1 --1 s s Sm m s s --s --s .. ? f f . . s s R C C type e post t post t post t post t pre e cir r post t post t cor r prepar r cor r post t post t post t post t cir r post t post t post t pre e post t pre e jpost t cir r post t post t cir r post t Det t D f f D l l --cD D . 9 9 cD D D l l --D l „ „ D m m (Df) ) Df f Df+ + --gap p --R P ^ ^ --( G A . ) ^ ^ --N --N --( W \ * ) ) ++ N --RPwh h +N N --(GD)? ? --N --N (GD)? ? --(GA.) ) --N --N ( G A ^ ^ RPwh h N N (GD)? ? O R E L L RCs,, , RCS R R --R C » » (RA(add)s) ) --RAfNR^X X RCgo o RCjD D --RA(T)S S --RCS R R RC C RA(add)„ „ --RA(addX, , RA(T)S S R M C L L R C N R J J --RCSR R --Source e L,(D,K) ) S,(KCJ.) ) (P) ) (D,KC,L) ) G,L,P,(Cu, , D.K4CC) ) Cu,(L) ) (KCJ>) ) P P D,L,(K) ) L L (L) ) (K) ) (KC) ) (L) ) (P) ) (Cu) ) (KC) ) (L) ) (P) ) P, , (CJC^CC) ) (L) ) L L L L Cu u (seee m.) L,P,(D,K) ) (L) ) (KC) )

Keenann reports the use of R P ^ as a prepositional object (instead of RQp). Seee the footnote on French.

Moreoverr there is deletion of the agreement affix on V that corresponds to the relative gap. Theree is a debate concerning circumnominal relatives in Japanese. Murasugi (2000) claims that they doo not exist, i.e. that they are adverbial adjuncts that are misanalysed.

Seee the footnote on Tamil.

Theree is is whole series of tense markers specialized for relative clauses, hence the Korean RC is not participial. .

(18)

TYPOLOGICALL DATA

_{381 1}

language e Lushai i Macedonian n Malagasy y Malay29 9 Maninka/ / Mandinka a

Maori i

Marathi i Man31 1 Mbama a Mbum32 2 Mbundu u (Loanda) ) Minang--Kabau u Mohave e Mongolian n Moore e Murinh--Patha a Nahuatl33 3 Nama34 4

s,o,v v

sov v

svo o

vos s

svo o

sov v

vso o

sov v

sov? ?

svo o

sov v

sov? ?

svo o

sov v

svo o

sov v

ace/ / erg g erg g ace e (ace) ) acee / (erg) ) ace e ace e Ca a se e s s s s .. ? --__ 7 s s --s+ + s s --s --s ace?? j fs RC C type e post t pOStpar r post t post t post t cor r post t cor r post t prepar r post t post t post t post t cir r cor r pre,,» » err r post t cir r post t pre e Det t (Df) ) --Dl l Df f cD D cD D Dl l --DnH-f? ? Dfs s cD D Df+ + Df f Dl l Dfs s g»P P --RP* * --NN + RPSP P +N N RP» » --N --N N N N N --N --N --(GDW W C/REL L RA(add)s s RA(T)SS ? --RCSR R (RC») ) RCsp p --RA(T)S S RA(NRT)>> ? RACAT^ ^ RCSR R --(RA(Agr)p) ) --. ? ? (RCSR) ) --Source e (L) ) P P CKJCCLJ» » KC,P, , (CJ>) )

(oa) )

(KC) ) L,(D) ) (L) ) (L) ) (G4») ) (Cu) ) (L) ) (KC) ) L L (L) ) (L) ) Cu u (L) ) LJ> > (seee fn.)

Thee distinction between Malay and Indonesian is more political than linguistic. Hence see also the referencess on Indonesian.

Seee the footnote on Bambara.

Thee non-nominalizing tense-replacing affix is in fact not specialized for relativization.

32 2

Thee relative clause starts and ends with a particle, which might be compared with the discontinuous determinerss in Yucatecan and Godié, as suggested in Lehmann(p. 159), hence Dm+f However, if II understand correctly, simple nouns are not accompanied by these elements.

33 3

II think the agreement morphemes on the verb are like clitic doubles, not resumptive pronouns. P(p. 246)) exemplifies the use of relative pronouns (RPv*), but this involves free relatives (and possibly adverbiall relatives) only. Finally notice that Nahuatl uses a definiteness marker as a general subordinator,, hence its function is RCSR here.

34 4

Namaa is described in Hagman (1973), Olpp (1977) and Rust (1965). There is a determiner suffix and aa construction-final (rudimentary) Case ending. (There can also be an additional initial demonstrative.)) A relative clause contains a gap, unless in postpositional contexts, where a resumptivee pronoun shows up. Hagman (1973:232) shows that in some of these cases the

(19)

language e Navaho o Norwegian n Ojibwa35 5 Oromo o Oskian n Palauan n Papago--Pima a Polish h Ponapean n Portuguese e Roviana a Rumanian n Russian n Russian n (Medieval) ) Rwanda a Saho-Afar r Sanskrit Sanskrit (Vedic) )

s,o,v v

sov v

svo o

sov v

svo o

vso o

svo o

svo? ?

svo o

sov v

ace/ / erg g (ace) ) ace e (ace) ) ace e ace e ace e ace e Ca a se e --s --s (s) ) --s --s s s s s --s --s RC C type e cir r pre e post t pOStpar r post t pre e post t post t pre e .JP9?L-_. . pre e post t post t post t post t post t post t cor r post t pre e cor r cir r post t pre e Det t --Dl(+) ) Dm, , Dms s Dfs s --Dfs s Dl l Dl l D l l --cD D --cD D gap p N N --RPd/wh h --(GD) ) RPwh h --RPwh h --RPwh h RPwfa a --RPwh h +N N --RPSP P +...N N N N C/REL L RAfNRaddX X (RCS R) ) --RCT), , ( R C S R / ) ) ++ (R) (RCSRJ-) ) + R R --RCso o RC» » R CS R R --R C S --R --R RCSR R --R CS R R (RA(ditic)pU U --Source e CuAL,(K) ) (CuJXK^) ) S S

D,a> >

P,(L) ) (L) ) (L) ) (D) ) P,(KC) ) G G S,(L) ) KC C S,(D,KC) ) C,P, , (D,KC,L) ) (K) ) G,(CJC) ) (L) ) D,L L (L) ) .... continued

postpositionn can be deleted instead of inserting a resumptive pronoun, but it is not confirmed in the otherr grammars. Furthermore, free relatives are always false free relatives. Interestingly, all three authorss mention a construction which shows a false free relative mat is in apposition to a DP, and thereforee has an appositive meaning. This is mistakenly referred to as a postnominal relative in Lehmannn (1984:103). The FR in apposition behaves as usual, exept for one mysterious property: it hass an optional, clause-initial relative complementizer particle.

Inn fact, what is called a relative tense particle here, is a specialized auxiliary. Perhaps it must be consideredd as RCSR nowadays.

Apartt from the optional relative complementizer and an optional resumptive pronoun, there is a relativee particle R (derived from kana 'this') which has a free position (!) in the relative clause. In postnominall RCs it is optional, in prenominal RCs obligatory.

37 7

(20)

TYPOLOGICALL DATA

_{383 3}

language e Schwyzer--dütsch h (Zurich) ) Serbo--Croatian n Shona a Shoshoni i Sinhala a Slovenian n Spanish3 9 9 Sumerian n Swahili i Swedish h Tagalog g Tamil l Telugu u Tewa a (Arizona) ) Thai i Tibetan n Tigré é Tongan n

s,o,v v

sov v

svo o

sov v

svo o

sov v

svo o

svo? ?

sov v

svo o

sov v

vso o

ace/ / erg g ace e ace e ace e ace e (ace) ) erg g (ace) ) ace e ace e erg g erg g Ca a se e s s s s --s+ + s s --s --s --s/f f ? ? V) ) s<+) ) s s f f RC C type e post t post t post t post t pOStpar r pre e post t post t post t post t post t J P 9 Ë . . . . pre e cor r .JPregar... . cor r cir r post t JPrejiar... . cir r post t pre e post t Det t Dl l --_? ? --Dl l --Etas s . ? ? _ ? ? cD D cD D Df f D l l D l l gap p (GD) ) RP\vh h --(GA.W W --RPg, , (GD) ) RPwb b --(GA.)^ ^ --RF\vb b --RPwh h + N N RPwh h N N --N --N --(GA,)^ ^ C/REL L R C S R R --RCSR R (RACAgr.jd)) ) RACNRaddX X RA(T)S S R CS R R R C S R R --RA(NRa d d)s s RACCL.diV» » RCgp+CLs s ( R C S R ) ) --RCsR^s s RsR.sJ J RA(T)S S RA(T)S S RCjp p RACNRT)»/ / RAp p --Source e S,(KC) ) P,(K) ) (KC^J») ) L L (L) ) (KC) ) P,(K,KC) ) S,(KCX) ) LJ» » L,P,(K) ) S,(K,KC, , L) ) (C,K,KC, , L) ) L,(KC) ) (L4CC) ) L,P P (Cu) ) (L) ) G,K,(DX) ) (KX) ) '(D) '(D) (L) ) KC C

Regardingg RA(T), see the footnote on Tamil. Seee the footnote on French.

II don't know if the tense-replacing affix is the normal participial form or a form specialised for relativisation.. Lehmann's glosses suggest the former.

(21)

language e Turkish42 2 Tzeltal l Ukrainian n Umbrian n Urhobo o Utee / Paiute Vai i Vietnamese e 44 4 Wappo o Warlpiri i Welsh h Wolof46 6 Yaqui i Yavapai i Yoruba a Yucatecan n Yurok48 8 Zapoteco o

s,o,v v

sov v

vos s

svo o

sov? ?

vso? ?

sov v

svo o

sov v

vso o

svo o

sov v

svo o

vso o

ace/ / erg g ace e ace e ace e ace e erg g ace e ace e Ca a se e ) ) --s --s --s --s --s --s s s --S<+) ) s+ + --RC C type e prep* * post t post t post t post t post t post t cor r post t cor r cir r cor r post t post t post t cir r post t post t post t pre e post t Det t --Dl l --Dm m D m ? ? cD D cD D --cD D Dl l Df? ? Dl l Dfs s Df f Dl+fs s Dl l Dl l gap p --RPwh h R P ^ ^ --GD D --N+ + RPwh h --N --N N N N N (GA,) ) .. ? .. ? N, , --R P ? ? C/REL L RA(T)S S RA(NRT)S S --RCSR R RCSRj) ) RAfNRaddX X --RCSR R --RCSRJ, , (RCSR) ) RMCL L RA(add)s s (RA(add)p) ) RCg,, , --RCSR R --Source e D,G,L,(C C ,KJCC,P) ) (D) ) (K) ) P P (L) ) (K4CC) ) G G L L

OW W

L,(K) ) (CuJCX) ) DX,(C, , K4CC) ) (K,KC,L) ) L L L,(K) ) L L (K,KC,L) ) L L P P (D) )

Figuree 1 shows the distribution of the relative clauses in the sample on a world map. II must repeat here that statistical conclusions cannot be based on this sample without furtherr processing. Nevertheless some general statements are justified. Postnominal relativess are dominant around the world. The other types (prenominal, circum-nominall and correlative) are rarer, but they do occur in different language families in differentt parts of the world.

RA(T)) is the normal participial form. It is used if the head is subject or genitive in the relative clause.. Otherwise, the nominalization strategy is used.

Seee the footnote on Bambara.

NN is in situ or fronted as a topic. There is no relative pronoun.

NN is in situ or fronted as a topic. There is no relative pronoun. RCSR is a prefix on AUX; it may be precededd by topic/subject (cf. Dagbani).

Thee classifier is combined with a determiner.

Yucatecann has discontinuous determiners, as indicated, cf. Godié.

Inn addition, the verb may be in attributive mood, which is reserved for relative clauses. The definite articlee equals the relative subordinator. Hence the question is whether D l or RCSR is missing in prenominall relatives. In my view it makes more sense that RCSR fails - assuming that the RC splits DD and N - but Berman (P (p. 257)) suggests the opposite, i.e. D fails.

42 2 43 3 44 4 45 5 46 6 47 7 48 8

(22)

TYPOLOGICALL DATA

% , ,

A$$ <

" t ff f' o

\ 3 é é

1 1

(

%:. %:.

ff

D

-J -J

* v .:

.* *

ww ï:-: rt:'* ^ ; ' C M M JJ p i - ,

,'

!

>« «

'' :' , s?ü ;- < S S , 211 1

*\1; ;

^ry y

.... f #'' " S ï t

ff \

(23)

Thee remaining tables in this appendix contain useful selections of table 2, the main

table.. Footnotes and information on strategies not belonging to particular selections

aree not repeated and must be looked up in table 2 if relevant.

Tabless 3 through 7 contain all main types of relatives.

Tablee 3 is a hst of all languages in the sample with circumnominal relatives. There

aree no relative pronouns, because these would trigger the promotion of the head,

whichh would lead to another main type. Notice that circumnominal relatives with a

frontedfronted internal head as a main strategy are found only in Yavapai. However,

accordingg to Lehmann (1984:121) it is a secondary strategy in Gaididj, Mohave,

Diegueno,, Latin and Sanskrit.

49

Tablee 3. Circumnominal relatives.

language e American n Sign n Language e Ancash h Quechua a Arrernte e (Eastern) ) Ayacucho o Quechua a Crow w Cuzco o Quechua a Dagbani i Diegueno o Dogon n (Donnoo So) Dogon n (Togoo Ka) Gaididj j Greek k (ancient) ) Hopi i Huanuco o Huallaga a Quechua a

s,o,v v

svo o

sov v

vso o

sov v

svo o

sov v

svo o

sov v

ace/ / erg g ace e erg? ? ace e ace e ace e ace e ace e ace e ace e Case e s+ + s+ + s+ + . ? ?

--s+ + s s

--s+ + s s s s s+ + RC C type e cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r Det t

--_? ?

--Df? ?

--Df+ + Dfs+ + Dr}s)+ + Df+ +

--gap p N N N N N N N N N N N N N N N N N N N N N N N N N N N N C/REL L RA(NRT)S S RAS S

--RCSR R (RA(Agr)p) )

--RA(add)s s

--other --other strat--egies s post t post, , pre™ ™ pre e post t pre e post t pre, , post, , cor r cor r post, , cor r pre, , post t pre e

(24)

TYPOLOGICALL DATA 387 language e Huanca a Quechua a Hurric c Imbabura a Quechua a Japanese e Kiowa a Lakota a Latin n Mohave e Moore e Murrinh-Pata a Navaho o Sanskrit Sanskrit (Vedic) ) Tewa a (Arizona) ) Tibetan n Wappo o Yavapai i

s,o,v v

sov v

sovv /

osv v

sov v

sov? ?

sov v

svo o

sov v

ace/ / erg g ace e erg g ace e ace e ace e ace e ace e erg g ace e ace e Case e s+ + s+ + s+ + s+ + --s --s s+ + --s --s s s s s s+ + RC C type e cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r cir r Det t --Df+ + Dfs s Df+ + --Df f --Dfs s g»P P N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N N, , C/REL L RA(add)s s RA(NR)S S RACNIUOs s --(RACAgrW W --RA(NRadd)t t --(RA(add)p) ) other other stmt--esies s cor, , pre, , post t pre e pre e post t post t cor r post t pre e cor, , pre, , post t post, , pre™ ™ cor r

(25)

Thee following table contains a list of all correlatives in the sample. Notice that there

iss no relative pronoun in Diegueno, Gaididj, Mohave, Wappo and Warlpiri.

Tablee 4. Correlatives.

language e Avestic c Bambara a Bengali i Diegueno o Erzya a Farsi i Gaididj j Greek k (ancient) ) Gujarati i Hindi i Hittite e Hurric c Kalaa Lagaw Ya a Kannada a Maninkaa / Mandinka a Marathi i Mohave e Russian n (Medieval) ) Sanskrit t (Vedic) ) Tamil l Telugu u Vai i Wappo o Warlpiri i

s,o,v v

sov? ?

sov v

sov? ?

sov v

svo o

sov v

sovv /

osv v

sov v

svo? ?

sov v

ace/ / erg g ace e ace e ace e ace e ace e ace e ace e acee / (erg) ) ace e erg g ace e ace e acee / (erg) ) ace e ace e ace e ace e ace e ace e erg g Case e s s --s --s s s s s s+ + s s s s s s s s s+ + s s Sm m s s s s s s s s Sf+) ) Sf+1 1 --s --s s s R C C type e cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r Det t cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D g»P P RPP + N N + R P ^ ^ R P ^ + N N N N R P ^ + N N N N RPwh h + . . . N N R P ^ + N N R P S P + N N R P ^ + N N R P ^ + N N (RPwh) ) ++ N R P ^ + N N N + R P ^ ^ RPsp+N N GN N RP^h+N N R P * * + . . . N N RPwh+N N RPwh+N N N + R P ^ ^ N N N N C / R E L L --RCS R R --(RA(add),) ) --R C S --R .0 0 other r strat--egies s post t post t post t cir, , post, , pre e post t cir r post, , cir r post t pre, , cir, , post t pre™ ™ post t cir r cir, , pre, , post t pre™ ™ pre™ ™ cir r

(26)

TYPOLOGICALL D A T A 389

Tablee 5 contains a list of all prenominal relatives in the sample.

Tablee 5. Prenominal relatives.

language e Abkhaz z Alekano o Amharic c Ancash h Quechua a Ayacucho o Quechua a Basque e Burmese e Chinese e (Mandarin) ) Cuzco o Quechua a Diegueno o Finnish h Hewa a Hopi i Huanuco o Huallaga a Quechua a Hurric c iio o Imbabura a Quechua a Japanese e Kannada a Korean n Lahu u Mari Mari Mongolian n Nama a Navaho o Oromo o Palauan n

s,o,v v

sov v

sovv /

(SVO) )

sov v

svoo /

(SOV) ) SOV V

sov v

svo o

sov v

sovv /

osv v

sov v

sov? ?

sov v

svo o

ace/ / erg g erg g erg g ace e ace e erg g ace e ace e ace e ? ? ace e ace e ace e erg g ace e ace e ace e ace e ace e ace? ? Case e --s --s --s<+> > s s S<+) ) --s --s s s s ? ? s s s s s s s+ + --s --s s s SM M s s f f

u u

--RC C type e pre e pre e pre e p r e ^ ^ pre e pre e pre e pre e pre e pre? ? p r e ^ ^ pre? ? pre e pre e pre e pre e pre e pre e prewar r pre e pre e prepar r pre™ ™ pre e pre e pre e pre e Det t Dm, , --Dffls s --Dfs s .. ? --Dfs s --. ? ? Dm m (DO O Dfs s --Dfs s gap p --. ? ? (GD) ) --N --N --Ni i --(GDW W --C/REL L RACAjsrt, , R A f S R ^ ^ RA(NRT)S S RACNIWtf f RCNRJ--RCMRJ J --RA(add)s s RA(add)s s --RA(NR)S S --RA(T)S S RA(T)S S RCNRJ J RA(T)S S RA(NRT)SS ? --RAfNRaddX X ( R CSR / ) ) ++ R other r strat--egies s cir, , post t cir r cir r cir, , post, , cor r post t cir, , post t cir r cir, , post, , cor r cir r cir r cor r cir r post t post t

(27)

language e Papago--Pima a Saho-Afar r Sanskrit t (Vedic) ) Sinhala a Tagalog g Tamil l Telugu u Tibetan n Tigré é Turkish h Yurok k

s,o,v v

sov v

sov? ?

sov v

ace/ / erg g ace e ace e ace e erg g ace e Case e s s s/f? ? S,+> > S(+> > s s V) ) --RC C typ« « pre e pre e pre e pre e pre e pre™ ™ pre™ ™ prepar r pre e prepar r pre e Det t .. 9 .. 9 .. ? Dl l --Dl l g»P P --C/REL L --RsR.s/ / RA(T)S S RA(T)S S RA(NRT)^ ^ RA» » RA(T)S S RA(NRT)S S --other r strat--egies s post t cor, , cir, , post, , post t cor r cor r cir, , post t post t post t

Thee following table contains a list of all participial relatives in the sample, both

prenominall and postnominal.

Tablee 6. Participial relatives.

language e Ancash h Quechua a Finnish h Kannada a Man n Mongolian n Tamil l Telugu u Tibetan n Turkish h Cahuilla a Djirbal l Greenlandic c Lushai i Ojibwa a Shoshoni i

s,o,v v

sov v

svo o

sov v

sov? ?

sov v

osv v

sov? ?

sov v

svo o

sov v

ace/ / erg g ace e ace e 9 9 acc c ace e acc c erg g acc c acc c erg g erg g erg g acc c Case e V) ) s? ? S(+) ) Sm m sm m s s V) ) s s s+ + s s s s s+ + RC C type e prep„ „ prepar r pre™ ™ prepar r pre™ ™ preDar r preDar r prepar r prepar r post™ ™ post™ ™ post™ ™ post™ ™ post™ ™ post™ ™ Det t --. 9 9 _? ? .. 9 .. ? --Dl l --gap p --C/REL L RA(NRT)S S --RA(T)S S RA(T)S S RA(NRT)S? ? RA(T)S S RA(T)S S RACNRrW W RA(T)S S RA(NRT)S S RA(NRT)S S RA(T)S S RA(T)S S RA(T)S? ? R(T), , RA(T)S S other r strat--egies s cir, , post t post t cor r cor r cor r cir, , post t post t post t post t post t

(28)

TYPOLOGICALL DATA 391

Tablee 7 is a list of all postnominal relatives in the sample. This is the largest group.

Tablee 7. Postnominal relatives.

language e Akkadian n Akan n Albanian n Ambaee (East) American n Sign n Language e Ancash h Quechua a Arabic c (classical) ) Arabic c (Tunisian) ) Ashéninca a Avestic c Bainouk k Bambara a Batakk Toba Bengali i Bicolano o Bora a Bulgarian n Cahuilla a Catalan n Coptic c Crow w Czech h Dagbani i Danish h Diegueno o Djirbal l Dutch h Egyptian n (ancient) )

s,o,v v

sov v

svo o

sov v

vso o

sov? ?

svo o

sov v

vos s

sov v

vso o

sov v

svo o

sov v

svo o

vso o

svo o

sov v

osv v

sov v

vso o

a c e / / erg g ace e ace e ace e ace e ace e ace e ace e ace e ace e ace e ace e (ace) ) (ace) ) ace e erg g (ace) ) Case e s s s s S(+) ) s s s s --s --s . ? ? --s --s p p s s s s s s --.. ? --s --s s+ + --R C C type e post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post post pOStoar r post t post t post t post t post t post t post t post™ ™ post t post t Det t - ? ? Dm, , D lp p --Df f D l l . 9 9 --D l l Df f D l l --D l l D l l D l l gap p (GA,) ) GD D RPwh h --G D ? ? (GAJ J (RP) ) --RP P --RPwh h (GDW W R P . . --RP* * --RPwh h --(GD,) ) RPwh h --RPdAvh h GD D --RPd/wfa a --C/REL L RCAT T R Q R R --RCSR R RM M (RC) ) RA(add)s s R M C L L --RCSR R --RC» » RMCL^X RMCL^X --RCsR R RA(NRT)S S RCSR R --R M ? ? R Ca a RCSR R ( R CSR ) ) --RA(T)S S --(RM) ) other r strat--egies s ch--oir, , pre™ ™ cor r cor r cor r cir r cir r cir, , cor, , pre e

(29)

language e Fnglish h Ewé é Farsi i Finnish h French h Frisian n (Northern) ) Fulfulde e (Adamawa) ) Gaelic c Ganda a Geez z Georgian n German n Godié é Greek k (modern) ) Greek k (ancient) ) Green--landic c Guarani i Hausa a Haya a Hebrew w Hindi i Hopi i Huichol l Hungana a Hungarian n Hurric c Iban n Icelandic c Igbo o Indonesian n

s,o,v v

svo o

sov v

svo o

sov v

svo o

vso o

svo o

vso o

svo o

sov v

svo o

svoo /

vos s

svo o

sov? ?

svo o

sov v

vsoo /

svo? ?

svo o

sovv /

osv v

svo o

a c e / / erg g (ace) ) ace e a c e ? ? (ace) ) (ace) ) ace e ace e ace e ace e ace e eig g ace e acee / (erg) ) ace e ace e erg g ace e Case e --s --s s ? ? --s --s s s s s --s --s s s s s --p --p s s s s --s --s s+ + s s --R C C type e post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t postp„ „ post t post t post t post t post t post t post t post t post t post t post t post t post t post t Det t D l l Df f D ms s (+D1) ) D l l D l l --D l l Dms+f f D l l --D l l D lp p D l l D l l .. ? D l l D ms s Df f gap p RPwh h --(GD) ) RP P --RF\vh h --GD D --(GAS) ) RPwb b --RPdAvh h (GD) ) (GD) ) RPwh h --(GDW W (GD) ) (GAS) ) RPSP P --( G A ^ ^ R P P --C / R E L L --(RCS R) ) R M M R C S R R --RCSR R --RCS R R R CS R? ? R C S R R RA(SR/ / AgTadd),, , RM„ „ --R C S --R --R --R --R R C S R R --RA(T)S S RA(add)s s R C S R R RCSR,P P --RA(add)s s RA(addSR)p p R M C L L (RCS R) ) RA(add)s s RCS R R RCsp p other r strat--egies s cor r pre„ar r cor, , cir r cor r cir, , pre e pre, , cir, , cor r

(30)

TYPOLOGICALL DATA

_{393 3}

language e Italian n Ivatan n Jacaltec c Javanese e Kalagan n Kekchi i Kera a Khmer r ^Central) ) Kinuku u Kiribati i Komso o Kongo o Kupsabiny y Lakota a Lango o Latin n Ligurian n (Genoese) ) Lushai i Macedonian n Malagasy y Malay y Maori i Marathi i Mbama a Mbum m Mbundu u (Loanda) ) Minang--Kabau u Moore e Nahuatl l Norwegian n Ojibwa a Oromo Oromo Oskian n Palauan n

s,o,v v

svo o

vso o

svo o

vso o

vsoo /

svo? ?

svo o

vos s

sov v

svo o

vso o

sov v

svo? ?

svo o

sov v

svo o

vos s

svo o

vso o

sov v

svo o

sov v

svo o

ace/ / erg g (ace) ) erg g ace e erg g ace e erg g ace e (ace) ) acee / (erg) ) ace e (ace) ) Case e --s --s --1 --1 s s --- ? ? --s --s s s s s . ? ? --- ? ? s s --s --s --RC C type e post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t posW W post t post t post t post t post t post t post t post t post t post t post t post t pOStpar r post t post t post t Det t D l l Dl l --D1D D Df f --(Df) ) --Dl l Df f Dl l --DnH-f? ? Df f Dl l Dl(+) ) Dnis s Dnij j g»P P --RPwh h --(GA,^ ^ --(GD)? ? --(GD)? ? --(GA.) ) --(GA.^ ^ RPwfa a (GD)? ? --RPg, , --RP* * --RPd/wh h --(GD) ) C / R E L L R CS R R --RCgo o (RA(add)t) ) RC» » RC» » RCsR R RC C RA(add)„ „ --RA(add)„ „ R M cL L --RCsR R --RA(add)s s R A ( T )S? ? --RCsR R (RC.) ) RQ,, , --RCSR R --__ ? ( R CSR ) ) (RCS R) ) --R(T)! ! (RCSRX> > ++ (R) other r strat--egies s cir r cir r cor r cir r pre e pre e

(31)

language e Papago--Pima a Polish h Ponapean n Portuguese e Roviana a Rumanian n Russian n Rwanda a Sanskrit Sanskrit (Vedic) ) Schwyzer--dütsch h (Zurich) ) Serbo--Croatian n Shona a Shoshoni i Slovenian n Spanish h Sumerian n Swahili i Swedish h Tagalog g Thai i Tibetan n Tongan n Turkish h Tzeltal l Ukrainian n Umbrian n Urhobo o Utee / Paiute Vietnamese e

s,o,v v

SVO O

svo o

SVO O

vso o

svo o

sov v

svo o

sov v

svo o

sov v

svo o

svo? ?

svo o

sov v

vso o

sov v

vos s

svo o

sov? ?

vso? ?

svo o

a c e / / erg g ace e (ace) ) ace e ace e ace e ace e ace e ace e ace e (ace) ) erg g (ace) ) erg g erg g ace e ace e ace e Case e s s (s) ) --s --s s s --s --s s s s s --s+ + s s --s --s --s/f? ? s s f f V) ) s s --s --s RC C type e post t post t post t post t post t post t post t post t post t post t post t post t post t POStpar r post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t Det t --Dfs s Dl l Dl l Dl l --Dl l --.. 9 --Dl l --Dms s .. ? Dl l D l l --Dm m D m ? ? g»P P RPwh h --RPwh h --RPwh h --RPwb b --(GD) ) RJ*wh h --( G A ^ ^ --RPg, , (GD) ) --RPvA A --( G A J ^ ^ --RPwh h --( G A ^ ^ RPwh h RPwh h --GD D --C/REL L --RC» » RC» » RCSR R --RCSR R --RCSR R --RCSR R (RA(cünc)pU U RCSR R --RCSR R (RACAgiw)) ) RACNRaidX X R A ( T )S S R CS R R R CS R R --RACNRadd), , RACCLadd)^ ^ RCsp+CLs s (RCSR) ) --RCsR.s s RC» » --RCsR R RCSR_D D RACNR^Js s R C S R R other r strat--egies s pre e cor, , cir, , pre e pre e cir, , pre™ ™ preoar r

(32)

TYPOLOGICALL DATA

_{395 5}

language e Welsh h Wolof f Yaqui i Yoruba a Yucatecan n Yurok k Zapoteco o

s,o,v v

vso o

svo o

sov v

svo o

vso o

a c e / / erg g ace e Case e --S(+) ) --R C C type e post t post t post t post t post t post t post t Det t D l l D f ? ? D l l D f f Dl+fs s D l l gap p (GAJ J . ? ? _ ? ? --R P ? ? C/REL L (RCSR) ) R M C L L RA(add)s s RCa, , --R CS--R --R other r strat--egies s pre e

Thee following eight tables contain lists of languages with relative pronouns,

particles,, etc.

Tablee 8 contains all languages in the sample that have relative pronouns. The table is

dividedd into two sections: correlatives and postnominal relatives. The other main

typess do not have relative pronouns.

Tablee 8. Relative pronouns.

language e Avestic c Bambara a Bengali i Erzya a Greek k (ancient) ) Gujarati i Hindi i Hittite e Hurric c Kalaa Lagaw Ya a Kannada a Maninka/ / Mandinka a Marathi i Russian n (Medieval) ) Sanskrit t (Vedic) ) Tamil l

s,o,v v

sov? ?

sov v

sov? ?

svo o

sov v

sovv /

osv v

sov v

svo? ?

sov v

a c e / / erg g ace e ace e ace e ace e acee / (erg) ) ace e erg g ace e ace e acee / ace e ace e ace e Case e s s --s --s s s s s s s s s s+ + s s S M M --s --s s s s s V> > R C C type e cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r cor r Det t cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D cD D gap p RPP + N NN + R P ^ RPso+N N RPwh+N N + . . . N N RPsn+N N R P S P + N N RPw h+N N R P ^^ + N (RPw h)+N N RPwh+N N NN + R P ^ RP.P+N N R P ^ + N N RPsp+.-N N RPwh+N N C/REL L --(RA(add)s) )

(33)

--language e Telugu u Vai i Albanian n Arabic c (Tunisian) ) Avestic c Bambara a Bengali i Bulgarian n Catalan n Czech h Danish h Dutch h Enghsh h Finnish h French h Georgian n German n Greek k (ancient) ) Hindi i Hungarian n Italian n Latin n Macedonian n Marathi i Norwegian n PoUsh h Portuguese e Rumanian n Russian n Serbo--Croatian n Slovenian n Spanish h Swedish h Tzeltal l Ukrainian n Zapoteco o

s,o,v v

SOV V

sov v

svo o

VSO O

sov? ?

SOV V SOV V SVO O

svo o

sov v

svo o

sov v

svo o

sov v

svo o

svo? ?

svo o

sov v

svo o

vos s

svo o

vso o

ace/ / erg g ace e ace e ace e ace e ace e ace e (ace) ) ace e (ace) ) (ace) ) (ace) ) ace? ? (ace) ) ace e ace e ace e acee / (erg) ) ace e (ace) ) ace e ace e acee / (erg) ) (ace) ) ace e (ace) ) ace e ace e ace e ace e (ace) ) (ace) ) ace e Case e * + > > s s s s s s --s --s s s --s --s --s ? ? --s --s s s s s s s s s --s --s s s s s --s --s --s --s s s s s s s --s --s R C C type e cor r cor r post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t post t Det t cD D cD D Dm* * --Dl l --Dl l Dl l Dl l Dl l Dl l --Dl l Dl l --Dl l Dl l Dl l --Dl l Dl(+) ) Dm, , --Dl l Dl l --Dl l Dms s Dl l --gap p RPwh h N + R P ^ ^ RPwh h (RP) ) RP P RPwh h RPsp p RPsD D RPwh h RPwh h RP<i/wh h RPd/wb b RPwh h RP P RPwh h RPwh h RPd/wh h RPwh h RP* * RP P RPwh h RPwh h RP™, , RPsp p RPd/wh h RPwh h RPwh h RPwh h RPwh h RPwh h R P » » RPwh h RPwh h RPwh h RPwh h R P ? ? C/REL L --(RC) ) --(RCSR) )