Category Archives: Articles on Balochi language

Impersonal Constructions in Balochi

(Research Paper)

Prof. Dr.Carina Jahani
Uppsala University, Sweden
Serge Axenov, St. Petersburg, Russia

Behrooz Barjasteh Delforooz
Uppsala University, Sweden
and University of Sistan and Baluchestan,
Zahedan, Iran

Maryam Nourzaei
University of ‘Olum va Ta
qiqāt, Fars, Iran


Impersonal constructions are interesting from a typological perspective. Siewierska (2008: 3–4) finds that “[t]he semantic characterizations of impersonality centre on two notions”, either “the lack of a human agent controlling the depicted situation or event” or “situations or events which may be brought about by a human agent but crucially one which is not specified.” The present article focuses on grammatical constructions for situations or events brought about by a non-specified agent in one Iranian language, namely Balochi. It draws upon four Balochi corpuses available to the authors, comprising four different dialects of Balochi and consisting of altogether approximately 130,000 words.

There are three constructions for a non-specific agent found in the corpus, those with the verb in 3PL, those with the verb in 2SG, and those with a passive verb. It seems that the 3PL construction allows the speaker to distance himself/herself from the event somehow in narrative texts, where the speaker and addressee are not included in the referential framework of this construction. The 2SG construction, on the contrary, allows an unrestricted impersonal interpretation in narrative texts. However, in procedural texts, the 2SG and 3PL constructions are used interchangeably to include the speaker, and probably also the addressee.

The 2SG construction in narrative texts and the 2SG and 3PL constructions in procedural texts are open to a truly impersonal interpretation. Thus, the 3PL construction does follow the referential properties described by Siewierska (2008: 14–17) in narrative texts but has wider referential properties in procedural texts. In Balochi, the referential properties of the passive construction seem, on the contrary, not to be as unrestricted as Siewierska (2008: 23) suggests.


1. Introduction

Impersonal constructions are interesting from a typological perspective.1 Onishi (2001: 45) notes that impersonal constructions need to be investigated for a large number of languages with different typological profiles and from different linguistic areas in order to draw far-reaching conclusions about these kinds of constructions and the typological constraints that apply to them. Siewierska (2008: 13–14) also pays attention to the lack of data for impersonal constructions in, e.g., grammatical descriptions of specific languages. The aim of the present article is to provide data concerning impersonal constructions for one such specific language, namely Balochi.2

Balochi is an Iranian language, thus belonging to the Indo-European language family, and is spoken in south-eastern Iran, south-western Pakistan, and southern Afghanistan, as well as in the UAE, Oman, and other parts of the Arabian Peninsula, Turkmenistan, India, and East Africa. In her article “Ways of Impersonalizing”, Siewierska (2008: 3–4) discusses the concept ‘impersonal’ and finds that this term has been used in a wide and not entirely well defined sense. Some scholars “conceive of impersonality in semantic terms, others adopt a syntactic approach, and yet others a morphological perspective.”

She finds that “[t]he semantic characterizations of impersonality centre on two notions”, either “the lack of a human agent controlling the depicted situation or event” or “situations or events which may be brought about by a human agent but crucially one which is not specified.” In the second category she pays particular attention to third person plural impersonal constructions and verbal impersonals and to the referential properties of these different constructions. Among Siewierska’s conclusions are that the referential range of the 3PL construction is more restricted than for other constructions and that 3PL impersonal constructions denote third person referents among which the speaker and/or addressee are hardly ever included.3 She also concludes that verbal impersonals are generally of a less restricted character when it comes to referential properties, and normally include the speaker, and that the most open reference is found in agentless passives, which she finds referentially unrestricted (Siewierska 2008: 23).

Blevins (2006) takes a morphological approach and describes the characteristics of morphologically marked impersonal constructions found in, e.g., Balto-Finnic and Celtic languages, which “represent a distinctive grammatical strategy for ‘suppressing’ reference to the subject” (Blevins 2006: 236). There are no morphological impersonal constructions of this kind (see also Blevins 2003: 486–489) in Balochi. Kitagawa and Lehrer (1990), on the other hand, ground their study in different uses of pronouns and define the concept of impersonality in semantic terms in connection with these pronouns. They discuss the distinction between referential, impersonal and vague uses of pronouns, where “[r]eferential uses identify specific individuals” , “[a]n ‘impersonal’ use of a pronoun applies to anyone and/or everyone”, and “[a] ‘vague’ use applies to specific individuals, but they are not identified, or identifiable, by the speaker” (Kitagawa and Lehrer 1990: 742).4 The distinction between ‘impersonal’ and ‘vague’ uses will here be applied to whole constructions rather than only to pronouns, particularly since Balochi is a pro-drop language.

The present article takes a semantic approach and focuses on the second category specified by Siewierska, namely situations or events brought about by a non specified (impersonal or vague) human agent. This should, however, not be taken as the position of the present researchers on what should be regarded as impersonal constructions.

Another interesting type of impersonal construction comprises those with a non-canonical subject, that is, a subject in the genitive or dative case (see Onishi 2001), which are frequent in Balochi and which will be the subject of a forthcoming study.
This investigation draws upon four Balochi corpuses available to the authors, namely Serge Axenov’s corpus of tales, reality-based stories, and procedural texts (dealing with, e.g., weaving, cooking, farming, etc.) from Turkmenistan (abbreviated BT), Behrooz Barjasteh Delforooz’s corpus of tales and reality-based stories from Sistan (abbreviated BS), Maryam Nourzaei’s corpus of tales, a reality-based story, and a procedural text in Koroshi, a dialect of Balochi spoken in Fars (abbreviated BK), and Carina Jahani’s corpus of modern short stories from Pakistan (abbreviated BP). These four corpuses comprise approximately 130,000 words.5 The texts that contain the greatest number of impersonal constructions with a non-specified human agent are procedural texts, but the written texts (BP) also have a considerable number of such constructions. The tales contain, on the whole, fewer impersonal constructions. The examples are transcribed in a phonemic representation modelled on the system used by Jahani and Korn (2009).6 All examples are marked for dialect.

Different constructions for a non-specified human agent found in the corpus will be classified and discussed below. Special attention will be given to the referential properties of the agent in each construction. There could be a vague delimitation of possible agents in the context, but there could also be a totally impersonal reference to any possible agent, including the speaker and the addressee in these constructions (generic reference, see Siewierska 2008: 9–10). This difference will be discussed for each example below.

The purpose of the present article is thus two-fold: to analyse the nature of the constructions used for non-specified human agents in the corpus of Balochi under study, and to discuss whether Siewierska’s (2008) conclusions about referential properties for the agent in different types of constructions hold for this corpus.


2. Constructions with a non-specified human agent found in the corpus

In this corpus, two main types of constructions where the human agent is not specified are found, one type with an active verb and one type with a passive verb. There are three different constructions with an active verb, 3PL, 2SG, and 3SG constructions.


2.1 Constructions with an active verb, narrative texts

Constructions with an active verb are particularly common in procedural texts, but they are also found in reality-based stories and, although much less commonly, in tales. Since it seems that the constructions operate somewhat differently in narratives and procedural texts, the analysis of the procedural texts is separate from that of the narrative texts (see section 2.3). In narrative texts the verb is found either in 3PL, 2SG, or 3SG. In the ergative construction (ex. 3) the agent is expressed by a pronominal clitic. Only a limited number of examples have been included, but these examples are chosen to be representative of all the occurrences of the particular construction.


2.1.1 Verb in the third person plural

Ex. 1

Ā_____ drōgburr_____ uškit_____ ki _____bi_____ plān _____šār-ā

DEM   liar _____hear.PT.3SG_____ CLM ____in____so.and.so_____town-OBL

drōgburr=ē_______ ast_____ wa____ āī____ tārīp-ān-ā_____ bāz=a



That liar heard that in such and such a town there was a liar who was widely praised (lit. and they praise him a lot). (BT)

The speaker here is the person in the story telling the addressee (the first liar) about a second liar. The presence of a speaker is, in fact, very weak, since the verb uškit ‘he heard’ is used to refer to what was said. The addressee is definitely excluded from the referential properties of the verb kanant ‘they do’. He was not even aware of the second liar until he heard about him from the speaker. It is not totally clear whether the speaker also would praise this second liar or not, i.e. if the speaker is part of the referential framework of the verb kanant ‘they do’ or not.

Possible agents are to be found in the context of the story, which means that this is a vague rather than an impersonal or, to use Siewierska’s terminology, generic construction.

Ex. 2

Šallāx_____ ēšī____ čarm-īēn_____ dāšt

Whip___ DEM.OBL ____leather-ADJZ-ATTR _____have.PT.3SG

ġadīm-ā____ zābul____ ta____ bā____ kurt-ant____ čarm-īēn


A whip, he had a leather whip; in the past in Zabol, you know, they were sold, (whips) made of leather. (BS)

This sentence is found in a reality-based story. Possible agents are tradesmen in Zabol in former times, which makes the agent of kurtant ‘they did’ vague rather than impersonal. In this example, the speaker and the addressee must be seen as excluded from the referential framework of the verb, since the speaker is a storyteller from the region, and the addressee is a linguistic researcher.

Ex. 3

marō_____ zahr=eš_____ rētk-a=Ø _____mā ______xorāk=at

today____ poison=PC.3PL____ pour.PT-PP=COP.PR.3SG____ in_____ food=PC.2SG

Today there is poison poured into your food. (BK)

The context of this sentence from a folktale is that the stepmother wants to kill her stepson, but his horse has supernatural powers and is able to warn him. The speaker is the horse and the addressee is the boy. Possible agents in this ergative construction with an agent clitic instead of the verb in 3PL are the people present in the story who want to kill the addressee. Thus, both the speaker and the addressee are excluded from the referential framework, since neither of them would have put poison in the food. These restrictions of the agent makes the sentence vague rather than impersonal.

Ex. 4

Xān___ mnī___ pād-ān____ kawš-ān-ī____ tā____iškar

Khan___ PRON.1SG.GEN____ foot-PL ___shoe-PL-GEN__ in__ live.ember

rēt-ag=ant_____ ša ______dušmanāīā

pour.PT-PP=COP.PR.3PL_____ from____ enmity-OBL

Khan, my feet! Somebody has poured live embers in my shoes out of enmity. (BS)


This example comes from a reality-based story, and the incident with live embers being poured into the speaker’s shoes took place at a wedding. Possible agents are people present at the wedding party, thus a vague subject from which speaker is excluded. It is also clear in the context that he does not suspect that the addressee (the Khan) would be the agent.


Ex. 5

nimāzliq ___bi___ awā____ bāl kurt=u____ ēšānā_____ āwurt

prayer.rug___ in air___ wing __do.PT.3SG=and __DEM.PL.OBJ___ bring.PT.3SG
am=ōdā____ ki___ wazīr-ay___ jinikk-ā___ šōd-ant

EMPH=there ___CLM___ wizier-GEN___ girl-OBJ.VCL ___wash.PR-3PL

The prayer rug took off into the air and brought them to the place were the wizier’s daughter was being washed (lit. where they wash the wizier’s daughter). (BT)

This sentence is from a folktale. It is the narrator’s voice, which means that the speaker is excluded from the possible agents, as are the addressees, i.e. the audience, who are also outside the framework of the story. Possible agents of the verb šodant ‘they wash’ are people in the story who could be involved in washing the wizier’s daughter, which means that the agent is vague.

Ex. 6

xolāsa _____ar=r-ant ____ahmad-ī____ rannā

in.short_____ VCL=go.PR-3PL_____ NP-GEN ____after

To make a long story short, Ahmad is asked to come (lit. they go to get Ahmad). (BK) The sentence is from a narrative section in a folktale. This excludes the narrator and the audience, i.e. the speaker and the addressee, from the referential framework. Possible agents are the people at the court of the king who needs Ahmad’s services and therefore sends for him. The construction is thus vague rather than impersonal.


It is clear from ex. 1–6 that the 3PL construction is used with a vague rather than an impersonal (generic) human agent. The speaker and the addressee seem to be excluded from the referential properties of this construction in narrative discourse, which means that Siewierska’s conclusion for the 3PL holds in this type of text (see also below).


2.1.2 Verb in the second person singular

Ex. 7

bēšakkā ____ki ___pa___ xudā___ ta ____yakk=ē____ b-day-ay

undoubtedly _____CLM__ for God__ PRON.2SG__ one=IND___ SUB-give.PR-2SG

xudā____ da=a ___dant

God ten=VCL____ give.PR.3SG

Undoubtedly, if you give one (unit of something) for the sake of God, He will give you tenfold (back). (BS)

This is a generic statement meaning ‘Whoever gives something to God will get tenfold back’. The speaker, in this story the prophet Moses, definitely includes the addressee, a poor man who actually showed generosity and was rewarded, and there is no reason to believe that he would exclude himself either. Possible agents are not restricted to the framework of the story, which means that we are dealing with an impersonal construction.

Ex. 8

na-zān-ay _____čīā_____ āī ______ čamm______ ham=ē_____ kišk-ā

NEG-know.PR-2SG ____why___ DEM.GEN ____eye EMPH=DEM___ side-OBL



Nobody knows why his eyes were fixed in this direction. (BP)

This sentence is about a man who is expecting his son to come back home even long after the son has been killed. The agent of the verb nazānay ‘you don’t know’ could be anyone within the story, i.e. anybody who knew this man. But it could also be anyone hearing or reading this story. It therefore seems that an impersonal interpretation is possible here.

Thus, the speaker and the addressee can be included as possible agents in this example.

Ex. 9

Ē____ nimāzliq-ay____ sarā____ ki___ nind-ay ___ā___ bāl=a kan-t

DEM __prayer.rug-GEN __on ___CLM sit.down.PR-2SG___ DEM ___wing=VCL__ do.PR-3SG

When you sit down on this prayer-rug, it takes off. (BT)

The sentence is uttered by a man who wants to sell a magic rug, and the addressee is a person who wants to buy this rug. The intention is, of course, not that it will take off only if this buyer sits down on it, which means that the verb ninday ‘you sit down’ is not to be interpreted as referring only to the addressee. It is an impersonal construction with general preferentiality, including both the speaker and the addressee in the context where it is uttered.

Ex. 10

doros=en____ bās=en_____ ġarīb-pasand_____ be-bey

correct=COP.PR.3SG____ must=COP.PR.3SG____ stranger-accepting SUB-become.PR.2SG

Surely one must accept strangers. (BK)

This sentence is from a reality-based story and the intention of the person who said it is that everybody, including himself and the addressee, should accept strangers. It is open and generic in its referential properties, thus an impersonal construction.

Ex. 11

na-ma-bī-yā______ čūbān-ī______ kan-ey

NEG-IMP-become.PR-3SG_____ shepherd-NOMZ

It is impossible to be a shepherd. (BK)

This sentence is uttered by a shepherd boy, and the addressee is his father. The boy wants to quit being a shepherd and argues that it is an impossible job. It is therefore clear that the speaker in particular is included here. The statement is, however, made in such a general way that also the addressee (the father) and anyone else who would attempt to be a shepherd can be included. It can therefore be interpreted as an impersonal construction.


Ex. 7–11 show that the 2SG construction in narrative text has wider referential properties than the 3PL construction. In all the examples an impersonal interpretation is possible. It is interesting to note ex. 11, where the speaker refers to himself in particular with this construction. However, this example also allows for an impersonal interpretation.

2.1.3 Verb in the third person singular

Ex. 12

guš-īt ___ki ____yag____ bādišā=yē=at

say.PR-3SG ___CLM ___one king=IND=COP.PT.3SG

The story goes (lit. he/she says) that there was a king. (BS)

Ex. 13

ē____ š-ī____ ančēn____ sawt=ē___ēširā___ allā=i pāk

DEM ___say.PR-3SG___ such.ATTR___ voice=IND ___DEM.OBJ____ Allah=IZ pure



He, the story goes, the Holy God had given him such a (wonderful) voice… (BS)

Ex. 14

š-īt___ yak___ xān=ē=at

say.PR-3SG __one __Khan=IND=COP.PT.3SG

The story goes that there was a Khan. (BS)

The construction with the verb in 3SG occurs only in narrative texts in one part of the corpus, namely BS, and only for the verb ‘to say’ (in full or reduced form), but it is common in BS, both in fiction and reality-based stories.7 It is found in the introduction of a story, but also later on in the narration. It is quite clear from the way this verb is used that it is linked to epistemic modality and expresses that the narrator does not have first-hand information about what follows after the introductory verb of saying. Rather, he8 expresses some uncertainty about the contents of the narration, but not to the extent that the listener feels that he expresses outright doubt about it. Thus, this construction does not include the speaker or the addressee in its referential framework and is therefore not impersonal. It has a vague reference to people who may have been eyewitnesses to the very story about to be told or being told.


2.2 Grammatical passive construction, narrative texts

The grammatical passive construction in Balochi consists of an infinitive or a past participle with an auxiliary verb, either ‘to become’ (ex. 15–17, 19–20) or ‘to come’ (ex. 18).9 The passive is normally not used with an overt agent in Balochi (see e.g. Farrell 1995: 231, Baranzehi 2003: 100, Axenov 2006: 200) but there are two such examples in the whole corpus, which thus do not belong in the discussion of a non-specified agent.10

Ex. 15

dawlatxān-ārā___ kayz__ u___ band-ay___ sazā ___day-ag

NP-OBJ____ prison ___ and ___ prison-GEN ___ punishment ___give.PR-INF



Dawlatkhan should not be punished by imprisonment (lit. imprisonment should not be given to Dawlatkhan). (BP)

This is the verdict in a murder case. The person writing it issues an order to those involved in the process, particularly to the addressee (the receiver of the verdict), who would be the actual person to imprison Dawlatkhan. The agent here is, however, vague rather than impersonal, including the addressee as well as others who are part of the process, though not the writer himself, who stands above the process.

Ex. 16

agan māt=ē____ kuš-ag _____ma-būt-ēn______ du_____ čār

if mother=PC.3SG _____kill.PR-INF___ PROH-become.PT-SUB.3SG __ two__ four

rōčā___ rand___ zahg ___allamā___ wadī ___būt-ag=at
day-OBL after child surely born become.PT-PP=COP.PT.3SG

If the mother had not been killed, the child would definitely have been born a couple of days later. (BP)

A woman has been killed and the prosecutor is trying to find out more about the murder. The sentence above is spoken by the father of the killer (i.e. the agent), quoting the midwives about the birth of the child with whom the woman was pregnant.

Either the speaker did not know that his son (actually the father of the illegitimate child, who tried to conceal his adultery) killed this woman, or he did not want to disclose this information. He would hardly include himself as a possible agent though, and the addressee, the prosecutor, is definitely not one of the potential killers, which makes the agent vague rather than impersonal.

Ex. 17

ham=ē___ rōčā___ bēgāh-ay___ wahd-ā____ āsmān-ay___ dēmā

EMPH=DEM ___day-OBL___ evening-GEN___ time-OBL___ sky-GEN ___on

jāgah___ jāgah=ē ___jambar ___ham ___gind-ag___ ātk-ag=at

Place___ place=IND___ cloud__ also__ see.PR-INF __come.PT-PP=COP.PT.3SG

That same day, in the evening, one could see clouds in a few places in the sky. (BP) This sentence is in a narrative part of a modern short story. Here the agent includes everybody who is part of the framework, including the ‘omnipresent narrator’. The persons to whom the story is told (the addressees), i.e. the readers of this short story, are not a natural part of the framework, however.

Ex. 18

diga __āl___ bi___ man___ mālūm___ na-bū___ ki

other __state__ to__ PRON.1SG__ evident NEG-become.PT.3SG CLM

mnī ___mard___ kušt-a ___bū __yā__ na

PRON.1SG.GEN__ man __kill.PT-PP___ become.PT.3SG__ or__ no

It was actually not clear to me if my husband was killed or not. (BS)

This sentence is from a folktale and the speaker is the wife of the man who may have been killed. The addressees are her parents. The potential agents are to be found within the framework of the story, but exclude the speaker and the addressees.

The construction is therefore not an impersonal construction but a construction with a vague human agent.

Ex. 19

dēb-ay ___sarag ___ki___ sist-a____ būt ___dēb___ murt

demon-GEN ___head____ CLM ___remove.PT-PP___ become.PT.3SG__ demon __die.PT.3SG

When the head of the demon was removed, the demon died. (BT)

This sentence is found in a narrative section of a folktale. The potential agents of this construction are to be found within the framework of the story. The speaker and the addressee are excluded from the referential framework of this construction with a vague human agent.


In this corpus, the grammatical passive is the preferred strategy for a vague human agent in written texts (BP), where no instances of the otherwise common 3PL construction are attested. In the oral texts, the grammatical passive is rather rare, although not totally absent. It thus seems that the grammatical passive construction plays the same role in written literary style as the 3PL construction does in oral literary style to denote vague human agents excluding the speaker and the addressee in most instances (ex. 16, 18–19), but that it also can be used with wider referential properties to include the speaker (ex. 17) or the addressee (ex. 15). It is, however, not used in impersonal contexts in the same way as the 2SG construction is used.

Siewierska’s (2008: 22) conclusion that the most open reference is found in agent less passives, which she finds referentially unrestricted, is thus not readily applicable to this corpus.

2.3 Constructions with an active verb, procedural texts

In procedural texts, which in this corpus are available for BT and BK, the two constructions 3PL and 2SG are used interchangeably. They are therefore not separated into different sections here. Two slightly longer examples (ex. 20–21) from procedural texts are presented to illustrate the way these constructions are used in procedural texts. They are taken from a text about weaving and a text about traditional cures for various diseases.

Ex. 20

masalan __har___ raŋ=ē___ ke ___gēš ___bokān=et estefāda

for.example___ every __colour=IND__ CLM more__ want.PR=PC.2SG __use

kan-ey ____ā____ raŋ=at___ gēš=a __kan-ey___ hālā___ yā ___DEM colour=PC.2SG___ more=VCL do.PR-2SG__ now_ or

zard ___yā___ ġermez___ yā___ ke___ ez… har___ raŋ=ēābī

Yellow___ or ___red ___or___ CLM___ from… every___ colour=IND…

aksaran___ ġālī-bār-ey ____zamīn-ā____ ġermez=a ____kan-an___t hā

Mostly____ carpet-PL-GEN ground-OBJ red=VCL___ do.PR-3PL yes

ġermez=a___ kan-an

red=VCL ___do.PR-3PL

You know, any colour you want to use more, you dye more wool in that colour (lit. make that colour of yours more) now, either yellow or red or…any colour…light blue. You (lit. they) mostly make the ground of the carpet red. Yes, you (lit. they) make it red. (BK)

Here the narrator starts out by using constructions in the 2SG (bokān=et estefāda kaney, gēš=a kan-ey) and then switches to the 3PL construction (ġermez=a kan-ant, ġermez=a kan-an) in the very same passage. It is clear that she includes herself in the referential properties of these constructions, since she is a weaver and therefore sometimes herself performs the tasks she describes. She potentially also includes the addressee, the linguistic researcher, in case she would like to try the craft of weaving.

If the addressee was a weaving apprentice, she11 would definitely be included in the referential framework of the construction. It is thus possible to apply an impersonal interpretation both to the 3PL and the 2SG construction in this example.

Ex. 21

pa a__dardīā ___pas-ay___ pōst-ā ____gwarā=a kan-ant___ pas=ē

for___ bone.ache-OBL__ sheep-GEN __skin-OBJ__ on=VCL do.PR-3SG__ sheep=IND

ki____ kuš-ay___ pōst-ay-ā______ kašš-ay____ pōst-ā

CLM ___kill.PR-2SG ___skin-PC.3SG-OBJ.VCL___ pull.PR-2SG skin-OBJ.VCL

patāy-ant=u_____ garm=a ___sōč-ant ___šap-ā___ nājō_-ay

fold.PR-3PL=and____ warm=VCL___ burn.PR-3PL night-OBL____ sick-GEN

puččān-ā___ kašš-ant=u____ pōst-ā ___bi ___gwaray-ā

clothes-PL-OBJ.VCL___ pull.PR-3PL=and____skin-OBJ___ to___ on.PC.3SG-OBL.VCL



For pain in the bones, you (lit. they) put on a skin from a sheep. When you kill the sheep, you pull off its skin. You (lit. they) fold the skin and heat it up. At night you (lit. they) pull off the clothes of the sick person and you (lit. they) put the skin on him. (BT)

This example allows for a similar interpretation as ex. 20. The speaker, who himself knows about traditional medicine and therefore can be expected to cure people with pain in the bones, is included in the referential framework. He starts out with a 3PL construction (gwarā=a kanant), then uses two 2SG constructions (kušay, kaššay), and then again four 3PL constructions (patāyant, sōčant, kaššant, dayant). The addressee, although in this case a linguistic researcher, i.e. an outsider, could probably also be included in the referential framework of this construction in case he would need to treat somebody with pain in the bones. It is also likely that the very same construction would be used to instruct somebody from within the culture who would like to learn this skill (see also ex. 20).



An interesting observation is that the two constructions with 3PL and 2SG verbs seem to be interchangeable in procedural texts, making the referential framework of the 3PL construction able to also include the speaker and the addressee, thus allowing for an impersonal interpretation. The distinctive features of the 3PL construction versus the 2SG construction encountered in narrative texts are thus not present in procedural texts. This observation contradicts Siewierska’s conclusion that the 3PL construction does not readily include the speaker and the addressee within its referential framework.


3. Conclusions

The present corpus proved to be a rich source for the investigation of constructions with an impersonal or vague human agent in Balochi. There are three main constructions found in the corpus, those with the verb in 3PL, those with the verb in 2SG, and those with a passive verb.12

It seems that in Balochi the 3PL construction allows the speaker to distance himself/ herself from the event somehow, particularly in narrative texts, but that this interpretation cannot be applied to the 3PL construction in procedural texts. Thus, the 3PL construction does follow the referential properties described by Siewierska (2008: 14–17) in narrative texts but allows for wider reference in procedural texts.

The referential properties of the passive construction do not seem to be as unrestricted as Siewierska suggests. In most examples of the passive construction a vague interpretation lies closer at hand than an impersonal interpretation. There is one example (ex. 15) where the addressee is definitely included and one where the speaker can be seen as included (ex. 17), but in most examples both the speaker and the addressee are definitely excluded.

The 2SG construction in narrative texts allows for an impersonal interpretation. It is also used totally interchangeably with the 3PL construction to include the speaker, and probably also the addressee, in procedural texts, which makes an impersonal interpretation possible for both these constructions in this type of texts. Thus, the 2SG

construction in the examples from narrative texts (ex. 7–10) and the 2SG and 3PL constructions in the examples from procedural texts (ex. 20–21) are open to a truly impersonal (generic) interpretation.

These conclusions are similar to the conclusions drawn by Shokri (in the present volume) about the use of 3PL and 2SG impersonal constructions in Mazandarani, another Iranian language closely related to Balochi.

List of abbreviations and symbols

– :                     separates a morpheme

= :                    separates a clitic

Ø:                    zero morpheme

1:                     first person

2:                     second person

3 :                    third person

ADJZ:             adjectivizer

ATTR:            attributive

BK:                 Balochi, Koroshi (Maryam Nourzaei’s corpus)

BP:                  Balochi of Pakistan (Carina Jahani’s corpus)

BS:                  Balochi of Sistan (Behrooz Barjasteh Delforooz’s corpus)

BT:                  Balochi of Turkmenistan (Serge Axenov’s corpus)

CLM:              clause linkage marker

COP:               copula

DEM:             demonstrative

EMPH:           emphatic particle

GEN:              genitive

IMP:               imperfective

INCL:             inclusive

IND:               indefinite

INF:                infinitive

IZ:                   izāfa
LOC:               locative

MIR:              mirative particle

NEG:               negative

NOMZ:           nominalizer

NP:                  proper noun

OBJ:               object

OBL:               oblique

PC:                  pronominal clitic

PL:                  plural

PP:                  past participle

PR:                  present

PROH:            prohibitive

PRON:            personal pronoun

PT:                  past

SG:                  singular

SUB:               subjunctive

VCL:               verb clitic

1 In 2009, a corpus-based linguistic project comprising several languages belonging to different language families was initiated at the Department of Linguistics and Philology, Uppsala University, the aim of which is to study, among other grammatical features, impersonal constructions.
2 Sincere thanks to Agnes Korn, Frankfurt am Main, for comments on an earlier version of this article and to Guiti Shokri for interesting discussions during the writing process.
3 Siewierska (2008) does not treat 2SG impersonal constructions (see below).
4 Siewierska (2008: 9) uses the term ‘impersonal’ the way Kitagawa and Lehrer employ the term ‘vague’ reference, and ‘generic’ for the concept Kitagawa and Lehrer call ‘impersonal’. The present article follows Kitagawa and Lehrer.
5 These four corpuses will henceforth be referred to as ‘the corpus’.
6 Nasalization is not taken into account here. Pronoun forms are analysed as one unit due to the great variation in these forms. The DEM.OBJ can, e.g., occur as ārā, āīrā, āīārā, āīā. The COP.PR.3SG is explicitly given as Ø in the present perfect, but not the Ø personal ending in the past tense 3SG.
7 The 3PL is occasionally found in BS the same context, but not nearly as frequently as the 3SG.
8 All these stories were told by male storytellers.
9 For the passive construction in Balochi, see also Jahani and Korn 2009: 662–663.
10 E.g. pādišā ki kušt-a būt ša ragjan-ay dastā king CLM kill.PT-PP COP.PT.3SG from bloodletter-GEN hand mušmā ā ragjan-ā bi dār-ā jan-an PRON.1PL.INCL DEM bloodletter-OBJ to wood-OBL.VCL hit.PR-1PL When the king has been killed by the bloodletter (lit. at the hand of the bloodletter), we will hang that bloodletter. (BT)
11 In this culture it is only the women who weave.
12 The construction with 3SG verb is limited to one specific verb and is also found only in BS. Another interesting impersonal construction has, in fact, been observed by the authors in spoken Balochi, but it is not attested in the corpus. This construction uses mardum ‘people’ with the verb in the singular for an impersonal human agent. The same subject with a plural verb would denote a vague human agent (mardum= a kant ‘one does, you do (impersonal)’ versus mardum=a kanant ‘people do (vague)’).

Axenov, Serge (2006). The Balochi Language of Turkmenistan. A corpus-based grammatical description.

(Acta Universitatis Upsaliensis, Studia Iranica Upsaliensia 10). Uppsala: Uppsala University.

Baranzehi, Adam Nader (2003). “The Sarawani Dialect of Balochi and Persian Influence on It”. In Jahani,

Carina, and Korn, Agnes (eds), The Baloch and Their Neighbours: Ethnic and Linguistic Contact in

Balochistan in Historical and Modern Times. Wiesbaden: Reichert, pp. 75–111.

Blevins, James P. (2003). “Passives and Impersonals”. Journal of Linguistics, 39, pp. 473–520.

Blevins, James P. (2006). “Passives and Impersonals”. In Brown, Keith (ed.-in-chief), Encyclopedia of

Language and Linguistics. Oxford-Amsterdam: Elsevier, pp. 236–239.

Farrell, Tim (1995). “Fading Ergativity? A Study of Ergativity in Balochi”. In Bennett, David C. et al.

(eds), Subject, Voice and Ergativity. London: School of Oriental and African Studies, pp. 218–262.

Jahani, Carina, and Korn, Agnes (2009). “Balochi”. In Windfuhr, Gernot (ed.), The Iranian Languages.

London and New York: Routledge, pp. 634–692.

Kitagawa, Chisato, and Lehrer, Adrienne (1990). “Impersonal Uses of Personal Pronouns”. Journal of

Pragmatics, 14, pp. 739–759.

Onishi, Masayuki (2001). “Introduction: Non-canonically marked subjects and objects: Parameters and

properties”. In Aikhenvald, Alexandra, Dixon, Robert M. W., and Onishi, Masayuki (eds), Noncanonical

Marking of Subjects and Objects. Amsterdam: Benjamins, pp. 1–51.

Siewierska, Anna (2008). “Ways of impersonalizing. Pronominal vs. verbal strategies”. In Gómez

González, María de los Ángeles, Mackenzie, J. Lachlan, and González Álvarez, Elsa M. (eds), Current

Trends in Contrastive Linguistics. Amsterdam: Benjamins, pp. 3–26.

Published by Orientalia Suecana LIX (2010)

Comments Off on Impersonal Constructions in Balochi

Posted by on January 3, 2014 in Articles on Balochi language


50 years of the Balochi Academy, Quetta

By: Aseer Abdul Qadir Shahwani

Translated by: Hamid Ali Baloch

The first valuable attempt for the establishment of the Balochi academy and the promotion and popularization of the Balochi language was made in 1957, at Kalat in the name of “Balochi Diwan” and Haji Abdul Qayyum was nominated as the first convener of the Balochi Academy.
On august, 1958, a huge concourse of tribal chiefs, literary men, political leaders and poets was held at Mustung, and the objective of this gathering was to collect financial, ethical, literary and political support for the Balochi academy.
In 1961, it was established as a regular institution. In the early years of its establishment the chairmen and their sincere colleagues left no stone unturned to institutionalize the academy but became vain.
In 1995, Mr. Jan Muhammad Dashti, an expert and well-acquainted man was elected as the Chairman of the Balochi Academy, and he led a u-turn in the destiny of the Balochi Academy.
Fifty years before, when the scion of knowledge and Literature for the promotion and development of the Baloch, Balochi and Balochistan had planted in the shape of Balochi academy, this absolutely formed it a fruitful plant.
During the periods of its development, the Balochi academy made the Balochi language, literature and culture conspicuous by its writings, debates and Conferences.
In 1958, a Constitutional draft had been prepared and in the light of the draft it was given a legalized shape later. A new article for the honorary members (the Baloch intellectuals and literary men who are living in Pakistan or abroad) had been added in the constitution of the Balochi Academy. The date for the annual meeting has been fixed for the members on the first Saturday of the year as usual, and the committees are organized for the election in the same Saturday.
For the welfare and progress of the Balochi Academy, eight committees such as, Finance committee, constitution committee, Orthographic committee; Dictionary committee, encyclopedia committee, Seminar committee, publishing committee and marketing committee have been organized to tackle down the maters of the Balochi academy.
The provincial government used to specify a grant of approximately Rs.50, 000 annually and after all this amount was enhanced to Rs.95, 000. But it is sorry state that the amount was not enhanced till thirty years, and in 1997 when the then chief minister of Balochistan Sardar Akhtar Mengal (an eminent learned man) came into power, he increased the grants of the Balochi academy to Rs.100, 0000. In 2010, the amount rose to Rs. 200,0000and in 2011, the provincial government increased the grant to Rs.500, 0000.
The Balochi academy has published at least three hundred books on different topics such as literature, linguistics, religion, culture, autobiographies, personalities and politics, science, ethics, economics, travelogue and folk literature.
The name of Conference Hall and the rooms of the Balochi Academy have been kept on the historical names of different places such as, Sarlatt, Rabāt, Miyānī, Gokprosh, Bampoor, Talār, Trātānī and Kōhang.
The Balochi Academy has been published approximately 280 books from its birth till now, which are mainly consisted of the folk literature, culture, poetry, grammar and biographies. Many of the published books of the Balochi academy have not been only concluded as texts books in Schools and colleges but as texts books for Federal and Provincial competitive examinations.
The official website of the Balochi Academy is:

In 2001 the Balochi academy started compiling a Balochi-Balochi Dictionary. For the compilation of this dictionary a team of literary men and experts was made in the supervision of Jan Mahamd Dashti. The editorial board was consisted of the following editors:
Jan Muhammad Dashti (Editor in Chief) and editors are: Ulfat Naseem(Panjgoor), Gulzar Khan Murree(Kohlu), Ghaus Bahar(Ormara), Ghani Pahwal(Karachi), Dr. Abdul Saboor Baloch(Turbat), Prof. Muhammad Yousuf Baloch((Noshki), Dr. Aini Baloch(Kharan), Prof. Wahid Buzdar(Koh e Sulaiman), Sajid Buzdar(Dera Ghazi Khan), Molvi Abdul Haq Baloch(Turbat), Yousuf Gichki( Turbat) and Ashraf Sarbazi(Iran).
The dictionary is in its last stage of completion.
On 27thMay 1998, the Balochi academy conducted a Literary Seminar on the occasion of its Silver Jubilee. The Balochi academy is providing its publications approximately 70 libraries from different parts of the province.
Beside, the Balochi academy used to carry out functions and ceremonies time by time in the past and invited eminent personalities in different occasions. Sardar Akhatar Mengal was among the prominent figure, who was invited as chief guest in the inauguration function of the Balochi Academy Complex on 5th June, 1999.
The Balochi Academy has conducted Seminars on the topics such as, Balochistan: Literature and Culture, Natiq Makkorani, Balochi Orthographic System, Pashto Literary Seminar (21st April, 2004), the role of Literature in the changing world and rehabilitation of peasants. Apart from this, the Balochi Academy conducted book exhibitions on knowledge festival at Islamabad and Askari Park Quetta. The Balochi academy has also the honor to carryout book display ceremonies of other institutions in its Conference Hall. Condolence congregations have been managed on the death of poets and literary men in the Balochi academy and lectures have been delivered on the literary contributions of the deceased literary men.
After 50 years of its establishment, the Balochi Academy celebrated its Golden Jubilee on 31st July- 1st august 2011. In this regard, the 2nd International Conference on the Balochi Language, Literature, Culture and History is going to be held and the national and international Scholars will deliver their research papers in the said Conference.
It’s very interesting to note that after the construction of the Balochi Academy premises, the intellectuals came to visit the Academy from different parts of Sindh, Punjab, Khyber Pakhtoonkhwa, Iran, the USA and the United Arab Emirates and noted their views in the visitors’ book of the Balochi Academy.
The future planning and the vision of the Balochi academy are clear. Our mission is to leave no stone unturned for the promotion and popularization of the Balochi language, literature, culture and history and to bring the Balochi language on the line of advanced languages.
In this regard, we are en-route to carry out some productive plans such as:
► To make an acceptable and standardized Orthography for the Balochi language.
►to hold national and international conferences concerning to the Balochi language, Literature, Culture and History.
►to establish a spacious library on the basis of modern science and technology and to preserve manuscript and hand written books, publications, and inaccessible research books, films, audios, videos and cassettes.
►to establish a more effective and functional research cell, which fulfill the needs of modern research.
►to extend the needs of Balochi dictionary and translate the Balochi-Balochi dictionary into Balochi-Urdu, Balochi-Persian, Balochi-Arabic, Balochi- Brahvi, Balochi- Sindhi and Balochi- Pashto.
►to write down the three thousand years ancient history of the Baloch nation on the basis of research and compile a book on it.
The eminent personalities to whom the chairmanship has been honored are; Muhammad Sardar Khan Gishkori, Haji Abdul Qayyum, Malik Muhammad Ramazan, Bashir Ahmad Baloch, Abdullah jan jamaldini, Jan Muhammad Dashti, Abdul wahid Bandeeg and Siddiq Baloch respectively.

♣♣♣♣ ♣♣♣♣ ♣♣♣♣


Restrictive Relative Clauses in Balochi and the Marking of the Antecedent, Linguistic Influence from Persian?

Prof Dr Carina Jahani

Prof Dr. Carina Jahani


An interesting morphosyntactic feature that has been observed in e.g. New Persian1 (JAHANI 2000) and Balochi (BARANZEHI 2003:102, AXENOV 2006:250-251) is that thesame suffix2 that marks indefinite restrictive selection out of a generic unit or a
plurality (the so-called yPa-ye va .hdat) is also attached to head nouns3 of restrictive relative clauses.

A restrictive relative clause is one that “determines and restricts the extension of the head”, as opposed to a non-restrictive (or descriptive) relative clause, which “merely gives some additional explanation of it” (GREN-EKLUND 1978:53), e.g. “the families that have a car could offer other families a ride to the outing” as opposed to “my father, who has a car, offered another family a ride to the outing”.4 A restrictive relative clause thus restricts, or selects, one or several objects from among a generic unit or a plurality at hand by first specifying a larger set, the domain of relativisation (“families” in the example above), and then restricting that domain “to some subset of which a certain proposition is true” (i.e. “having a car”) (FABB 1994:3520).

Head nouns of restrictive relative clauses can either be generic or definite. In the first case the restriction is open (indefinite, non-specific). The subset for which the relative clause is true is thus an open class, which means that it can also be empty. In the sentence “a book that makes you understand yourself better is definitely worth reading” the domain of relativisation is “a book”, i.e. a generic head noun, and the restriction is open, which means that the subset of which the proposition “makes you understand yourself better” holds true is not delimitated. It could hold true of many books, or even of no book at all, in which case the subset would be an empty class.

In the second case the restriction is closed (specific) which means that the subset selected from the domain of relativisation is a closed class and delimitated by means of the restrictive relative clause, e.g “the book that you bought yesterday is really worth reading”. In this case the domain of relativisation, i.e. the head noun “the book”, is definite and the subset of which the proposition “you bought yesterday” holds true is a closed class, here containing only one object. If the noun is in the plural, “the books that you bought yesterday are really worth reading”, the subset will contain more than one object (see also LEHMANN 1984:261). The distinction between open and closed restriction (discussed in brief by GREN-EKLUND 1978:52), is particularly relevant to the marking of the antecedent in Iranian languages.

A previous study of relative clauses in Classical and Modern Persian (JAHANI 2000) shows that the grammaticalisation of the yPa-ye va .hdat, which in Modern Persian is pronounced -P., as a marker of antecedents of restrictive relative clauses had already started when early Persian prose was written (from the end of the 9th century A.D.), but it was not completed until the Modern Persian language emerged in the 20th century.

This process has thus taken about 1000 years or more. Persian is the New Iranian language that has the longest literary tradition and it therefore leaves a particularly goodwritten material for diachronic studies of grammaticalisation processes like the onedescribed above.

Quite on the contrary, Balochi has a very short written literary tradition but a long oral literary tradition. We can assume that the language of epic poetry and tales transmitted orally from generation to generation is somewhat normalised and conventionalised for mnemonic purposes, just like the language of written literature (see UTAS 2005:65 and 2006:209-210). Even so, a certain amount of structural change can be expected over time in oral literature. Oral texts have in the case of Balochi been recorded in written form only from around 1900 onwards (e.g. in DAMES 1891) but it is reasonable to assume that they reflect different layers of language structure including that of the time when they were recorded. Thus, unless these different layers can be discerned in the texts, they provide material that allows us to make diachronic studies reaching back over a period of only slightly more than 100 years.5 When it comes to non-literary, spoken Balochi, and the spoken variant of most other Iranian languages (possibly excluding Persian, where observations about the spoken language were made rather early by European travellers, see PERRY 1996), it is even harder to make diachronic studies due to the fact that very few descriptions of spoken language and hardly any recordings are available to us from a date earlier than the mid-20th century.

In Balochi we find the same suffix as in Persian marking indefinite restrictive selection out of a generic unit or a plurality. However, due to the retention of the mid-open socalled ma.ihPul vowels /Pe/ and /Po/ in the Balochi phonemic system, the pronunciation of this suffix is -Pe in Balochi, which was also the pronunciation in Classical Persian.6

The purpose of the present article is to investigate whether an extension of the use of the suffix denoting indefinite selection to mark antecedents of restrictive relative clauses similar to the one observed in Persian has also taken place in Balochi. If there is such a marking, in what variants of Balochi does it occur? In these cases, should it be seen as an internal development in Balochi parallel to that of Persian or can it be attributed to linguistic influence from Persian?

To answer these questions a number of texts and recordings of written and spoken Balochi from different parts of Balochistan will be investigated. Different dialects of the three main groups . Eastern, Southern and Western Balochi . will be represented. Most written texts belong to the oral literature genre, but some are fiction and factual prose. Since writing fiction and factual prose in Balochi dates mainly from the 1950s onwards, it is assumed that texts belonging to these genres represent more or less the same grammatical structures as the spoken language. The oral literature may represent considerably older grammatical structures (see above). However, the Balochi language data at hand do not allow any diachronic conclusions, nor do they permit a study of diverging sociolects, genderlects or generatiolects (see JAHANI 2003:130). Variants of Balochi will therefore be analysed with a focus on geolects (geographical variation). Dialects of Balochi that stand in a continuous close contact with Persian will be compared with dialects spoken in areas where Persian is not a dominant language (see below).

It may also be possible to find traces of historical contacts between Balochi and Persian by comparing the marking of the antecedent in dialects that today are spoken far from Persian-speaking areas, and thus not now under direct Persian influence, with the way the antecedent was marked in older stages of Persian.

As for glossing, the singular case ending -Pa is glossed OBJ or OBL depending on its function. The ending -aPn or -aP is glossed PL.OBL for all dialects except those spoken in Iran, where it is glossed PL, since it has spread also to the direct case in these dialects.7 There is also a plural ending -PanPa (PL.OBJ) that is used for direct and indirect objects in most dialects of Western Balochi. In Iranian Balochi a merger of the direct and oblique cases is underway (JAHANI 2003:121-125). Therefore forms glossed as OBL may occur as the subject (in the non-ergative domain) in Iranian Balochi. Verb stems are marked as either present8 (PRS) or past (PST). Only grammatical morphemic analysis is carried out and word formation (of compound words, causative verbs etc.) is ignored.

2.Marking of the head noun in Persian

In Classical Persian we find head nouns of restrictive relative clauses both with and 7 For case systems in different Balochi dialects, see JAHANI (2003) and Korn in this volume. 8 Or rather non-past, since present also includes future. See e.g. AXENOV 2006:175.
without the suffix -P.. As mentioned, antecedents with this suffix are already found in the earliest Classical Persian texts, but antecedents without -P. are equally common. However, antecedents without -P. are normally qualified by a demonstrative or indefinite pronoun (used as attributive adjectives), e.g. Pan “that”, P.n “this” or har “every”. My investigation of Persian (JAHANI 2000) clearly shows that the suffix -P. in early Classical Persian was more common on head nouns of an open restriction than on head nouns of a closed restriction. Since there are certain constraints as to what pronouns in Persian can take the suffix -P., only the statistics dealing with nouns as heads are presented here.

In 62 percent of all cases, a head noun of an open restriction is marked with the suffix -P., whereas in the case of a closed restriction only 32 percent of the head nouns take this suffix. This could indicate that the -P. “was initially used on indefinite antecedents in accordance with its use on any noun to mark indefiniteness” (JAHANI 2000:52), and that from there it spread also to marking head nouns of a closed restriction. If only nouns are counted we get the percentages shown in Table 1 for antecedents with or without -P. in Classical and Modern Persian.9

Table 1: Marking of head nouns in restrictive relative clauses in Persian

Pattern10 \ period (A.D.) 10th-12th cent. 13th-15th cent. 16th-19th cent. 20th cent.
A. Noun phrase + ke 54% 43% 30% 12%
B. Noun phrase + -P. + ke 46% 57% 70% 88%

In Modern Persian it is thus the marking with the suffix -P. that is totally predominant. It is very rare to find a demonstrative pronoun co-occurring with this suffix in the literary style, but in dialogue parts of texts, as well as in spoken Modern Persian, a double marking on the antecedent, i.e. both a demonstrative pronoun and the suffix -P., has been noted (JAHANI 2000:49-51). The most likely explanation is that this is a retention of the pattern with a demonstrative pronoun from early Classical Persian with the addition of the -P.. This makes the spoken language more explicit than the written language in this grammatical structure, signalling the head noun both with the demonstrative pronoun and the -P..

3. Linguistic contact between different dialects of Balochi and Persian

Balochi has probably been in contact with and influenced by spoken Persian (Old, Middle and New Persian) from early historical times, possibly as long as 2000 years or more. This can be seen by the fact that Balochi, a north-western Iranian language, shares phonological innovations with both the old stage (from about the 6th century B.C. onwards) and the middle stage (from about the 3rd century B.C. onwards) of the south-western Iranian language Persian (KORN 2003:58-59). There are also several layers of loanwords from Old, Middle and New Persian found in all dialects of Balochi (KORN 2005:49, 330).

At the court in Kalat, where the administrative language was Persian,11 Balochi was also in contact with written Classical Persian, and in traditional religious education Persian literature was taught in addition to Arabic. We therefore assume that in historical times Balochi, at least to some degree, was in touch with and under influence of literary Persian. Even so, Balochi is likely to have been primarily in contact with the spoken form of the Persian language. However, Southern and Eastern Balochi, are spoken far from Persian speaking areas, so contact with Persian seems to have been less intense in these dialects than in Western Balochi. There is both phonological and morphosyntactic evidence of deeper influence from Persian in different dialects of Western Balochi than in Eastern and Southern Balochi. For example, we find more frequent use of prepositions and a more profound breakdown of the ergative construction in Western than in Southern and Eastern Balochi.12

Nowadays there is heavy influence both from spoken and written Persian on those dialects of Balochi (both Western and Southern) that are spoken in Iran (see also MAHMOODZAHI 2003:147-152 and JAHANI 2005:158-160). This influence is, of course, but a natural result of modernisation, which involves more and more intense contacts with the state language through such dynamics as education, mass media, government employment and increased travelling.

Though to a lesser degree, the same is true of the dialect (Western Balochi) spoken in Afghanistan where modernisation has been slower than in Balochi speaking areas of Restrictive Iran. Here it is the Afghani (Dari) variant of Persian that is predominant, but in recent years also the Iranian variant (Farsi) reaches Balochi speaking areas in south western Afghanistan through radio and TV broadcasts from Iran.

Groups of Baloch started to migrate mainly from Afghanistan northwards into Turkmenistan in search of better pastures and more political freedom from about 1900 until 1934, when the Soviet border was closed (MOSHKALO 2000:97-99). Those Baloch who settled in Turkmenistan speak Western Balochi, but there contact with Persian hasbeen replaced by contact with and influence from Russian and Turkmen.

3.Marking of the head noun in different Balochi dialects13

The head noun of a restrictive relative clause in the Balochi texts analysed in this study either takes the suffix -Pe (which in dialects heavily influenced by Persian is sometimes even pronounced -P.) or is preceded by a demonstrative or an indefinite modifier (a pronoun or an adjective), or takes a double marking, i.e. both a preceding demonstrative / indefinite modifier and the suffix. The emphatic particle (h)am is frequently added to a demonstrative pronoun preceding the head noun. The five patterns found are demonstrated in Table 2.

Table 2: Patterns for marking the head noun in Balochi
Without the suffix -¯e
i. Demonstrative pronoun (DEM) + Head Noun + ∅
ii. Indefinite modifier (IND.MOD) + Head Noun + ∅
B. With the suffix -¯e
i. Demonstrative pronoun (DEM) + Head Noun + -¯e
ii. ∅+ Head Noun + -¯e
iii. Indefinite modifier (IND.MOD) + Head Noun + -¯e

In all these patterns both open and closed restriction is possible, although the presence of an indefinite pronoun almost invariably denotes an open restriction and a demonstrative pronoun very often implies a closed restriction.

The suffix -¯e can co-occur with the case endings -r¯a and -¯a, in which case the -¯e precedes these endings. In Iranian Balochi a head noun with the plural suffix (cf. p. 4) can take the suffix -¯e, which is added after -¯an.

Conjunctional use of head nouns is not described in this study, e.g. wahd-¯e ki “when, lit.: at the time that”, har zam¯an-¯e ki “whenever, lit.: every time that”, pa x¯atir-¯e ki “because”, am¯a rang ki “just like, lit.: in the same way that”. Likewise, examples whose head noun has an adverbial function in the main clause are omitted from the study.14

4.1 Balochi from Afghanistan15

Balochi of Afghanistan is here represented by an autobiographical account in oral form written down and published by BUDDRUSS (1988) as well as by 34 transcribed pages of tales and a conversational text (an interview) collected by Behrooz Barjasteh Delforooz between 2000 and 2006.16

Without the suffix -¯e

i. DEM + head noun (altogether 15 examples):
1 ¯e ˇc¯a-n¯a wa ¯e n¯ašt¯ay¯ı ki ¯a ¯awurt-at
DEM tea-PL.OBJ and DEM breakfast SUB DEM bring.PST-COP.PST.3SG
man wa ¯a w¯art-an
PERS.1SG and DEM eat.PST-1PL
“She and I drank the tea and ate the breakfast that she had brought.” (BUDDRUSS 1988:32)
2 ¯e ˇiin¯ınz¯ag am-¯a ˇc¯adir-¯a ki man wat-r¯a bir
d¯at-ag-un am-¯eš-¯ı-r¯a kaš-a kan-t

“This woman is pulling this cover with which I have covered myself.” (BUDDRUSS 1988:30, 32)

ii. IND.MOD + head noun (altogether 2 examples):

3 ar ˇcinka s¯al ki ta g¯o man gw¯az¯ent-ay
every how many year SUB PERS.2SG with PERS.1SG pass.PST-2SG
ša s¯al-ay d¯em¯a t¯ı tanx¯a da-y-¯ın
from year-GEN in front of PERS.2SG wages give.PRS-GL-1SG
“I will give your wages on a yearly basis for as many years as you (will) have passed with me
(when you leave).” (BUDDRUSS 1988:26)

from year-GEN in front of PERS.2SG wages give.PRS-GL-1SG
“I will give your wages on a yearly basis for as many years as you (will) have passed with me (when you leave).” (BUDDRUSS 1988:26)

With the suffix -¯e

DEM + head noun + -¯e (altogether 16 examples):

4 ¯a malang-¯e ki ta gušt-ay amm¯a kušt-an-¯e
We killed that dervish whom you talked about.” (Barjasteh Delforooz) ENCL.3SG

iii. Head noun + -¯e (altogether 9 examples)

5 amm¯a bi zub¯an-¯a bal¯oˇc¯ı-y-ay-¯a (y)¯asi¯aw¯an ¯ıš-¯e-r¯a
PERS.1PL.EXCL to language-OBL Balochi-GL-GEN-OBL miller DEM-OBL-OBJ
guš-an ki b¯ıst-u-ˇc¯ar sahat bi y¯asi¯ab-ay sar¯a ništ-a17 wa kas-¯e
say.PRS-1PL SUB twenty-four hour to mill-GEN on sit.PST-PP and person-IND
ki ša watan-ay t¯a galla k¯ar-¯ıt pa ¯art-¯a
SUB from homeland-GEN in wheat bring.PRS-3SG for flour-OBL
¯a-w-¯an¯ı muz-¯a z¯ur-¯ıt wa ¯a-w-¯an¯a ¯art-a kan-t
DEM-GL-PL.GEN wages-OBJ take.PRS-3SG and DEM-GL-PL.OBJ flour-V.EL do.PRS-
“In the Balochi language we call that one a miller who sits at the mill 24 hours per day and who takes payment from anyone who brings wheat to grind from his native land and grinds it into flour.” (BUDDRUSS 1988:24)

6 ˙ gab¯ul kurt-ag-ant ki ˇc¯ız-¯e-r¯a ki d¯ın guš-¯ıt
accepting do.PST-PP-COP.PRS.3PL SUB thing-IND-OBJ SUB religion say.PRS-3SG
w¯a ˙ gi¯ıyat d¯ar-¯ıt
truth have.PRS-3SG

“They have accepted that what religion says holds true.” (Barjasteh Delforooz)

iv. IND.MOD + head noun + -¯e (altogether 6 examples):

7 t¯ı n¯an wa t¯ı puˇc wa digar masrap-¯e ki
PERS.2SG.GEN bread and PERS.2SG.GEN clothing and other consumption-IND SUB
d¯ar-ay muˇc-¯a man-a da-(y)-¯ın
have.PRS-2SG all-OBJ PERS.1SG-V.EL give.PRS-(GL)-1SG
“I will give [you] everything, your food and your clothes and the other needs of consumption that
you have.” (BUDDRUSS 1988:26)


There are 31 cases of a head noun marked with -¯e and 17 cases without the suffix. It is interesting that the predominant constructions are the ones with a demonstrative pronoun before the head noun (with or without the suffix -¯e), a construction also common in spoken Persian. In this dialect, the suffix -¯e can combine with the case marking suffix -r¯a (ex. 6, where the head noun is the direct object of the relative clause), but it is more common for the case marking of the head noun to be omitted when it is marked with -¯e (ex. 4, where the head noun is the direct object of the main clause).

4.2 Balochi from Turkmenistan18

Balochi of Turkmenistan is here represented by 35 pages from Zarubinfs tales (ZARUBIN 1932:1-35) as well as by 35 pages of oral texts (folktales and ethnographic texts) recorded and transcribed by Serge Axenov between 1989 and 2000.19

A. Without the suffix -¯e
i. DEM + head noun (altogether 6 examples):

8 ta p-¯e k¯ar-¯an ki man-a kan-¯ın baxt
PERS.2SG for-DEM work-PL.OBL SUB PERS.1SG-V.EL do.PRS-1SG luck
“You don’t have any success in the things that I am doing.” (Axenov)

9 am-¯e mardum ki asan-¯ı gis-¯a šut-at pa p¯adiš¯a
EMPH-DEM person SUB NP-GEN house-OBJ20 go.PST-COP.PST.3SG for king
b¯az u b¯az t¯ar¯ıp-¯ı kurt asan-¯ı ˇian¯en-¯a
much and much laudation-ENCL.3SG do.PST.3SG NP-GEN wife-OBJ
“The person who had gone to Asan’s house praised Asan’s wife highly before the king.”
(ZARUBIN 1932:15)

ii. IND.MOD + head noun (altogether 5 examples):

10 ša mušm¯a ar kass ki t¯ar¯ı p¯adiš¯a-¯a d¯em¯a d¯ıst
from PERS.1PL.INCL every person SUB morning king-OBJ first see.PST.3SG
b-guš-¯ıt ay mihrab¯an-¯en p¯adiš¯a …
SBJ-say.PRS-3SG oh merciful-ATTR king
“The one of us who meets the king first in the morning should say: Oh merciful king…” (Axenov)

B. With the suffix -¯e
i. DEM + head noun + -¯e (altogether 16 examples):

11 am-¯e ˇc¯ız-¯e ki man tr¯a guš-¯ın ta
am-¯a mn¯ı abar-¯a b-z¯ur-ay
“You should accept my word, that thing which I tell you.” (ZARUBIN 1932:18)

ii. Head noun + -¯e (altogether 4 examples):
12 ˇiin¯enz¯ag kit¯ab-¯e ki bi ¯a-¯ı t¯a zin¯ak¯ar¯ı u gandak¯ar¯ı-ay
woman book-IND SUB to DEM-OBL in adultery and evildoing(=adultery)-GEN
gun¯a u ˇiiz¯a nimist-a k¯ar-¯ıt u dant
sin and retribution write.PST-PP bring.PRS-3SG and give.PRS.3SG
bi p¯adiš¯a-ay dast-¯a
to king-GEN hand-OBL
The woman brings the book in which it is written about the sin of adultery and its punishment and gives it to (lit.: in the hand of) the king.” (Axenov)

iii. IND.MOD + head noun + -¯e (altogether 7 examples):

13 ar ˇc¯ız-¯e-r¯a ki mn¯ı dil b-kašš-¯ıt wad¯ı-a kan-¯ın
every thing-IND-OBJ SUB PERS.1SG.GEN heart SBJ-pull.PRS-3SG found-V.EL do.PRS
“I find whatever my heart longs for.” (Axenov) -1SG
14 uštur ar s¯un-¯e ki d¯emay-a kap-¯ıt raw-t
camel every direction-IND SUB in front of.ENCL.3SG-V.EL fall.PRS-3SG go.PRS-3SG
“The camel goes in whatever direction it pleases (lit.: whatever direction shows up before the camel it keeps going).” (Axenov)


There are 27 cases of a head noun marked with -¯e and there are 11 without the suffix. The predominant construction is the one with a demonstrative pronoun before the head noun, which takes the suffix -¯e. It is not unexpected that this dialect shows profound similarities with the Balochi of Afghanistan, from which it branched off about 100 years ago. In this dialect, the suffix -¯e can combine with the case suffix -r¯a (ex. 13, where the head noun is the direct object both of the relative clause and the main clause) but more often the case marking of the head noun is omitted when it is marked with -¯e (ex. 11, where the head noun is the direct object both of the relative clause and the main clause, and ex. 12, where the head noun is the direct object of the main clause)

.4.3 Western Balochi from Pakistan . dialect of Noshke (Nushki)

The town of Noshke in Western Pakistan is the birthplace of two famous contemporary Baloch poets, Gul Khan Nasir and Azat Jamaldini (see JAHANI 1996). Other persons who have supported and developed the Balochi language and its literature also originate from Noshke, notably Abdullah Jan Jamaldini (Azat’s brother) and Aqil Khan Mengal. The Noshke dialect is the basis of the most comprehensive course book in Balochi produced to date, A Course in Baluchi (1969) by Muhammad Abd-al-Rahman BARKER and the above mentioned Aqil Khan MENGAL. The dialect is represented here by ABDULLAH JAN JAMALDINI’s text (1957:8-12) about Balochi embroidery that appeared in the magazine M¯aht¯ak Bal¯oˇc¯ı21 and a biographic text (about his brother) by the same author published in Azat Jamaldini’s collection of poems Ružn (AZAT JAMALDINI 1985:6-42), as well as by Gul Khan Nasir’s preface to Grand, one of his collections of poems (GUL KHAN NASIR 1971:10-27). In addition to these texts, four folktales collected in Noshke by AQIL KHAN MENGAL (1973:10-14, 35- 43, 81-93, 102-112), conversational texts (BARKER/MENGAL 1969/I:425-456, 499-526), ethnographic texts (BARKER/MENGAL 1969/II:95-99, 115-119) and the story “W¯aˇja panˇc-kuš” (BARKER/MENGAL 1969/II:172-181) have been included.22

A. Without the suffix -¯e
i. DEM + head noun (altogether 15 examples):23
15 malang ¯a-¯ı hamr¯ah b¯ut am-¯a malang ki
dervish DEM-OBL company become.PST.3SG EMPH-DEM dervish SUB
“The dervish joined him on the road, the very dervish who had been at the graveyard.” (AQIL
KHAN MENGAL 1973:91)
16 arab¯ı-ay ham-a¯ harf-㯠ki a¯-w-a¯n¯ı tawa¯r-a¯
m¯a ˇca wat-¯ı gu ˙t ˙t-¯a dar kurt na-kan- ¯ ˜u …
PRON.1PL24 from REFL-GEN throat-OBL out do.PST NEG-do.PRS-1PL
ˇca bal¯oˇc¯ı abˇiad-ay pa ˙t ˙t¯ı-¯a dar kan-ag bi-bant
from Balochi alphabet-GEN embroidered headband-OBL out do-INF SBJ-be.PRS.3PL “Those letters from Arabic whose sound we cannot produce (lit.: get out from our throat) should be removed from the Balochi alphabet (lit.: embroidered headband of the Balochi alphabet).”
(GUL KHAN NASIR 1971:23)
ii. IND.MOD + head noun (altogether 2 examples):

17 har kass-a¯ ki … mašhu¯r-e˜¯ ru¯s¯ı raka¯sa tama¯ra¯ xa¯nu¯m-a¯
every person-OBL SUB famous-ATTR Russian female dancer NP lady-OBJ
d¯ıst-ag-at ¯ay-r¯a macl¯um-int ki

t¯aˇiik u uzbak ˇiin¯en-¯an¯ı pašk u d¯oˇc ˇcinkas ˇca bal¯oˇc¯ı-¯a
NP and NP woman-PL.GEN dress and embroidery how much from Balochi-OBL
“Whoever has seen the famous Russian dancer Mrs. Tamara knows (lit.: it is evident to him/her) how similar the Tajik and Uzbek women’s dress and embroidery is to the Balochi (one).”
18 ¯az¯at-¯a har ˇc¯ı zarr ki ša wat-¯ı r¯ozg¯ar-¯a
NP-OB every thing money SUB from REFL-GEN work-OBL
ˇca man-¯ı pag¯ar-¯a u ˇca d¯ost-¯an kumak u w¯am
from PERS.1SG-GEN salary-OBL and from friend-PL.OBL help and loan
rast ¯a-¯ı pa “bal¯oˇc¯ı” m¯aht¯ak-¯a xarˇc kurt-ant
arrive.PST.3SG DEM-OBL for NP monthly magazine-OBL expense do.PST-3PL “All the money that Azat got (lit.: that reached Azat) from his own work, from my salary and
from friends as support or loan he spent on Monthly Balochi.” (AZAT JAMALDINI 1985:19-20)

B. With the suffix -¯e
i. DEM + head noun + -¯e
There are no examples of this construction in the texts investigated for this dialect.

ii. Head noun + -¯e (altogether 1 example):
19 mard-¯e ki ˇca panˇc maz¯ar-ay kuš-ag-¯a baˇiˇi na-w¯art
man-IND SUB from five tiger-GEN kill.PRS-INF-OBL bend NEG-eat.PRS.3SG
e¯ yakk-e˜¯ maza¯r-ay kuš-ag pa a¯-¯ı hicˇcˇ cˇ¯ı-h-e¯
DEM one-ATTR tiger-GEN kill.PRS-INF for DEM-OBL no thing-GL-IND
b¯ut na-kant
“A man who is not afraid of (lit.: does not bend from) killing five tigers, to kill this one tiger
cannot be any problem (lit.: thing) for him.” (BARKER/MENGAL 1969/II:174-175)

iii. IND.MOD + head noun + -¯e (altogether 4 examples):
20 yakk diga ˇc¯ı-¯e ki grand-ay niwištag-¯an mã šumay d¯em¯a
one other thing-IND SUB NP-GEN writing-PL.OBL in PERS.2PL.GEN in front of
kayt ¯a “hamza” u “he”-ay k¯armarz kan-ag-int
come.PRS.3SG DEM hamza and he-GEN use do.PRS-INF-COP.PRS.3SG
“Another thing that appears to you in the writings of (the book) Grand is the use of (the letters)
hamza and he.” (GUL KHAN NASIR 1971:24)

First, it is interesting to note that there are notably few restrictive relative clauses to be found in the texts selected for this dialect (see also Table 3 below). It will require further studies to determine whether this is representative of the dialect. When this dialect is compared with the dialects described above (in 4.1 and 4.2) we note that a different construction predominates, namely the one without the suffix -¯e. There are 17examples of this pattern, and only five of the pattern with the suffix -¯e, four of which are in combination with an indefinite pronoun. There are no co-occurrences of the demonstrative pronoun and the -¯e, neither are there any examples where -¯e occurs together with a case ending. It is thus interesting to note that in this dialect, which is not deeply influenced by Persian today, the structure of relative clauses is more similar to early Classical Persian than to spoken or written Modern Persian.

4.4 Southern Balochi from Pakistan25

Since the mid-20th century a significant number of Baloch in Karachi and other urban centres in the south of Pakistani Balochistan such as Turbat, Mand and Habb have engaged in literary and cultural activities. Most of the new generation of writers, in fact, use the southern dialect. It is here represented by seven articles on the Balochi language from SABA DASHTYARIfs two-volume work BalPo.cP. zubPanPe26 Pakubat (BZA). These selections are by Saba Dashtyari himself (BZA II:13-28), Sayad Hashmi (BZA I:309- 333), Muhammad Beg Begal (BZA II:105-128), Badal Khan Baloch (BZA II:129-146), Siddiq Azat (BZA II:182-187), G. R. Mulla (BZA II: 597-603) and Karim Dashti (BZA II:604-611). Also included are two short stories written by Ghaws Bahar (GHAWS BAHAR 2003:5-22) and a part of a novel by Ghani Parwaz (GHANI PARWAZ 2000:5-25).

A. Without the suffix -¯e

i. DEM + head noun (altogether 25 examples):

21 likwa ˙r say bundar¯ı rahband- ¯ ã bahr kan-ag bant
writing system three fundamental system-PL.OBL division do.PRS-INF be.PRS.3PL
bz㯠awl¯ı ham-a¯ zuba¯n ki cˇa cˇapp-e¯n pahna¯t-a¯
namely first EMPH-DEM language SUB from left-ATTR side-OBL
r¯ast-¯en n¯emag-¯a nib¯ıs-ag bant
right-ATTR side-OBL write.PRS-INF be.PRS.3PL
“Writing systems can be divided into three main groups: the first (one is) those languages that are written from left to right.” (Muhammad Beg Begal in BZA II:117)
22 har kas-¯a ki turk¯ı zub¯an-¯e t¯ar¯ıx want-ag ta ¯a every person-OBL SUB Turkish language-GEN history read.PST-PP SUB DEM
ˇiw¯an¯ı-¯a sarpad-int ki ˇca kam¯al ataturk-¯a s¯ar¯ı ham goodness-OBL knowing-COP.PRS.3SG SUB from NP-OBL before also
arab¯ı alifb¯a-h-¯an¯ı b¯abat¯a bahs u tr¯an b¯u-ag-¯a-at Arabic alphabet-GL-PL.GEN about discussion and talk be.PRS-INF-OBL-COP.PST.3SG
“Everybody who has read (about) the history of the Turkish language is well aware of (the fact that) even before Kemal Atatürk there had been an ongoing discussion about Arabic alphabets (i.e. alphabets employing the Arabic script).” (Badal Khan Baloch in BZA II:136)

ii. IND.MOD + head noun (altogether 2 examples):

23 ¯e guš-ag muškil b¯ıt ki ¯e dur¯ah-¯en zub¯an
DEM say.PRS-INF difficult be.PRS.3SG SUB DEM all-ATTR language
sarˇiam u k¯amil- ¯ ˜e zub¯an-ant y¯a gu ˙ra¯ lahte˜¯
total and complete-ATTR language-COP.PRS.3PL or then some.ATTR
ancˇe˜¯ zuba¯n-ant ki e¯waka¯ e¯ diga zuba¯n-a¯n¯ı
such.ATTR language-COP.PRS.3PL SUB only DEM other language-PL.GEN
g¯alw¯ar u lahˇia-ant
speech and dialect-COP.PRS.3PL
“It is difficult to say whether all these languages are languages in their own right (lit.: total and complete languages) or if there are some “languages” of such a kind that they are only dialects of these other languages.” (Saba Dashtyari in BZA II:21)

B. With the suffix -¯e
i. DEM + head noun + -¯e
There are no examples of this construction in the texts investigated for this dialect.
ii. Head noun + -¯e:
There are no examples of this construction in the texts investigated for this dialect.

iii. IND.MOD + head noun + -¯e (altogether 5 examples):

24 har kas-¯e ki ˇca angr¯ez¯ı zub¯an-¯a z¯antk¯ar-int
every person-IND SUB from English language-OBL knowledgeable-COP.PRS.3SG
¯a ˇca angr¯ez¯ı zub¯an-¯e b¯erahband¯ı-¯an ham ˇiw¯an z¯antk¯ar-int
DEM from English language-GEN irregularity-PL.OBL also good knowledgeable-
“Anyone who knows English is also well aware of the irregularities of the English language.”
(Sayad Hashmi in BZA I:327)


Very few restrictive relative clauses are present in the fiction texts (the text by Ghani Parwaz contains none), though there are more in the factual prose texts. As the data show, also in this dialect the construction without the suffix -¯e predominates. There are 27 examples of this pattern, and only 5 of the pattern with the suffix -¯e, all in combination with an indefinite pronoun in open restriction and with a head noun in the singular.
The -¯e does not combine with the case ending -¯a (OBL and OBJ SG) in the data examined for this dialect.27 When the construction without -¯e occurs with an indefinite pronoun in open restriction the head noun has a plural reference. There are no co-occurrences of the demonstrative pronoun and the suffix -¯e in the texts investigated here.

4.5 Eastern Balochi28

Eastern Balochi is the dialect that has been used the least as a literary medium among the Baloch themselves. On the other hand, it has been better described (by British officials) than the other Balochi dialects (see e.g. KORN 2005:33-34). Eastern Balochi is here represented by texts from DAMES’ A Text Book of the Balochi Language, Part I “Miscellaneous stories” (1891:1-30), as well as by two articles from Bal¯oˇc¯ı zub¯an¯e ¯akubat written by speakers of Eastern Balochi; Aziz Muhammad Bugti (BZA I:389-395) and Surat Khan Marri (BZA II:223-230). These two persons, however, live in Quetta and their written language is characterised by a considerable amount of dialect mixture, possibly intentionally so to make it easier for non-Eastern speakers to read their texts (see also JAHANI 1989:108). Since the patterns found in these texts conform to those found in Dames’ texts, we can assume that they can serve as examples of Eastern Balochi patterns for relative clauses.

A. Without the suffix -¯e

i. DEM + head noun (altogether 25 examples):

25 mã haw-㯠mard gir-㯠ki mã wa\ khuš khan-ã¯
“I will marry a man whom I myself select.” (DAMES 1891:Part I:5)
26 ma th ¯arum¯e ˙ g- ¯ ˜e šaf-¯a haw- ¯ ã mardum-¯a ki gind- ¯ ã r¯oš-¯a gu ˙d¯a
in dark-ATTR night-OBL EMPH-DEM person-OBJ SUB see.PRS-1SG day-OBL then
sad mar nyã¯w㯠b¯ı\-¯ı phaˇiya¯ kha¯r-a¯n-¯ı
a hundred man among be.PRS.3SG-ENCL.3SG recognition bring.PRS-1SG-ENCL.3SG
“Whoever I see in the darkness of the night (lit.: in the dark night), I will recognise in daylight,
even if he be among a hundred men.” (DAMES 1891:Part I:17)

ii. IND.MOD + head noun (altogether 2 examples):

27 ba¯z kam cˇus-e˜¯ mardum bant ki a¯ bunge¯ˇi¯ı zuwa¯n u
very few such-ATTR person be.PRS.3PL SUB DEM basic language and
ham-¯a zuw¯an ki tarr¯en-a ˙ g-¯en-¯ı
EMPH-DEM language SUB translate.PRS-INF-COP.PRS.3SG-ENCL.3SG
yak ˙dawl-¯a b-z¯an-t
one manner-OBL SBJ-know.PRS-3SG
“There are very few such people who know the source language and the language into which they
translate equally well.” (Surat Khan Marri in BZA II:224)29

B. With the suffix -¯e
i. DEM + head noun + -¯e
There are no examples of this construction in the texts investigated for this dialect.
ii. Head noun + -¯e (altogether 1 example):
28 khas-¯e-¯a ki ya r¯upi¯a-¯e saud¯a khu\-¯a ˇcy¯ar r¯upi¯a
person-IND-OBL SUB one rupee-IND business do.PST-PP four rupee
saud¯agar-¯a d¯a\-a-iš
businessman-OBJ give.PST-PP-ENCL.3SG30
“A person who did business worth one rupee, to him the businessman gave four rupees.” (DAMES
1891:Part I:12)

iii. IND.MOD + head noun + -¯e (altogether 2 examples):
29 cˇo¯š-e˜¯ mard-e¯ b¯ı ki guna¯h cˇ¯ı na-khu\-a
such-ATTR man-IND be.PRS.3SG SUB sin any NEG-do.PST-PP
“It should be such a man who has not committed any sin.” (DAMES 1891:Part I:5)

In the Eastern dialect also the construction without the suffix -¯e predominates. Again, there are 27 examples of this pattern. Only three fit the pattern with the suffix -¯e, all of them in open restriction and with a head noun in the singular. Note, however, that although all occurrences with -¯e are in an open restriction, not all open restrictions have this suffix (see ex. 27). When the construction without the suffix -¯e occurs with an indefinite modifier in open restriction the head noun has a plural reference. In the data for Eastern Balochi the suffix -¯e is found in combination with the case marker -¯a (ex. 28, where the head noun is the agent in an ergative construction in the relative clause). As in all the other variants of Balochi spoken in Pakistan, there are no co-occurrences of the demonstrative pronoun and the suffix -¯e in Eastern Balochi.

4.6 Balochi from Iran31

The only data available in writing for Iranian Balochi are a few stories published in ELFENBEIN (1966) and some articles written by Iranian Baloch and published in books and journals edited in Pakistan. The language of the latter is to a certain degree influenced by Pakistani dialects, where the writers lived when they wrote these articles, and also by the “semi-standard” literary language used in Pakistan. Therefore only oral data have been analysed here, namely 4.5 hours of recordings in the Western and Southern Balochi dialects as well as in the Central Sarawani dialect. The recordings consist of tales, anecdotes, monologues and conversations on various topics and also of a certain amount of elicited data. About 40 Baloch from 15 to 80 years old acted as language consultants, and they all agreed to have their speech recorded. Most of them are educated, but about one hour consists of recordings of illiterate elderly Baloch. Both genders are well represented.32

On the whole, it is notable that the number of relative clauses in these data is very low and particularly so in the recordings from non-educated speakers. Only a limited number of the speakers use more than the odd one relative clause in their speech, and then mostly when they talk about a more formal subject, such as literature, history or contemporary society. For Central Sarawani only one relative clause was found in more than one hour of recordings of 16 different persons talking about various subjects. There are no observable differences in my data between patterns of restrictive relative clauses in the two main dialect groups represented in Iran, Western and Southern Balochi. Therefore all the Iranian dialects will be dealt with as one unit for this particular grammatical construction. Further investigations may reveal dialect differences when it comes to marking the antecedent between different variants of Iranian Balochi, possibly geolectal (based on geographical distribution) or, maybe more likely, sociolectal (based on education, occupation and other social factors).

A. Without the suffix -¯e/-¯ı

i. DEM + head noun (altogether 5 examples, 2 in elicited data)

30 ¯a mard¯enzahg ke ¯ed¯a yaht mahm¯ud-¯ı pes-at
DEM man SUB here come.PST.3SG NP-GEN father-COP.PST.3SG
“The man who came here was Mahmud’s father.” (Elicited) (W.Bal.)
31 ¯a ket¯ab ke ta man-a neš¯an d¯a
DEM book SUB PERS.2SG PERS.2SG-OBJ sign give.PST.3SG
xeyl¯ı x¯ub-¯en-¯e-at
very good-ATTR-IND-COP.PST.3SG
“That book which you showed me was a very good one.” (Elicited) (W.Bal.)33

32 ham-a¯ mardom ke m㯠deha¯t-a¯n ˇia¯mnend-㯠a¯y-a¯n
EMPH-DEM people SUB in village-PL living-COP.PRS.3PL DEM.OBL-PL
waš-¯en o ˙dawld¯ar-¯en bal¯oˇc¯ı kan-ant
good-ATTR and beautiful-ATTR Balochi do.PRS-3PL
“Those people who live in villages, they speak good and beautiful Balochi.” (S.Bal.)

ii. IND.MOD + head noun:

There are no examples of this construction in the texts investigated for this dialect.

B. With the suffix -¯e/-¯ı

i. DEM + head noun + -¯e/-¯ı (altogether 8 examples):

33 ham-¯a kas-¯e ke n¯an-¯ar¯a wat-¯ı-¯a dar y¯ar-¯ı
EMPH-DEM person-IND SUB bread-OBJ REFL-GEN-OBJ out bring.PRS-3SG
mard¯enzahg mar¯oˇc¯ı [sard¯ar] ham-¯a-¯ı en
man today chief EMPH-DEM-OBL COP.PRS.3SG
“Anyone, (any) man, who earns his own bread, it is he who is a chief nowadays.” (W.Bal.)
34 ¯a ˇc¯ız-¯an-¯e ke … l¯azem-o b-z¯ur-¯en-eš
DEM thing-PL-IND SUB necessary-COP.PRS.3SG SBJ-take.PRS-1PL-ENCL.3PL
“Let’s take (with us) the things that we will need (lit.: are necessary).” (Ce.Sar.)
ii. Head noun + -¯e/-¯ı (altogether 10 examples, 1 in elicited data):
35 l¯o ˙t-¯ın darb¯are-ye m¯em¯an wa ezzat-¯e ke m¯em¯an d¯ar-¯ı
want.PRS-1SG about-I ˙ Z guest and honour-IND SUB guest hold.PRS-3SG
kessa-¯e bo-goš-¯ın
tale-IND SBJ-say.PRS-1SG
“I want to tell a story about guests and the honour which a guest should be paid.” (W.Bal.)
36 ket¯ab-¯ı ke ta-r¯a neš¯an d¯at-on b¯az ˇiw¯an-¯en-¯e
book-IND SUB PERS.2SG-OBJ sign give.PST-1SG very good-ATTR-IND
“The book that I showed you is a very good one.” (Elicited) (W.Bal.)34
37 bal¯oˇcest¯an-¯e t¯ok¯a kas-¯e n¯e ke ¯a-¯ı n¯am-¯a
Balochistan-GEN in person-IND NEG.COP.PRS.3SG SUB DEM-GEN name-OBL
gõ¯ a¯šna¯ ma-b¯ı
with acquainted PROH-be.PRS.3SG
“In Balochistan there is nobody who is not acquainted with his name.” (S.Bal.)
38 gapp-a¯-¯ı ke man kan-㯠do sad sa¯l gwast-ag-e˜¯
talk-PL35-IND SUB PERS.1SG do.PRS-1SG two hundred year pass.PST-PP-ATTR
balo¯cˇ-a¯n¯ı gapp-ã¯
Baloch-PL.GEN talk-COP.PRS.3PL
“The things I say are about the Baloch of 200 years ago.” (S.Bal.)

ii. IND.MOD + head noun + -¯e/-¯ı (altogether 3 examples)

39 agar yak kas-¯e hat-at ke poˇcˇc-¯an-¯ı
if one person-IND be.PST-COP.PST.3SG SUB clothes-PL-ENCL.3SG
dert-at-ant o ˇc¯o gan¯ok-¯a gaˇiˇi-¯ı kort-at …
tear.PST-COP.PST-3PL and like crazy-OBL foam-ENCL.3SG do.PST-COP.PST.3SG
go ˙d¯a ta ham-¯ay-r¯a b-y-¯ar pa man
“If there is such a person (in the group) whose clothes are torn and who has foam around his
mouth like a lunatic, then bring this very one to me.” (W.Bal.)
40 har ens¯an-¯ı ke beˇi¯a-ye yak zab¯an ˇc¯ar-t¯a balad
every human-IND SUB instead of-I ˙ Z one language four-CLASS acquainted
b¯ı beˇi¯a-ye ˇc¯ar nafar-˜e
be.PRS.3SG instead of-I ˙ Z four person-COP.PRS.3SG
“Any person who instead of one language knows four is like four persons.” (S.Bal.)

In Iranian Balochi the construction with -¯e/-¯ı is predominant. There are 21 examples of this pattern, but only five of the pattern without the -¯e/-¯ı. Perhaps the construction without -¯e/-¯ı is regarded as “more correct”, since one informant used it in elicited data (exs. 30, 31), in spite of the fact that the sentence to be translated into Balochi was given in Persian, i.e. with the head noun marked with the suffix -¯ı. Iranian Balochi dialects are the only ones that add the suffix -¯e/-¯ı after the plural suffix (exs. 34, 38).
There are no co-occurrences in the data of an indefinite modifier and a head noun without the suffix -¯e/-¯ı. Furthermore, in these data no case endings (-¯a or -r¯a) are found on any antecedents, neither those marked with -¯e/-¯ı, nor those without this suffix.36


This investigation shows that the Balochi dialects are neatly divided into three groups when it comes to the marking of the head noun of restrictive relative clauses with or without the suffix -¯e/-¯ı (see Tables 3 and 4). The three dialects where the pattern without the -¯e is totally predominant, and where this marker is found only if the restriction is open (indefinite, non-specific) and the noun occurs in the singular are those spoken in Pakistan. These dialects have had only limited influence from Persian, and the influence has been mainly from Classical Persian. It is interesting to note that all these dialects show a lower percentage of the suffix -¯e than even early Classical Persian does. The fact that -¯e only occurs in open restriction further strengthens the argument that this ending is historically the same as the marker of indefinite singularity (see section 1.).

Table 3: Marking of the head noun in different Balochi dialects (occurrrences in the

Dialect A. Without the suffix -¯e/-¯ı B. With the suffix -¯e/-¯ı Total
i. DEM +
+ head
i. DEM +
head + -¯e/-¯ı
ii. head +
iii. IND.MOD
+ head + -¯e/-¯ı
E.Bal. 25 2 – 1 2 30
S.Bal.Pak. 25 2 – – 5 32
W.Bal.Pak. 15 2 – 1 4 22
BA 15 2 16 9 6 48
BT 6 5 16 4 7 38
BI 5 – 8 10 3 26
Total 91 13 40 25 27 196

Table 4: Marking of the head noun with or without the suffix -¯e/-¯ı in different Balochi

dialects in comparison with Classical and Modern Persian (in percentages)
Language variant A. Without the suffix -¯e/-¯ı B. With the suffix -¯e/-¯ı
E.Bal. 90% 10%
S.Bal.Pak. 84% 16%
W.Bal.Pak. 77% 23%
BA 35% 65%
BT 29% 71%
BI 19% 81%
10th-12th cent. texts (early Classical) 54% 46%
13th-15th cent. texts (late Classical) 43% 57%
16th-19th cent. texts (Transitional) 30% 70%
20th cent. texts (Modern) 12% 88%
contemporary spoken37 8% 92%38

The second group are those dialects that have historically been in closer contact with Persian and have been strongly influenced by it (mainly in the Afghani variant Dari).
These are the dialects spoken in Afghanistan and Turkmenistan, which show a picture similar to that of Persian from the 16th to 19th centuries. In this context it would also be interesting to investigate spoken Persian of Iran and Afghanistan more closely, but that investigation is outside the scope of this article. It seems, though, that spoken Persian has been the main source of influence on these dialects, since the double
marking of the head noun (both with the demonstrative pronoun and the suffix -¯e) that is common here is also more common in spoken than in written Persian.
The remaining Balochi dialects, i.e. those spoken in Iran, are nowadays under particularly heavy influence from Persian, both in its written and spoken form. Virtually the whole province of Sistan and Balochistan has electricity and TV is common even in remote areas. Furthermore, most Baloch children nowadays attend at least primary school. The two predominant structures are those with the same structure as the most common construction in spoken Modern Persian (DEM + head noun + -¯e/-¯ı) and in written Modern Persian (head noun + -¯e/-¯ı). The percentage of occurrences with the suffix -¯e/-¯ı is almost as high in Balochi dialects of Iran as in Modern Persian. Subordinate constructions are rather rare in Balochi in general, relative clauses occur more frequently in written than in spoken styles, and written styles are a comparatively new phenomenon in Balochi. Therefore, one might wonder whether the whole structure of relative clauses has been copied from Persian. If so, it probably took place very early, i.e. before the 10th century, since the demonstrative pronoun and the suffix -¯e do not co-occur in those dialects of Balochi that have not been under continuous strong influence from Persian, whereas they do already in early Classical Persian.
In this context it is also interesting to note that relative clauses seem to be more common in formal language than in spoken language or tales in oral literary style. This study has not attempted to quantify how frequently relative clauses occur in different types of Balochi texts, nor has it made any comparison with Persian regarding the frequency of relative clauses. The main reason for this is that the material for the different dialects is somewhat divergent, e.g. orally recorded data for one dialect and written texts for another, or mainly folktales for one dialect and factual prose for another. A comparison of different types of text in Balochi and/or Persian texts will make an interesting study and may reveal considerable intralinguistic stylistic variation as well as differences between the two languages when it comes to using restrictive relative clauses.


1, 2, 3 1st, 2nd, 3rd person I ˙ z i ˙z¯afa
ATTR attributive NEG negation particle
BA Balochi from Afghanistan NP proper noun (nomen proprium)
BI Balochi from Iran OBJ object (direct and indirect objects)
BT Balochi from Turkmenistan OBL oblique (agent, locative and prepositional)
BZA I, II SABA DASHTYARI 1995, 1998 Pak. Pakistan
Ce.Sar. Central Sarawani PL plural
CLASS classifier PP past participle
COP copula PROH prohibitive particle (negation of
DEM demonstrative pronoun imperatives and subjunctives)
E.Bal. Eastern Balochi PRON personal pronoun
EMPH emphatic PRS present stem
ENCL enclitic pronoun PST past stem
ex., exs. example(s) REFL reflexive pronoun
EXCL exclusive S.Bal. Southern Balochi
GEN genitive SBJ subjunctive
GL glide SG singular
INCL inclusive SUB subordinator (clause linker)
IND indefinite V.El. verbal element
IND.MOD indefinite modifier WBal. Western Balochi
INF infinitive


1. The term “New Persian” is used here to denote the Persian language after the conquest of Islam, or rather after A.D. 850, written mainly in Arabic script. Below only “Persian” will be used in this sense. “Classical Persian” denotes the language down to the end of the 15th century, and “Modern Persian” the language in the 20th century. The time between 1500 and 1900 is described as the “transitional period” between the Classical and the Modern stages of the Persian language.

2. This is the opinion of Rubinˇcik (see WINDFUHR 1979:35-36), HINCHA (1961:173) and LAZARD (1966:263). My own investigation of New Persian (JAHANI 2000) indicates that this is indeed the case, since the suffix -¯ı marking the antecedent of restrictive relative clauses is attached mainly to singular nouns denoting an open (indefinite) restriction in early Classical Persian, a natural place to find a y¯a-ye va ˙hdat also on nouns that are not antecedents of restrictive relative clauses. This suffix has then been grammaticalised as a marker of both singular and plural antecedents with either open or closed restriction in Modern Persian. The findings in Balochi presented in this article strengthen the argument that marking the antecedent with the suffix is an extended use of the y¯a-ye va ˙hdat. A purely synchronic description of, e.g., Modern Persian may, however, label them as two different suffixes, because in this variant of Persian they are used in two very different ways: as a marker of
indefinite selection and as a marker of antecedents of restrictive relative clauses.

3 .The term “antecedent” is also used in this article to denote the head noun of relative clauses.

4. COMRIE (1989:143) regards restrictive relative clauses “more central to the notion of relative clause” than non-restrictive relative clauses, and FABB (1994:3520), following this argument, argues that nonrestrictive relative clauses are a structural imitation of restrictive relative clauses.

5. The earliest written texts in Balochi that we know of are three manuscripts in the British Library, dating from the 19th century (ELFENBEIN 1983:1-2).

6. 6 Some modifications of transcription have been carried out in the example sentences. Stress marking has been removed if the original source has it. The vowels are consistently represented by the symbols a, i, u (short vowels) and ¯a, ¯ı, ¯u, ¯e, ¯o (long vowels) for other dialects than those spoken in Iran, and with the same symbols for the long vowels but a, e, o for the short vowels in Iranian Balochi. Also some other modifications have been carried out, such as the replacement of c and j with ˇc and ˇi, respectively (both used in the sources for the postalveolar affricates), as well as insertion of a glide even if a hamza is used in the text. There is no marking of different letters in the
Arabic script representing the same phoneme in Balochi. Dames’ transcription has also been modified to a certain extent; vowel length is marked as described above and digraphs are replaced by a single sign, e.g. š for sh and ˇc for ch. The superscript h denotes aspiration in Eastern Balochi.

7 case systems in different Balochi dialects, see JAHANI (2003) and Korn in this volume.

8. Or rather non-past, since present also includes future. See e.g. AXENOV 2006:175.

9. See also JAHANI (2000:49), where the figures include both noun and pronoun heads.

10. The noun phrase may also contain a demonstrative pronoun P.n/Pan. This is very common in pattern A, but rare in pattern B in written Persian (see JAHANI 2000:47-48).

11. In later times the administrative language was changed to English, see JAHANI (2005:153).

12. The fricatives /f/, /x/ and / . g/ are found in Eastern and Western, but not normally in Southern Balochi. They are part of the Eastern Balochi sound system whereas in Western Balochi they occur only in loanwords (see DAMES 1891: Balochi Grammar: 2, 4, and JAHANI 1989:82-84).

13. For the area where the different dialects are spoken, see map xy.

14. An example of this type is ˙ gullu ˙ gd¯ar yak r¯oˇc¯e ki sawd¯agir bi diga ˇi¯a¯e šutat, wat¯ı xizmatk¯ar¯er¯a d¯em d¯at ki sawd¯agiray ˇiin¯en¯a ¯a¯ı¯a by¯ar¯ıt “one day when the merchant had gone to another place, the employee sent a servant of his to bring the merchant’s wife home to himself” (BT).

15. For descriptions of this dialect, see NAWATA (1981) and BUDDRUSS (1988).

16. Sincere thanks to Behrooz Barjasteh Delforooz for putting this material at my disposal.

17. In the TAM-form perfect, the third person singular appears without the copula. The form ni.t-a is thus a finite verb form, namely perfect, 3SG.

18. This dialect has been described by, among others, SOKOLOVA (1953), SOKOLOV (1956) and AXENOV (2006). ZARUBIN (1932 and 1949) has published two volumes of folktales in this dialect with Russian translations and ELFENBEIN (1963) is a glossary for these.

19. Sincere thanks to Serge Axenov for putting this material at my disposal. Restrictive Relative Clauses in Balochi and the Marking of the Antecedent

20. I here follow Axenov’s analysis of the case system in Balochi of Turkmenistan (AXENOV 2006:69- 83). The oblique case ending in -¯a (SG) and -¯an (PL) is used only after a preposition in this dialect and the object case ending in -¯a (SG) and -¯an¯a (PL) is used not only for direct and indirect objects, but also for place and time adverbials, e.g. ˙ gad¯ım¯a bal¯oˇc g¯okay šangulakk¯an¯a li ˙ d˙duk-a muštant “in the old days the Baloch used to rub dung onto the cow’s hoofs”.

21. This text was published with English translation in ELFENBEIN 1990/I:410-419.

22. The genitive ending of nouns for this dialect is transcribed -ay (see BARKER/MENGAL 1969/I:15)
regardless of orthographic conventions used in particular texts.

23. Also counted here is one example with an ordinal (d¯om¯ı “the second”) instead of a demonstrative

24. In Balochi dialects (Western, Southern and Eastern) spoken in Pakistan, there is no distinction
between the first person plural inclusive and exclusive pronoun.

25. Southern Balochi dialects in Pakistan have been described by e.g. PIERCE (1874), MARSTON (1877), GRIERSON (1921) and FARRELL (1990).

26. The genitive ending is for Southern Balochi here invariably transcribed -¯e.

27. In the data from this dialect there are no examples of head nouns marked with -¯e functioning as
agents in ergative constructions or as direct objects, cf. ex. 28.

28. This dialect has been described by, among others, DAMES (1891), GRIERSON (1921) and GILBERTSON (1923). See also Bashir in this volume.

29. Note the change from plural (bant) to singular (tarr¯en-a ˙ g-¯en, bz¯ant) in the sentence.

30. There is no clear distinction upheld between the 3SG and 3PL enclitic pronouns in Eastern Balochi
(cf. ex. 26 and DAMES 1891:Balochi Grammar:17).

31. Descriptions of Balochi dialects spoken in Iran include BARANZEHI (2003), who describes Central Sarawani, and YUSEFIAN (2004), who describes Lashari. See also JAHANI (2003) for a description of the nominal system in Iranian dialects of Balochi.

32. I am grateful to Adam Nader Baranzehi, who has put all his recordings and transcriptions of Central Sarawani language data at my disposal. Many thanks also to Erik Olafsen, Gol Mohammad Arbabi, and the late Abdolhossein Yadegari for making recordings among people of different educational backgrounds and occupations in several places of Iranian Balochistan.

33. Cf. the parallel ex. 36.

34. Cf. ex. 31.

35. The plural suffix -¯a is a borrowing from colloquial Persian.

36. There are no examples in the data where the head noun is the direct object both in the relative clause and in the main clause. However, in the following sentences, found in a translation of stories from the Old Testament from English into Iranian Balochi, a definite head noun which is the direct object both of the main clause and the relative clause is in one instance found without case marking and in another instance with the case marker -¯a:

41. la ˙t ˙t-o-d¯ar-¯e ke pa wat-¯ı korb¯an¯ı-¯a tay¯ar-¯ı
stick-and-wood-IND SUB for REFL-GEN sacrifice-OBL ready-ENCL.3SG
kot-ag-at ish¯ak-¯ı sar¯a ¯er kot
do.PST-PP-COP.PST.3SG NP-GEN on down do.PST.3SG
“He put the wood that he had prepared for his sacrifice on Isac(’s shoulders).” (Genesis 22:6)

42. pahk¯en ˇiarr-o-d¯ar-¯e-¯a ke trongol na-košt-ag-at
all shrub-and-wood-IND-OBJ SUB hail NEG-kill.PST-PP-COP.PST.3SG
madag w¯art-ant
grasshopper eat.PST.3PL
“The grasshoppers ate all the vegetation that the hailstorm had not killed.” (Exodus 10:15)

37. This is a limited investigation based on interviews and two hours of recorded language data. See
JAHANI 2000:49-51.
38. Here the two patterns found are DEM + head noun + -¯ı (67%) and head noun + -¯ı (33%). The predominant pattern found in spoken Modern Persian is thus DEM + head noun + -¯ı. This result agrees with my own participant observation during more than 20 years of friendship with Persian speakers.


ABDULLAH JAN JAMALDINI (cAbdull¯ah J¯an Jam¯ald¯ın¯ı) 1957: “Bal¯oˇc¯ı d¯oˇc”. In: M¯aht¯ak Bal¯oˇc¯ı
(Jan. 1957), pp. 8-12
AQIL KHAN MENGAL (cA¯ qil Xa¯n Me¯ngal) 1973: Ge¯d¯ı kissaw 7. Quetta: Balochi Academy
AXENOV, Serge 2006: The Balochi Language of Turkmenistan [Studia Iranica Upsaliensia 10].
Uppsala: Uppsala University
AZAT JAMALDINI ( ¯ Az¯at Jam¯ald¯ın¯ı) 1985: Ružn. Karachi: Al-Maxzan Prin ˙tirz
BARANZEHI, Adam Nader 2003: “The Sarawani Dialect of Balochi and Persian Influence on It”.
In: JAHANI/KORN, pp. 75-111
BARKER, Muhammad A., and Aqil Khan MENGAL 1969: A Course in Baluchi. Montreal: McGill
University, 2 vol.
BUDDRUSS, Georg 1988: Aus dem Leben eines jun jungen Balutschen, von ihm selbst erzählt [Abhandlungen für die Kunde des Morgenlandes 48/4]. Stuttgart: Steiner
COMRIE, Bernard 1989: Language Universals and Linguistic Typology. Syntax and Morphology.
Oxford (UK) / Cambridge, Mass.: Blackwell (2nd ed.)
DAMES, M. Longworth 1891: A Text Book of the Balochi Language, consisting of Miscellaneous Stories, Legends, Poems, and a Balochi-English Vocabulary. Lahore: Punjab Government Press
ELFENBEIN, Josef 1963: A Vocabulary of Marw Baluchi [Quaderni della sezione linguistica degli Annali 2]. Naples
––– 1966: The Baluchi Language: A Dialectology with Texts [Royal Asiatic Society Monographs XXV]. London: Royal Asiatic Society of Great Britain and Ireland
Restrictive Relative Clauses in Balochi and the Marking of the Antecedent 165
––– 1983: A Baluchi Miscellany of Erotica and Poetry: Codex Oriental Additional 24048 of the
British Library. Naples: Istituto Universitario Orientale
––– 1990: An Anthology of Classical and Modern Balochi Literature. Wiesbaden; Harrassowitz, 2 vol.
FABB, Nigel 1994: “Relative Clauses”. In: Ronald E. ASHER, Joy M. Y. SIMPSON (eds.): The
Encyclopedia of Language and Linguistics 7. Oxford etc.: Pergamon Press, pp. 3520-3524
FARRELL, Tim 1990: Basic Balochi. An Introductory Course [Baluchistan Monograph Series 1].
Naples: Istituto Universitario Orientale
GHANI PARWAZ (G˙ an¯ı Parwa¯z) 2000: Mihre¯ ho¯ša¯m. Quetta: Balochi Academy
GHAWS BAHAR (G˙ aws Baha¯r) 2003: Karke¯nk. Quetta: Balochi Academy
GILBERTSON, George W. 1923: The Balochi Language. A Grammar and Manual. Hertford:
Stephen Austin & Sons
GREN-EKLUND, Gunilla 1978: A Study of the Nominal Sentences in the Oldest Upadi ˙sads
[Studia Indoeuropaea Upsaliensia 3]. Uppsala: Uppsala University
GRIERSON, George A. 1921: “Bal¯och¯ı”. In: Linguistic Survey of India X: Specimens of
Languages of the Eranian Family. Calcutta: Superintendent Gov. Printing, pp. 327-451
GUL KHAN NASIR (Gul X¯an Na ˙s¯ır) 1971: Grand. Mastung: Qal¯at Pabliširz
HINCHA, Georg 1961: “Beiträge zu einer Morphemlehre des Neupersischen”. In: Der Islam 37,
pp. 136-201
JAHANI, Carina 1989: Standardization and Orthography in the Balochi Language [Studia
Iranica Upsaliensia 1]. Uppsala
––– 1996: “Poetry and Politics: Nationalism and Language Standardization in the Balochi
Literary Movement”. In: Paul TITUS (ed.): Marginality and Modernity: Ethnicity and
Change in Post-Colonial Balochistan. Karachi etc.: Oxford University Press, pp. 105-137
––– 2000: “Restrictive Relative Clauses in Classical and Modern New Persian and the Marking
of the Antecedent”, in: Orientalia Suecana 49, pp. 33-56
––– 2003: “The Case System in Iranian Balochi in a Contact Linguistic Perspective”. In:
JAHANI/KORN, pp. 113-132
––– 2005: “State Control and its Impact on Language in Balochistan”. In: Annika RABO, Bo
UTAS (eds.): The Role of the State in West Asia [Transactions 14]. Stockholm: Swedish
Research Institute in Istanbul, pp. 151-163
JAHANI, Carina, and Agnes KORN (eds.) 2003: The Baloch and Their Neighbours: Ethnic and
Linguistic Contact in Balochistan in Historical and Modern Times. Wiesbaden: Reichert
KORN, Agnes 2003: “Balochi and the Concept of North-Western Iranian”. In: JAHANI/KORN, pp.
––– 2005: Towards a Historical Grammar of Balochi. Studies in Balochi Historical Phonology
and Vocabulary [Beiträge zur Iranistik 26]. Wiesbaden: Reichert
LAZARD, Gilbert 1966: “L’enclitique nominal -i en persan: un ou deux morphèmes?” In:
Bulletin de la société de linguistique de Paris 61, pp. 249-264
LEHMANN, Christian 1984: Der Relativsatz. Tübingen: Gunter Narr
MAHMOODZAHI, Moosa 2003: “Linguistic Contact in Iranian Balochistan in Historical and
Modern Times”. In: JAHANI/KORN, pp. 147-156
MARSTON, E.W. 1877: Grammar and Vocabulary of the Mekranee Beloochee Dialect. Bombay:
Education Society’s Press
MOSHKALO, Vyacheslav V. 2000: “Language and Culture of the Baloch in Turkmenistan”. In:
Carina JAHANI (ed.): Language in Society – Eight Sociolinguistic Essays on Balochi [Studia
Iranica Upsaliensia 3]. Uppsala: Uppsala University, pp. 97-103
NAWATA, Tetsuo 1981: Baluchi [Asian and African Grammatical Manuals 17b]. Tokyo: Tokyo
University of Foreign Studies
PERRY, John 1996: “Persian in the Safavid Period: Sketch for an Etat de Langue”. In: Charles
MELVILLE (ed.): Safavid Persia. The History and Politics of an Islamic Society [Pembroke
Papers 4]. London/New York: Tauris, pp. 269-283
PIERCE, E. 1874: “A Description of the Mekranee-Beloochee Dialect”. In: Journal of the Royal
Asiatic Society (Bombay) 11, pp. 1-98
SABA DASHTYARI ( ˙Sab¯a Dašty¯ar¯ı) 1995: Bal¯oˇc¯ı zub¯an¯e ¯akubat I. Karachi: Jadg¯al ˇc¯ap u šing
––– 1998: Bal¯oˇc¯ı zub¯an¯e ¯akubat II. Karachi: Sayad Hashmi Reference Library
SOKOLOV, Sergej N. 1956: “Grammatiˇceskij oˇcerk jazyka beludžej Sovetskogo Sojuza”. In:
Trudy Instituta Jazykoznanija 6. Moscow: Akademija Nauk, pp. 57-91
SOKOLOVA, Valentina P. 1953: “Beludžskij jazyk”. In: Valentina SOKOLOVA: Oˇcerki po fonetike
iranskix jazykov. Moscow / Leningrad: Nauka, pp. 7-77
UTAS, Bo 2005: “Semitic in Iranian: Written, Read and Spoken Language”. In: Éva Ágnes
CSATÓ, Bo ISAKSSON, Carina JAHANI (eds.): Linguistic Convergence and Areal Diffusion.
London/New York: RoutledgeCurzon, pp. 65-77
––– 2006: “‘Genres’ in Persian Literature 900-1900”. In: Gunilla LINDBERG-WADA (ed.):
Literary History: Towards a Global Perspective, 2: Literary Genres: An Intercultural
Approach. Berlin/New York: Walter de Gruyter, pp. 199-241
WINDFUHR, Gernot 1979: Persian Grammar. History and State of its Study [Trends in
linguistics. State-of-the-Art Reports 12]. The Hague etc.: Mouton
YUSEFIAN (Y¯usef¯ı¯an, P¯akz¯ad) 2004: Tou ˙s¯ıf-e rade-šen¯axt¯ı-ye zab¯an-e bal¯uˇc¯ı. (Unpublished PhD
thesis, Isfahan University) 1383 h.š.
ZARUBIN, Ivan I. 1932: Beludžskie skazki I. Leningrad: Nauka
––– 1949: Beludžskie skazki II. Moscow/Leningrad: Nauka


Balochi and the Concept of North-Western Iranian

Prof Dr. Agnes Corne

By: Prof Dr. Agnes Korn

1. Introduction
The aim of this paper is to examine the position of Balochi among the Western Iranian languages by looking at certain features of its historical phonology. These features will be examined for their validity, and the Balochi data will be assessed to determine its relationship to the traditional notion of North-Western Iranian languages.

1.1 Balochi as a North-Western Iranian language

From a historical point of view, Balochi has been regarded as a so-called North- Western Iranian language. This means that with regard to certain linguistic features (called isoglosses), Balochi shares a set of characteristics with e.g. contemporary Kurdish and Zazaki and the Middle Iranian Parthian, whereas Persian, Tajiki and some other languages – called South-Western Iranian – show something different under the same circumstances. The table shows examples of such features from the field of the development of the consonants from prehistoric times to the languages of today. The Balochi data is compared with that of two other North-Western Iranian languages, viz. Zazaki (chosen for its strong North-Western traits), and Kurdish (chosen for its similarities with Balochi, for which see below), and the South-Western Iranian Persian.1

Proto-Iranian2 example

*´z “know”
*´s “iron”

North-Western Iranian

z: B z¯an-, Z zonen-, K zan-
B z¯am¯at, Z zama, K zava
s: B ¯asin, Z asın, K hesin

South-Western Iranian

d: NP d¯an-
NP d¯am¯ad
h: NP ¯ahan

Such relationships between languages have often been visualised in the form of a


East West

North South
Old Iranian languages
Avestan Median Old Persian

Middle Iranian languages
Sogdian, Saka etc. Parthian Middle Persian

New Iranian languages
Ossetic, Pashto etc. Zazaki, Balochi, New Persian etc.
Kurdish etc.
Figure 1: simplified family tree of Iranian languages3

1.2 Balochi as a southern North-Western Iranian language

It has long been observed, however, that the relationships between Iranian languages are more complex than the family-tree would suggest: In some cases, the Balochi outcome of some Proto-Iranian sounds or combinations of sounds is not the one we might expect in a North-Western Iranian language (judging from the Parthian evidence), but rather the South-Western one as in Persian. The same may be said for Kurdish, but not for Zazaki which usually shows the features expected in a NWIr. language. Examples are:

Proto-Iranian example

*\r “three”
*´s ˘ u “dog”

North-Western Iranian

hr: Z hire
sp: South Z espe

South-Western Iranian

s: B sai, K sê, NP se
s: B sag, K seg, NP sag

The word for “three” is the only word with PIr. *\r4 which is found in all the fourlanguages, but it is not a good example, since the Balochi and Kurdish words for “three” are perhaps borrowed from Persian (PAUL 1998:1668). Better examples which show that Balochi has s for PIr. *\r are: ¯as “fire” (from *¯a\r-, obl. of *¯atar-, NP ¯azar), dialectal pis “father”, m¯as “mother”, br¯as “brother” (< obl. *pi\r- etc., GEIGER 1891:430, MORGENSTIERNE 1948:257), duks¯ıˇc/dusk¯ıˇc “sister-in-law” ( B s: s¯ah “breath, life” which (GEIGER 1891:430f.) might belong to OInd. ´sv¯asá-, unless it was (together with Pashto s¯ah, MORGENSTIERNE 1927:66) borrowed from some Indic language; s¯ıy- / s¯ıt7 “swell” is a cognate of OInd. ´sváya- according to GEIGER 1891:430 and BAILEY 1979:476a; šiš8 “louse” (NP sopoš, šepeš, Av. spiš) might go back to *´s ˘u- (via *siš, cf. Wakhi šiš, MORGENSTIERNE 1927:69, STEBLIN-KAMENSKIJ 1999:30, 330). Conversely, there is no good evidence for PIr. *´s ˘ u > B sp since the only examples which might be adduced here (B (i)sp¯et “white”9 and asp “horse”, may be Persian loanwords. Thus, one might think that a model which ranks the New Iranian languages according to their specific number of differences from Persian (i.e. the number of shared characteristics with Parthian) is more adequate for the description of relationships among the Western Iranian languages than a family-tree. Such a “scale of northernness” has been suggested by PAUL 1998:170, from which one may conclude that Balochi (and Kurdish even more) is a comparatively “southern” North-Western Iranian language:
PIE PIr. Pth. Gor. A”z. Zaz. Tal. Semn. Casp. CD Bal. Kd. Pers.
*´k/*´g *´s/´z10 s/z s/z s/z s/z s/z s/z s/z s/z s/z s/z h/d
*k( ˘u) pal *ˇc -ž- -ž- -ž- -ˇj- -ž- ˇj,ž -ˇj- ˇj,ž,z -ˇc- -ž- -z-
*g( ˘u)(h) pal *ˇj ž ž ž (y-) ˇj ž ˇj,ž ˇj (z-) ˇj,ž,z ˇj -ž- z
*´k ˘u *´s ˘ u 11 ? sip isb esp asb esp s esb s? s s
*tr/*tl *\r12 hr (ya)r (h)r (h¯ı)r h(*r) (h)r r r s s s
˘u *d ˘ u b b b b b b b b b(?) d d
(OIr.) *rd/*rz *rd/*r´z r(d)/rz ł,r/rz r/rz ¯r/rz r/rz l/l(rz) l/l l/l(rz) rd/rz ł/ł l/l
*s ˘u- *h ˘u- wx w h w h x(u) x(u) x(u), f w x(w) x(u)
*t ˘u *\ ˘u f u u w h h h h(u) h? h h
* ˘i- * ˘i- y- y- y- ˇj- ˇj- ˇj- ˇj- ˇj-(y-) ˇj- ˇj- ˇj-
Table 1: Western Iranian isoglosses (modified from PAUL 1998:170)
– PIr. intervocalic *ˇc (TEDESCO 1921 no. 4): The Balochi product is not ˇi as PAUL 1998:170 assumes, but ˇc (GEIGER 1891:423), e.g. PIr. * ˘uaiˇc- “sieve” > B g¯eˇc- (Z viˇi-, K bêžing (noun), NP b¯ız-); PIr. *ra ˘uˇcah- “light” > B r¯oˇc “day” (Z roˇi/rodz/roz/rož, K rož, NP r¯uz).14 – PIr. *ˇi (TEDESCO 1921 no. 3): e.g. PIr. *ˇian- “woman” > Z ˇiınêk, B ˇian, K žın, NP zan. – PIr. *d ˘ u (TEDESCO 1921 no. 12): Judging from do “two” and dar “door”,15 it has generally been assumed (thus also PAUL 1998:170) that Balochi shows PIr. *d ˘ u > d as do Persian and Kurdish (Zazaki has b: kê-ber/ˇcê-ver “(house-)door”). However, both words can also be explained as Persian loanwords. In this case, ipt¯ı/pit¯ı “other” B zird (Z ze¯rî, NP del) “heart”, PIr. *b ˙r´z- > B burz (Z berz, NP boland) “high”.16 – PIr. *h ˘u- (TEDESCO 1921 no. 17) gives B w (PAUL 1998:170 has v), e.g. war- “eat” (Z wer-, K xwar-, NP xwor-) or h, e.g. h¯ed “sweat” (K xwêh, NP xwei). – PIr. *\ ˘ u (TEDESCO 1921 no. 11): It is not clear if the outcome in Balochi, Kurdish and some other languages is h as PAUL 1998:170 assumes since the only examples are the numerals “four” and “forty” which are identical with (Early New) Pers. ˇcah¯ar/ˇc¯ar, ˇcihil/ˇcil and thus could be loans. Zazaki preserves w in ˇcewres/tsewres “fourty”.

– PIr. * ˘i- (TEDESCO 1921 no. 5): e.g. NP ˇiod¯a “separate”, Z ˇiıya, B ˇiit¯a, K ˇiihê.

2. Significance of the isoglosses

One might therefore ask if it is adequate to call Balochi a North-Western Iranian language at all, or rather, in which respect Balochi is North-Western Iranian. To answer this question, I will examine the isoglosses once more, focussing on two points which have not been taken into account so much in previous discussions about the Western Iranian isoglosses: the chronology of the sound changes in question and the distinction between archaism and innovation.

2.1 Chronology

Some of the sound changes producing the characteristic differences between South- Western and North-Western Iranian languages date back to Old Iranian times, i.e. they already distinguished Old Persian (South-Western Iranian) from Avestan (Non-South- Western Iranian). Other sound changes only came about in Middle Iranian times, i.e. they can be seen in Middle Persian and Parthian, but not in Old Persian and Avestan. It is important to consider the date of a feature since a common characteristic of e.g. Balochi and Middle Persian might tell us something different than one which Balochi shares already with Old Persian.
The approach presented in table 1 above suggests that in the beginning, all so-called North-Western Iranian languages showed some common traits and then one by one came under the influence of Persian, with Kurdish and Balochi being the first affected and Zazaki among the last. It will be seen, however, that this cannot be the case.

Table 2 shows the sound changes (those mentioned above as well as others often cited in works dealing with the grouping of WIr. languages) arranged according to their date of appearance from the oldest (top of the table) to the most recent ones (bottom).17

Proto-Iranian Old Iranian:

Avestan (Non-SW) Old Persian (SW)
*´s s \
(> Middle Persian h)
*´z z d
*pas-ˇca “behind” pas-ˇca pas¯a
*´s ˘ u sp s
*\r \r ç
(> Parthian hr) (> Middle Persian s)
Middle Iranian:
Parthian (NW) Middle Persian (SW)
*ˇc /V_V ž (OP ˇc >) z
*ˇI ž (OP ˇi > ) z
* ˘i- y (OP y >) ˇi
*r´ z rz (OP rd > ) l
*rd rd (OP rd > ) l
*d ˘ u b (OP duv >) d
*h ˘u- wx (OP uv >) xw
*\ ˘ u f (OP \uv >) h
*g, d /V_V g, d (OP g, d >) y
* ˘u- w (OP w >) w, g
*m /V_V m, w (OP m >) m
New Iranian:
Z, B, K New Persian
*šm18 m (> K v) (OP šm >) šm

Table 2: isoglosses in chronological order (bold letters: innovations)

– PIr. *pas-ˇca is no. 21 in TEDESCO 1921.
– PIr. *ˇi: It is not excluded that the OP signs ˇia, ˇii already stood for ža, ži (HOFFMANN
– PIr. *g, d (TEDESCO 1921 no. 8 and 6 respectively) disappear between vowels in Zazaki
and Kurdish, e.g. pê “foot” vs. B p¯ad, NP p¯a(y).
– PIr. * ˘u- (TEDESCO 1921 no. 18): e.g. B giˇcin- / NP guz¯ın- “collect”, Z vin-en-, vên-en- / B gind- / K bîn- / NP b¯ın- “see” (pres.), Z va / B gw¯at / K ba / NP b¯ad “wind”.
– PIr. *m: TEDESCO 1921 no. 19.
From these findings, we may conclude that table 1 is a synchronic one, summarizing the data from contemporary Western Iranian languages without reference to their history, and that the “scale of northernness” cannot be taken to hint at historical developments. Contrary to their position in table 1, e.g. the isogloss NWIr. ˇi, ž etc. vs.
SWIr. z (from PIr. *-ˇc, ˇi) is of a more recent date than SWIr. (OP) ç, later s vs. Av. \r, Pth. hr (PIr. *\r), and Persian ˇi- from PIr. * ˘i- cannot be a young phenomenon, but conversely, has to be even older than the voicing of the intervocalic stops as may be seen from Persian loanwords in Armenian, e.g. ˇiatuk “sorcerer” (HÜBSCHMANN 1897:232, cf. NP ˇi¯ad¯u, OInd. y¯atú- “sorcery”).

2.2 Archaism versus innovation

The next interesting observation is that in the majority of the cases, the (Old and Middle) Persian outcome of a certain PIr. sound (or sound combination) can be interpreted as an innovation (printed in bold letters in table 2) whereas Avestan and Parthian preserve the PIr. state of affairs. With regard to the features studied here, the North-Western Iranian languages show much fewer innovations. These are: Pth. b from *d ˘ u (where Persian has an innovation in the other direction, i.e. d), Pth. f from *\ ˘ u (Pers. h), the development of PIr. *h ˘u- to Pth. wx19 (MP xw), occasional w for postvocalic m and the reduction of šm to m. 19 Written wx- might be interpreted as the result of a metathesis *hw- > wh- or as a “device (…) to represent a new sound, viz. devoiced [ ° w]”, cf. English wh- from Old English hw- (MACKENZIE 1967:2629). In the latter case, the Pth. result would be already quite near to that of Zazaki, Balochi


Most of the NWIr. characteristics of contemporary languages are of the type “shared archaism”, i.e. these languages preserve the development already shown by Parthian, e.g. B, Z, K z from PIr. *´z (see 1.1). Shared archaisms tell us that from a historical point of view, Balochi, Zazaki etc. belong to the group of North-Western Iranian languages (i.e. come from a common ancestor). Therefore, this does not tell us anything about the history of a given language after the time the traits in question came about.

Independent innovations

The isoglosses which may be regarded as innovations will be examined next, taking into account data from Zazaki and Kurdish again. Common innovations may come about either independently or through contact of the languages in question. For each feature, it has to be decided into which category it most likely belongs. There are four sound changes which are likely to be independent innovations.
One change which is extremely common in languages all over the world is that of word-initial * ˘i- > ˇi which also occurs e.g. in Urdu, several Turkic languages and Low German. Moreover, this change is quite an ancient one in Persian (it must have happened in an old stage of Middle Persian, see p. 55). It is not likely that at that time,
Persian was in contact with the predecessors of all the Western Iranian languages which show ˇi from * ˘i- (even including Talyshi and Zazaki, see table 1), so this change is probably not the result of language contact. The weakening of intervocalic Old Ir. b, g and d to the corresponding fricatives b, g and d is common to Parthian and (some early stage of) Middle Persian (SUNDERMANN 1989:108).20 The further weakening of g and d to y (Persian) or zero (Zazaki, Kurdish) has probably occurred independently: the loss of postvocalic consonants, especially postvocalic voiced fricatives, is such a common phenomenon among the languages of the world that it seems safer to assume (also taking into account the peripheral position of Zazaki) that it has occurred independently in all three languages.
20 This sound change “is common throughout Iranian” (SIMS-WILLIAMS 1996:650). Note that if one considers the change of intervocalic b, d, g to fricatives as shared by all (Western) Middle Iranian languages, this implies that in some stage of Balochi, b, d, g must have reemerged from Middle Ir. b, d, g. In this case, the common assumption that the Balochi stops preserve the Old Iranian state of affairs would be subject to modification.
Something similar applies to the development of postvocalic m to w. This appears to be a Middle Iranian phenomenon since it has been claimed to be an optional development in Parthian already (TEDESCO 1921:208). On the other hand, this change seems to be a rather recent phenomenon in Kurdish and operates as a sound law for all instances of postvocalic m (e.g. nav “name”), including modern loanwords, e.g. ˇiiv¯at (Arabic ˇiam¯a‘at) “assembly” (cf. MACKENZIE 1961:70), and also including the outcome of the reduction šm to m (e.g. çav “eye” (cf. Z çım, B ˇcam(m)) vs. NP ˇcešm). As there is no trace of šm > m in Parthian yet, it seems that Parthian instances of m > w can have nothing to do with the Kurdish sound law and require some other explanation.21 In Balochi, the instances of w for m seem to be the result of a still less ancient process: m > w occurs in the Eastern dialects only, and only sporadically even there, and the Eastern dialects show a number of (quite recent, it seems) sound changes separating them from the other dialects.22 As the change of m > w is a very common one in the languages of the world (cf. e.g. Hungarian név “name” < Proto-Uralic *nime, cf. Finnish nimi), it is quite probable that it has come about independently in several Iranian languages. In Balochi, it might even have happened under the influence of Indic
The reduction of šm to m would represent the only common innovation of North- Western Iranian languages from post-Parthian times. However, it is also shared by a number of other Iranian languages as far away from each other as SW Iran (Bashkardi) and the Afghanistan-Pakistan border (Ormuri, an Eastern Iranian language). Since a phenomenon of language contact between all these is most unlikely for post-Middle Iranian times, it is safer to regard this change as an independent development, too. As in the case of common archaisms, sound changes which are the result of independent innovations do not tell us anything about the question of how the languages are historically related to one another, and which languages have been in contact with one another. Such sound changes are merely due to chance. They should therefore be deleted from the list of isoglosses. 21 In fact, the Parthian examples do not show the presumed change of m > w: angawan (otherwise anˇiaman) “assembly” does not exist, and abg¯aw¯ah belongs to abgaw- “increase” (Desmond Durkin- Meisterernst, personal communication). Parthian h¯aws¯ar “equal, similar” does not contain a variant of h¯am° “same”, but *h¯awat- “similar”, cf. MP h¯awand “like, similar” (SIMS-WILLIAMS 1998a:85). 22 The sound system of the Western and Southern dialects is very similar to that of Common Balochi (the ancestor language of the Balochi dialects), cf. GEIGER 1891:403. 23 Cf. Hindi/Urdu g㯠v “village” from Old Ind. gra¯ma- (HINÜBER 1986:107).

Innovations which Balochi shares with Persian

All features discussed above are of the type where Persian shows some innovativedevelopment. This also includes those features which Balochi shares with Persian, the ancient developments *´s ˘ u and *\r > s (if this is indeed the Balochi outcome, see 1.2 above) as well as the younger ones *\ ˘ u > h (if so, see 1.2) and * ˘u- > g(w)/b. Kurdish, in addition, shares with Persian the developments of *d ˘ u > d, *h ˘u- > xw, and perhaps also *r´z, *rd > l. Zazaki shows the “true” NWIr. development in all these cases.24 The characteristics that Balochi and other North-Western Iranian languages share with Persian may be attributed to the profound influence which Persian has exercised on the neighbouring languages all over the past centuries and already in Old Persian times.
Those languages which have been in contact with Persian show a number of similarities with Persian whereas Zazaki which is spoken farther off to the North-West has not been influenced by Persian very much. For the cases mentioned above, we may thus conclude that Balochi has taken over some innovations of Persian (and Kurdish even more).

Innovations which Balochi shares with Parthian

In addition, the evidence suggests that Balochi – in contrast to Kurdish – probably shares two innovations with Parthian: the development of b from *d ˘ u and of w/h from *h ˘u-. This might indicate that in Middle Iranian times, the ancestor of present-day Balochi has still been in contact with Parthian, but the ancestor of present-day Kurdish is likely not to have belonged within this sphere.

2.3 Summary of Western Iranian isoglosses

The table of potentially significant sound changes in WIr. languages thus looks as follows (the single line separates the NWIr. development to the left from the SWIr. One to the right): 24 Zaz. ˇi- < * ˘i- is probably not due to contact with Persian, but rather an independent innovation (see above).

PIr. Av. Parthian Zazaki Balochi Kurdish NP MP OP
*´s s s s s s h h \
*´z z z z z z d d d
*pas-ˇca “behind” pas-ˇca paš ? paš paš pas pas pas¯a
*´s ˘ u sp sp sp s(?) s(?) s s s
*\r \r hr hr s s s s ç
*ˇc /V_V ˇc ž ˇi ˇc ž z z ˇc
*ˇi ˇi ž ˇi ˇi ž z z ˇi
*r´z rz rz rz rz ? l l rd
*rd rd rd ¯r rd ? l l rd
*d ˘ u duu b b b(?) d d d duv
*h ˘u- xw wx w w,h xw xw xw uv
*\ ˘ u \b f w h? h? h h \uv
* ˘u- v w v g(w) b g, b g, w g, d
Table 3: Western Iranian isoglosses in Old, Middle and New Iranian times (in bold: innovations)

3. Conclusion

The findings from some characteristic sound changes thus show that the relationship among Western Iranian languages can neither be adequately represented by a family tree alone nor in the form of a “scale of northernness”.
The fact that some of the innovations which have occurred in the various stages of Persian have been taken over by Balochi (and some more by Kurdish) indicates that Balochi and Kurdish have been influenced by Persian since Old Iranian times. A small number of innovations that have occurred in Parthian are probably shared by Balochi, but not by Kurdish. This means that in contrast to Kurdish, Balochi might still have been in contact with other North-Western Iranian languages in Middle Iranian times.
This would mean that Balochi has been a North-Western Iranian language not only at the outset but remained so in Middle Iranian times.


Az. Azari OInd. Old Indic
B(al.) Balochi OP Old Persian
Casp. Caspian dialects Pers. Persian
CD Central dialects PIE Proto-Indo-European
Gor. Gorani PIr. Proto-Iranian
Ir. Iranian Pth. Parthian
K(d.) Kurdish Semn. Semnani
MP Middle Persian SW South-Western
NP New Persian Tal. Talyshi
NW North-Western W Western
obl. Oblique case Z(az.) Zazaki


1. A list of abbreviations is at the end of the article. For better comparison, the method of writing Zazaki and Kurdish consonants used here follows the tradition of Iranian studies and not the orthography used in Zazaki and Kurdish publications (e.g. š for ¸s). “Kurdish” as used here denotes Kurmanji.
2. Proto-Iranian (a language reconstructed by linguists) is the language from which all Iranian languages are thought to be descended in a way similar to members of a family from a common ancestor. The isoglosses shown in this table are no. 1 and 2 in Tedesco’s classical work on the internal grouping of the Western Iranian languages (TEDESCO 1921).
3. Avestan is grouped under Eastern Iranian here, although in Old Iranian times, the distinction was rather one between South-Western (i.e. mainly Persian) and the rest of the Iranian languages, so Avestan should more correctly be termed a “Non-South-Western” Iranian language (cf. also SIMSWILLIAMS1996:649f.).
4. This is isogloss no. 10 in TEDESCO 1921.
5. For Kurdish s from *\r, see MACKENZIE 1961:76f.
6. This isogloss is not found in TEDESCO 1921, but cf. e.g. PAUL 1998:166ff.
7. This word is only found in DAMES 1881, HITTU RAM 1881 and ELFENBEIN 1990/II.
8. šiš is only reported by ELFENBEIN 1989:635.
9. Balochi (i)sp¯et might have been borrowed from MP sp¯ed or Early NP saf¯ed. Devoicing of word-final consonants is comparatively common in loanwords in Balochi, cf. KORN 2001:319.
10. For references and examples, see above.
11. For references and examples (and s as the possible outcome in Balochi), see above. It should be kept in mind that the Kurdish outcome is far from certain. The Parthian outcome seems to be sp (as one would expect), cf. asp “horse”, isp¯ed “white”.
12. For references and examples, see above.
13. The entries for Balochi have been modified (see the notes below the table), a column for Proto- Iranian has been added; the symbols used for writing Proto-Indoeuropean (the ancestor language of the Iranian as well as several other branches of languages) have been adjusted to the system used here, otherwise the table is the one shown in PAUL 1998:170.
14. Parthian loanwords in Armenian and Georgian show that the development of intervocalic ˇc > ˇi > ž occurs within Parthian (i.e. early Parthian preserves -ˇc-) and that PIr. *ˇi gives ž in all stages of Parthian (GIPPERT 2000:2). The products of both thus merge in Middle Iranian times.
15. B dw¯azdah “twelve” and digar (etc.) “other” are Persian borrowings in any case as is shown by the -h of the former and the -g- of the latter.
16.According to MACKENZIE 1961:78, it is “not (…) possible to be certain which [i.e., l/ł or the preservation of rd, rz] is the true Kurdish development”.
17. The order in each section of the table hints at assumptions (not to be detailed here) concerning the relative age of the developments, but much of that remains speculative.
18. On the change of PIr. *šm, see below.
19. Written wx- might be interpreted as the result of a metathesis *hw- > wh- or as a “device (…) to represent a new sound, viz. devoiced [ ° w]”, cf. English wh- from Old English hw- (MACKENZIE 1967:2629). In the latter case, the Pth. result would be already quite near to that of Zazaki, Balochi
20. This sound change “is common throughout Iranian” (SIMS-WILLIAMS 1996:650). Note that if one considers the change of intervocalic b, d, g to fricatives as shared by all (Western) Middle Iranian languages, this implies that in some stage of Balochi, b, d, g must have reemerged from Middle Ir. b, d, g. In this case, the common assumption that the Balochi stops preserve the Old Iranian state of affairs would be subject to modification.
21. In fact, the Parthian examples do not show the presumed change of m > w: angawan (otherwise anˇiaman) “assembly” does not exist, and abg¯aw¯ah belongs to abgaw- “increase” (Desmond Durkin- Meisterernst, personal communication). Parthian h¯aws¯ar “equal, similar” does not contain a variant of h¯am° “same”, but *h¯awat- “similar”, cf. MP h¯awand “like, similar” (SIMS-WILLIAMS 1998a:85).
22. The sound system of the Western and Southern dialects is very similar to that of Common Balochi (the ancestor language of the Balochi dialects), cf. GEIGER 1891:403.
23. Cf. Hindi/Urdu g㯠v “village” from Old Ind. gra¯ma- (HINÜBER 1986:107).
24. Zaz. ˇi- < * ˘i- is probably not due to contact with Persian, but rather an independent innovation (see above).


BAILEY, Harold W. 1979: Dictionary of Khotan Saka. Cambridge.
DAMES, Longworth M. 1881: A Sketch of the Northern Balochi Language, containing a grammar, vocabulary and specimens of the language (Extra number of the Journal of the Asiatic Society of Bengal 1/1880). Calcutta.
ELFENBEIN, Josef 1989: “Baluchistan III: Baluchi language and literature”, in: Encyclopaedia Iranica 3, pp. 633-644.
––– 1990: An Anthology of Classical and Modern Balochi Literature. Wiesbaden (2 vols.).
GEIGER, Wilhelm 1891: “Lautlehre des Bal¯uˇc¯ı”, in: Abhandlungen der I. Classe der Bayrischen Akademie der Wissenschaften XIX/II, pp. 397-464.
GIPPERT, Jost 2000: “The historical position of Zaza revisited” (handout of a paper read at the First Workshop of Kurdish Linguistics, Kiel, 12th-14th May 2000).
HOFFMANN, Karl 1976: “Zur altpersischen Schrift”, in: Aufsätze zur Indoiranistik II. Wiesbaden, pp. 620-645.
von HINÜBER, Oskar 1986: Das ältere Mittelindisch im Überblick. Vienna.
HITTU RAM 1881: Biluchi Nameh, a Text Book of the Biluchi Language. Lahore.
HÜBSCHMANN, Heinrich 1897: Armenische Grammatik I: Armenische Etymologie. Leipzig.
KORN, Agnes 2001: “Archaismus und Innovation im Verbalsystem des Bal¯oˇc¯ı” (paper read at the Symposium Iran 2000, Bamberg), bamberg.pdf.
MACKENZIE, D. Neil 1961: “The Origins of Kurdish”, in: Transactions of the Philological Society, pp. 68-86 (= Iranica Diversa II, pp. 369-387).
––– 1967: “Notes on the Transcription of Pahlavi”, in: Bulletin of the School of Oriental and African Studies 30, pp. 17-29 (= Iranica Diversa I, pp. 35-47).
Balochi and the Concept of North-Western Iranian 61 MORGENSTIERNE, Georg 1927: An Etymological Vocabulary of Pashto. Oslo.
––– 1932: “Notes on Balochi Etymology”, in: Norsk Tidsskrift for Sprogvidenskap 5, pp. 37-53 (= Irano-Dardica, pp. 148-164). ––– 1948: “Balochi Miscellanea”, in: Acta Orientalia 20, pp. 253-292.
PAUL, Ludwig 1998: “The Position of Zazaki among West Iranian Languages”, in: SIMSWILLIAMS, pp. 163-177.
SCHMITT, Rüdiger (ed.) 1989: Compendium Linguarum Iranicarum. Wiesbaden.
SIMS-WILLIAMS, Nicholas 1996: “Eastern Iranian languages”, in: Encyclopaedia Iranica 7, pp. 649-652.
––– (ed.) 1998: Proceedings of the Third European Conference of Iranian Studies held in Cambridge, 11th to 15th September 1995. Part I: Old and Middle Iranian Studies. Wiesbaden.
––– 1998a: “Further notes on the inscription of Rabatak, with an Appendix on the names of Kujula Kadphises and Vima Taktu in Chinese (pl. 9-12)”, in: SIMS-WILLIAMS, pp. 79-92.
STEBLIN-KAMENSKIJ, Ivan M. 1999: E˙timologicˇeskij slovar’ vaxanskogo jazyka. St. Petersburg.
SUNDERMANN, Werner 1989: “Westmitteliranische Sprachen”, in: SCHMITT, pp. 106-113.
TEDESCO, Paul 1921: “Dialektologie der mitteliranischen Turfantexte”, in: Monde Oriental 15, pp. 184-258.

offprint from

Carina JAHANI, Agnes KORN (eds.) 2003: The Baloch and Their Neighbours. Ethnic and Linguistic Contact in Balochistan in Historical and Modern Times. Wiesbaden: Reichert, pp. 49-60


Few examples of Balochi and Brahui Morphological Similarities

Prof Dr. Abdul Razzaq Sabir

Prof Dr .Abdul Razzaq Sabir


Balochi a language from Northwestern Iranian language group of Indo European languages and Brahui a Proto Dravidian language and member of the northern group of Dravidian languages, spoken in a common region, despite having relation with different language groups have various linguistic similarities. The verbal system, like that of the Iranian languages, is based on the two stems present and past. Present stems are based on the imperative, present indicative, present subjunctive, agent nouns and present participle. In both languages the past stems are used in the preterit indicative, in the compound tenses, such as past indicative, past subjunctive, past perfect, perfect participle, pluperfect and infinitives.


In Balochi the infinitive is to be made from (imp+ag): kan+ag=kanag (to do), war+ag=warag (to eat), jan+ag=janag (to beat), di+ag=diag (to give). On the same pattern the Brahui infinitive is also made from (imp+ing) i.e. at+ing=ating (to bring), khall+ing=khalling (to beat), paach+ing=paaching (to skin), taf+ing=tafing (to tie), chikk+ing=chikking (to catch) etc1.


One of the most characteristic features of Brahui and Balochi languages is the formation of compound verbs i.e. nouns, adjectives and adverbs plus colorless verbs like kanag/kanning (to do), daiag/tinning (to give), booag/manning (to be come), janag/khalling (to beat).The following are examples of the compound verbs:
Brahui Balochi Meaning
kanning/kanag(to do)
Tawaar kanning tawaar kanag (to call)
Bahaa kaning Bahaa kanag (to sell)
jor kanning jor kanag (to make)
Hit kanning habar kanag (to speak)
Manning/beyag(to be come)
deer manning aap beyag (to became water)
khurt manning hurt beyag (to become grind)
Tinning/daig(to give)
dikka tinning dikka deag (to push)
hail tinning hail deag (to learn)
Doo tinning dast deag (to shake hand, to stop)
mon tinning dem deag (to send)
Du khalling dast janag (to touch)
Chakk khalling chakk janag (to turn and see back)
Sar khalling sar janag (to search dissolute)
Kan khalling Cham janag (to inform one by eye)
agent nouns
Agent nouns in both languages are formed by adding of /ok/ to the
present stem, e.g.
Brahui Balochi Meaning
Karok Kanok (doer)
Kunok Warok (eater)
Pulok Pulok (catcher)
Khalok Janok (beater)


Adjectives in Brahui commonly take suffix of /un/ and Balochi takes /en/ and precede the noun. Some common adjectives in Balochi are draajen (long), mazanan (big), sohren (red), kohnen (old), noken (new) while the main common adjectives in Brahui with the suffix of /un/ i.e. balloon (big), murgun (long), paalun (wet), peeun (white), kharrun (green), baasun (hot). In Brahui an adding of /ingaa/ is used after adjectives i.e. baasun /baasningaa/, paalun /paaluninga/ etc2.


The attributive adjective mazan in Balochi means (big) and for the abstract noun the prefix of mazan is commonly used in all the dialects of Balochi. The same morphological construction is used in Brahui with a minor phonetic change by deleting the ending consonant of /n/, and remaining maza is used for the purpose before the noun for example3:
Balochi Brahui Meaning
mazan ponz. maza baamus. (the person having a big nose)
mazan pad. maza paacha. (the person having big legs)
mazan sar. maza sara. (the person having a big head)
mazan dil. maza ust. (the person having big heart)Brave
mazan shaan. maza shaan. (the person having big dignity)


The Baloch and Brahui both languages distinguish two numbers singular and plural by the endings in Balochi /aan/ like:
gis (house) gisaan (houses)
chuk (child) chukaan (children)
kitab (book) kitabaan (books)
In Brahui the main number ending is /aak/ like:
uraa (house) uraak (houses)
chunaa (child) chunaak (children)
kitaab (book) kitaabaak (books)

Apart from the common main number ending in Brahui, there are two other ways are:
a)kaaffi:- In this formation the number ending is only /k/ like, doo (hand), dook (hands), khan (eye), khank (eyes), khaf (ear), khafk (ears).
b)gaaki:- The singular words ending with /a/ have /gaak/ added to them in order to make plural e.g, bala (grandmother), balaghaak (grandmothers), ghala (wheat), ghalaghaak (wheat, many), doosha (snake), dooshaghak (snakes) etc.


Balochi and Brahui both have no distinction of grammatical gender, in case of Brahui, all other Dravidian languages, except Toda and Brahui have kept the old gender system. About this construction in Brahui M.B. Emenue says that “this loss of gender system in Brahui is to be ascribed to Balochi influence on Brahui” 4. Both languages have common use of different words to distinguish between male and female.
piruk (grandfather), baluk (grandmother)
pis (father), maas (mother)
Braas (brother), gwaar (sister)
bachak (boy), jinik (girl or daughter)
Pira (grandfather), balla (grandmother)
Ilum (brother), ir (sister)
baava (father), lumma (mother)
maar (son), masir (girl or daughter)

Some male and female are distinguished by the use of additional words, such as, nar (male) and maadag (female) in Balochi, and naringaa (male) and maadaingaa (female) in Brahui such as:
nar shinik (male lamb), maadagen shinik (female lamb)
nar mazaar (lion), maadagen mazaar (lioness)
Nar tolag (male jackal), maadagen tolag (female jackal)
Nar maar (male snake), maadagen maar (female snake)

naringaa dusha (male snake), maadaingaa dusha (female snake)
naringaa sor (male lamb), maadaingaa sor (female lamb)
naringaa khakho (male crow), maadaingaa khaakho (female crow)
naringaa sher (lion), maadaingaa sher (lioness)


Baloch and Brahui both have three case system, direct, genitive, oblique, both in singular and plural, with the following case endings.
Direct Genitive Oblique
Singular unmarked -na e
Plural unmarked -ta te
Singular unmarked -ay a
Plural unmarked -i a


The case suffix /a/ in common in both languages. J. Elfenben a eminent linguist points out an other this case suffix of -a common in Balochi and Brahui languages. Balochi and Brahui both have a case suffix of /a/ and some times in Brahui /ga/ after a noun e.g.5
Balochi: man mastunga rain (i go to Mastung)
man gisa rain (I go to home)
Brahui: Ee mastungaa kaaava (i go to Mastung)
Ee uraghaa kaava (i go to home)


The echo words in both languages are formed either by changing the initial consonant of the word into the consonant /m/ or by adding /m/ to a word beginning with a vowel e.g. 6
Naan (bread) maan
Log (house or home) mog
Aap (water) map
Chuk (child) muk
Kaagad (paper) maagad
Iragh (bread) miragh
Uraa (house) muraa
Deer (water) meer
Chunaa (child) munaa
Kaaghaz (paper) maaghaz

In the other Iranian languages like Pashto has the same construction, and echo sign in Pashto also has the suffix of /m/ like:
Dodai (bread) modai
Kor (house) mor
Uba (water) Muba

In Balochi and Brahui the initial consonant of the followed by /m/ do not change:
Brahui Balochi
mom, mom (wax) mages, magas (fly)
malakh, malakh (locust) malakh, malakh (locust)
maee, maee (buffalo) maee, maee (buffalo)

Balochi and Brahui both languages some times have echo words in the sense of plural also.
Brahui Balochi
uraa, muraa gis, mis (some houses)
ulee, mulee usp, masp (some horses)
deer, meer aap, map (some water)
chunaa, munaa chuk, muk (some children)


The most common vocative interjections in Balochi and Brahui are e, o, eh, ya, oh and also va, eh, are, urc, pah, toba, ah etc.

USE OF kah.

In Dravidian languages neither the word ki nor the construction is found. It is said that Brahui has borrowed this either from Indo-Aryan or Iranian languages. According to J. Elfenbein “Iranian is by far most likely source for it in Brahui, since its use in the Indo-Aryan languages most likely to have influenced Brahui is much too restricted to account for the large variety of different functions it possesses in Brahui”7.
In Balochi and Brahui both “ki” (if) is used in the following forms:
a)In the meaning of /if/ likely
Brahui: ee ki makhaat oh hum makhaar.
Balochi: man ki handitun a hum handitant
(when I laughed they also laughed)
b)”ki nava”
Brahui: huris ki navaa tamos
Balochi: chon ma bi ki bikapai
(mind your steps)
c)taanki (unless/until) The both languages have borrowed this construction from Persian “ta an ki”.
Brahui: taanki o batane he inpara.(until he/she doesn’t come I will not go)


Other than first three numbers in Brahui asit (one), irut (two) and musit (three) all other numerals in both languages are same. The musit (three) has also become from si (three) of Persian. The numbers from ten to twenty are formed by the appropriate unit da (ten) with certain common phonetic changes:
yaanzda (eleven) dwaanzda (twelve)
Senzda (thirteen) Chaanrda (fourteen)
paanzda (fifteen)
In other instance, in both languages the addition of a unit with tens, hundreds, thousands is brought by means of the suffixial conjunction /o/ , with the large number placed in the first position followed by smaller number e.g.

Balochi Brahui
beest o hapt beest o haft (twenty seven)
chil o char chil o char (forty four)
shast o panch shast o panch (sixty five)
haptaad o sai haftad o sai (seventy three)
chaar sad o navad Chaar sad o navad (four hundred and ninety)
The ordinal numerals are formed from cardinals with the adding of the suffix /mi/ in Balochi and /miko/ in Brahui e.g.8
Brahui Balochi
Chaarmiko Chaarmi (4th)
Beestmiko Beestmi (20th)
See o shash miko See o shash mi (36th)
Chil o do miko chil o do dumi (42nd)
shast o panch miko Shat o panch mi (65th)


Some of the conjunctions used in Balochi and Brahui are same and some of them are of Persian and Arabic origin. Important Brahui/Balochi common conjunctions are under:
Brahui Balochi
Maga Maga (but)
Hum Hum (too, also)
Nai Na (neither…nor…)
wakhtas ki Wahde ki (when)
Aga Aga (if)
Gwaraa Gwaraa (with, near)
Padaa Padaa (behind)
Some other miscellaneous common conjunction in both languages are /ki/ (that) /o/ (and) ,/ya/ (or) etc.


Balochi and Brahui both languages have same sentence structure: minor sentences, major sentences, nominal, verbal and interrogative and compound sentences are to be seen with common structure.


1. Sabir, Abdul Razzak “Some Morphological and structural similarities of Brahui and Balochi languages” Proceedings of the International Symposium on “Linguistic contacts in Balochistan ancient and modern time” published by Department of Iranian Studies, Uppsala University Sweden.2004.pp-151-60.
2. –do—
3. –do—
4. Emeneau, M.B, “languages and linguistic area” edited by Anwar S. Dil essays Murrey, M.B Emeneau, Stanford University Press Stanford California 1980.Page 319.

5. Elfenbein. J “Notes on the Balochi Brahui linguistic commonality” Phiologica Society, Council 1981-82, Oxford pp-77-99. Page 85.

6. Sabir, Abdul Razzak ”Morphological similarities in Brahui and Balochi languages” IJDL, ISDL, Therivenanthapuram, Kerala S.India 1995.

7. Elfenbein. J “Notes on the Balochi Brahui linguistic commonality” Phiologica Society, Council 1981-82, Oxford pp-77-99.

8. Sabir, Abdul Razzaq” Balochi aur Brahui zubanoon ki rawabit” Ph.D dissertation submitted to University of Balochistan, 1994 p.218.


Marking of arguments in Balochi ergative and mixed constructions

Prof. Dr. Agnes Corne

By: Dr. Agnes Korn

1. Introduction

Balochi (Bal.) is a contemporary language of the Iranian (Ir.) branch of Indo-European languages and is spoken in Western Pakistan, South Western Afghanistan, South Eastern Iran and some other countries by several millions of people. Its dialects may be divided into a Western (WBal.), a Southern (SBal.) and an Eastern (EBal.) group.1 While the majority of Balochi dialects pattern ergatively in the PAST domain (see 1.1), many sentences show somewhat deviant constructions. These patterns are interesting from a typological point of view, specifically in their combination in one and the same language; they are the topic of this paper. The approach will be a comparative one, contrasting Balochi dialects with each other, and with data from earlier Iranian languages.

1.1 Ergative constructions

Nominative constructions are characterised by marking the subject of intransitive constructions (S) in the same way as the agent of transitive constructions (A) while the patient of transitive constructions (P) is marked differently. Ergative constructions, on the other hand, show identical marking of subject and patient, with the agent being marked differently (see e.g. PAYNE 1998:555). As a rule, ergativity in Iranian languages is of the split ergativity type, with nominative patterning of verbs forms from the present stem and ergative for the tenses formed from the past stem. These domains will be referred to as PRESENT and PAST domain, respectively, in what follows.2 So the shown in tables 1 and 2 coexist in the grammatical system in Ir. languages that show ergativity.3 The case used for the patient in nominative constructions is the same as the one used for the ergative agent (underlined).
1.: Marking of arguments in nominative constructions
2.: Marking of arguments in ergative constructions
1 This three way division of Balochi dialects follows JAHANI 2000:11 (see also KORN 2005:41 for more discussion). Although undeniably descending from a common protolanguage, it is questionable to which degree the Balochi dialects spoken today should be termed one language (see KORN, fthc. 3).
2 The two terms are capatalised to indicate that not all forms from the present stem are necessarily some sort of present tense, nor do all formations based on the past stem function as past. For details as to which constructions pattern ergatively and which ones nominatively in Balochi, see KORN (fthc. 1).
3 Cf. e.g. WINDFUHR 1992:31-32. It will be seen that this statement requires modification (as indeed mentioned by Windfuhr), see section 7.

1.2 The Balochi case system

Before embarking on the discussion of ergative Bal. constructions, a short look at the nominal system of Balochi is necessary. Table 3 shows the case system that I assume to underly all Bal. dialects.4 Apart from the vocative, there are four cases: direct, oblique and object case, genitive and vocative. The direct case has the ending -∅ both in the singular and the plural. In ergative constructions, the direct case is used for the patient while the oblique case (underlined) is used to mark the agent.

3.: Balochi case system

direct oblique object genitive vocative
-¯a -¯ar¯a -ai, -¯e, -¯ı, -a, -∅ -∅
plural -¯an -¯an¯a, -¯anr¯a -¯an¯ı -¯an
For the personal pronouns, it is necessary to list the forms of the three major dialect groups separately.

4.: Inflection of Balochi personal pronouns

direct oblique object genitive singular 1st WBal. man man¯a
SBal. man man¯a man¯ar¯a
EBal. mã, ma, m㯠ma¯ mana¯, man㯠ma¯ı, ma˜ı
2nd WBal. taw, ta tar¯a
SBal. tau, to¯ t(a)ra¯ tara¯ra¯ t(h)a¯ı, t(h)¯ı
EBal. thau, tha thar¯a
plural 1st5 WBal. (am)m¯a (am)m¯ar¯a
(am)mai, m¯e
SBal. m¯a m¯ar¯a
EBal. ma¯ ma¯r(a¯) ma˜ı¯
2nd WBal. š(u)m¯a šum¯ar¯a
šumai, šum¯e
SBal. šum¯a šum¯ar¯a
EBal. š(a)w¯a, š¯a š(a)w¯ar, š¯ar š(a)w¯a¯ı, š¯a¯ı
4 For discussion of this case system, see KORN (fthc. 3); for its history, see KORN (fthc. 2). For the case system of the Bal. dialects of Iran, see section 3.1. The transcription of Balochi has been put to unified system; the same applies to the glosses of the examples, some of which are based on the authors’, others are mine. Translations are meant literal rather than idiomatic to reflect the Bal. constructions. The left column of the exanmples specifies the dialect group and the subdialect (where known) of the sentences.
5 The forms am(m)¯a etc. are used in Afghan and Turkmen Balochi.

In most variants of Balochi, there is no distinction between direct and oblique case of the 1st and 2nd person pronouns. The WBal. dialects have only one form for the direct and the oblique case, which derives from the Middle Iranian oblique case. This form is classed as direct case in the remaining dialects, new oblique and object case forms being added to the system. However, even in the dialects which have a neo-oblique case, it is predominantly the forms deriving from the old oblique that are used as agent of ergative constructions (underlined).6 In addition, there are pronominal clitics. These are found in all functions of the oblique cases, including the agent of ergative constructions.7 For the 3rd person, demonstrative pronouns are used, which are for the most part inflected like nouns.

2. “Model” ergative constructions

Bal. ergative constructions of the standard type show the agent in the oblique and the patient in the direct case:
1 s¯abir-¯a ¯e haw¯al-∅ uškit
WBal. PN-OBL DEM news-DIR heard.PAST
(Pakistan) “Sabir heard this news.” (ELFENBEIN 1990/I:62 no. 5)8
2 ¯ay-¯a g¯ok-∅ kušt
SBal. DEM-OBL cow-DIR kill.PAST
(Karachi) “He/she killed the cow.” (FARRELL 1990:39)
3a haw¯e ˇc¯a\-∅ khay-¯a ˇia\-a
EBal. this.very well-DIR who-OBL strike-PERF
“Who has dug this well?”
3b haw¯e ˇc¯a\-∅ m¯a ˇia\-a
this.very well-DIR I.OBL strike-PERF
“I have dug this well.” (GILBERTSON 1923:121)
The manuscript Codex Additional 24048 of the British Library is the oldest known Bal. manuscript,9 it may date from around 1820 (ELFENBEIN 1983:1-4). As demonstrated in the examples quoted in what follows, Bal. ergative constructions at that period had more or less the same form as those of contemporary dialects. An example for the standard form is
4 mard-¯a ham¯e z¯al-∅ gipt
SBal. man-OBL this.very woman-DIR take.PAST
(19th c.) “The man took (i.e. married) this woman.” (CodOrAdd 24048: f. 1a, l. 3)10
In ergative constructions, the verb does not agree with the agent:
6 The same forms are also used after prepositions.
7 On the placement of these clitics, see DABIR-MOGHADDAM (fthc.).
8 This sentence is from a story in the dialect of Kharan in Pakistan.
9 For an edition, see ELFENBEIN 1983. In what follows, the text will be quoted according to folio (f.) and line (l.) of the manuscript plus page of Elfenbein’s edition. The transcription and the analysis are not always identical with those suggested by Elfenbein; glosses are mine.
10 ELFENBEIN 1983:10.

5 ã¯h-㯠to¯b¯ı ˇia\-a-∅
EBal. DEM-OBL.PL diving strike-PERF
“They have dived (lit.: have struck a dive).” (GILBERTSON 1923:59) Conversely, the verb may agree with the patient. There is no agreement in person11 of the verb with the patient in any dialect of Balochi, but 3rd person patient may agree with the verb in number, i.e. the 3pl. ending is optionally added if the patient is understood to be plural:
6 b¯anuk-¯a zahm-∅ kaššit drust-∅ ˇiat-ant
SBal. lady-OBL sword-DIR draw.PAST all-DIR strike.PAST-3PL
(19th c.) “The lady drew a sword [and] struck them all.” (CodOrAdd 24048: f. 4a, l. 2)12
7 ã¯h¯ı-a¯ kull-e˜¯ band¯ı-∅ yala ku\-ag-ant
EBal. DEM-OBL all-ADJ prisoner-DIR free do-PERF-3PL
(Marri) “He has freed all the prisoners.” (BASHIR 1991:104)13
8 z¯ı ã¯h¯ı-a¯ ma˜ı¯ ˇiarr-∅ šušt-ag-ã
EBal. yesterday DEM-OBL my clothes-DIR wash-PERF-3PL
(Marri) “Yesterday s/he washed my clothes.” (BASHIR 1991:104)14
Since the direct case has the ending -∅ both in the singular and the plural, agreement of the patient with the verb, i.e. the 3pl. ending of the verb, is the only indicator (besides the context) of plurality of the patient. Animacy and definiteness are not relevant here: plurality of animate as well as inanimate patients may be marked, neither need the patient be definite (see ex. 29). Ergative constructions that index the agent by way of a pronominal clitic have been treated as a separate type by some authors.15 It does not seem necessary to establish a separate type, though: pronominal clitics function as clitic form of the oblique case pronouns, so they may naturally also be used as ergative agent. These ergative constructions are indeed quite common. Some Bal. dialects use them for all persons, but in others, their use is limited to the 3rd person. It is significant that Bal. dialects where the distinction between direct and oblique case tends to be lost (see section 3.1) make ample use of the pronominal clitics, as their function is unmistakeably oblique.
9 p¯ıa¯la¯-∅=õ¯ zu¯rt-a
SBal. bowl-DIR=PRON.1SG seize-PERF
(Karachi) “I have taken the bowl.” (FARRELL 1990:54)
11 Since there is no gender in Balochi, there is obviously no agreement in gender either.
12 ELFENBEIN 1983:14. Elfenbein transcribes durust.
13 Bashir has kull¯e, which she interprets (Elena Bashir, personal communication) as containing the -¯e “one” (for
which see fn. 23), but it seems to make more sense to assume that the ¯e is nasalised, i.e. the suffix appearing on attributive adjectives.
14 BASHIR 1991:104 interprets these two sentences as showing past perfect, but it seems that they are regular present perfect examples with agreement of the verb with the object. Bashir’s EBal. examples are from an informant from the Marri tribe and appear to be elicited.
15 Thus e.g. MOŠKALO 1985:113-119, FARRELL 1995 and KALBA¯ S¯I 1988:78-82. I am grateful to Moritz Flatow for bringing the latter article to my attention.

10 b¯agp¯an-∅ gipt=¯ı
SBal. gardener-DIR take.PAST=PRON.3SG
(19th c.) “He seized the gardener.” (CodOrAdd 24048: f. 4a, l. 7)16
11 ma¯ı g¯oš-∅ buri\-ag-ant=iš
EBal. my ear-DIR cut-PERF-3PL=PRON.3PL
“They cut off my ears.” (GILBERTSON 1923:73)
The pronominal clitics may also occur in addition to an agent expressed as a noun:
12 ¯e sard¯ar-¯a g¯o man-∅ yak šart=¯ı kut-ag
SBal. DEM chief-OBL with I-DIR one bet=PRON.1SG do-PERF
(19th c.) “This chief made a bet with me.” (CodOrAdd 24048: f. 5b, l. 12)17
3. Marking of the agent
There are contexts in which the agent in the PAST domain is not in the oblique. This effects a marking of arguments that may be called neutral:18 the agent and the patient of transitive verbs, and the subject of intransitive verbs are all marked identically:
5.: Marking of arguments
in Balochi neutral constructions
This pattern is found under two entirely different conditions, viz. in the Bal. dialects spoken in Iran, and in all Bal. dialects in sentences with a pronoun of the 1st or 2nd pronoun as agent and a 3sg. as patient.

3.1 Ergative constructions in Iranian Balochi
Irrespective of their affiliation to one of the major dialect groups, the Bal. dialects spoken in Iran share a case system which markedly differs from that of other Bal. dialects, presumably due to the influence of Persian, whence they will be collectively termed “Iranian Balochi” (IrBal.) here. The genitive may be replaced by the e ˙z¯afe construction.19
16 ELFENBEIN 1983:14.
17 ELFENBEIN 1983:16-17. Elfenbein edits g¯on, but the photo of the manuscript seems to speak for g¯o, both variants of which are used in Balochi (cf. KORN 2005:181).
18 KALB ¯ AS¯I 1988:71 uses the term xon ¯s¯a “hermaphrodite, neutral”.
19 See JAHANI 2004 and 2003 for discussion of the IrBal. case system, for affiliation of IrBal. dialects and some of their features, see KORN 2005:256, for Bal. dialect groups, see also section 1.

6.: Case system of Iranian Balochi

nominative object genitive (or e ˙z¯afe)
singular -∅ -¯a(r¯a) -ey
plural -¯an -¯an¯a -¯an¯ı
Direct and oblique case tend to merge and yield a case that may be called nominative, with an ending -∅ in the singular and -¯an in the plural. The object case is used for indirect objects and for patients in the PRESENT domain. Being the conflation of the direct and the oblique cases, the nominative of Iranian Balochi marks both the agent and the patient of ergative constructions, and also the subject of intransitive verbs. So for Iranian Balochi, “neutral marking” means that agent, patient, and subject in the nominative case in the PAST domain.
Here are IrBal. examples for the subject in the nominative:
13 ost¯ad-¯an ez tehr¯an-∅ a yaht-ent
IrBal. teacher-NOM.PL from PN-NOM IPF come.PAST-3PL
(Sarawani) “The teachers were (lit.: were coming) from Tehran.” (BARANZEHI 2003:93)
14 pogol-㯠tawa¯r a ko
IrBal. frog-NOM.PL sound IPF do.PAST
(Sarawani) “The frogs were making noise.” (BARANZEHI 2003:103)
15 k¯ar-¯an=o tam¯am kapt-e-∅
IrBal. work-NOM.PL=PRON.1SG finish fall-PERF-3SG
“My works have become (lit.: fallen) finished.” (MAHMOODI BAKHTIARI 2003:143)20
The same case marks the agent:
16 e¯ sey-e˜¯ bacˇak-㯠(…) ro¯za=yeš wa¯rt-a
IrBal. this three-ADJ boy-NOM.PL fasting=PRON.3PL eat-PERF
(Sarawani) “These three boys have broken the fast.” (BARANZEHI 2003:94)
The plurality of the patient may still be marked on the verb:
17 n ¯ ˜u gw¯at-∅ ˇcan ˙d-¯ent-˜e
IrBal. now wind-NOM swing-CAUS.PAST-3PL
(Sarawani) “Now the wind swung [the clothes].” (BARANZEHI 2003:82)
18 mõ-∅ d¯at-˜e ramaz¯an-a ke ra-∅
(Sarawani) “I gave them to Ramazan, who [then] went.” (BARANZEHI 2003:83).
20 This sentence is not elicited according to MAHMOODI BAKHTIARI 2003:143 and indeed does not entirely
correspond to its Persian equivalent:
i k¯ar-h¯a=yam=r¯a tam¯am kard-e-am
NP work-PL=PRON.1SG=DO finish do-PERF-1SG
“I have finished my works (now that I am talking to you).”
It is noteworthy that in Iranian Balochi, the agent is expressed by a pronominal clitic in all persons wherever possible (see section 2.). Indeed, the use of these clitics is convenient in a system that would otherwise mark agent and patient identically.
19 ket¯ab=õ w¯ant
IrBal. book=PRON.1SG read.PAST
(Lashari) “I read (past tense) the book.”21
20 t¯an do s¯al dega ma-∅ l¯og=o zort-a
IrBal. until two year next I-NOM house=PRON.1SG seize-PERF
“I will have bought a house by the next two years.” (MAHMOODI BAKHTIARI
21 ˇcand wahd=¯e=yat ke yakk o degar=˜e na-d¯ıst-at
IrBal. some time=one23=
SUB one and other=PRON.1PL NEG-see-PPERF
(Sarawani) “It was some time since we had seen each other.” (BARANZEHI 2003:95)
22 n¯un=˜e belett-∅ gept
IrBal. now=PRON.1PL ticket-NOM take.PAST
(Sarawani) “Now we bought the ticket.” (BARANZEHI 2003:102)
23 dars-∅=en a wã¯
IrBal. lesson-NOM=PRON.1PL IPF read.PAST
(Khash) “We were studying.” (JAHANI 2003:125)
24 zekk-∅=¯ı t¯al¯an kort er ham-¯e tagerd
IrBal. pouring do.PAST from this.very mat
(Sarawani) “She poured out a goat skin on the mat.” (BARANZEHI 2003:83)
The agent is expressed both by a noun and a pronominal clitic specifically when the agent
is a 3sg.:
25 tam¯am-e s¯ıst¯an o bal¯oˇcest¯an-∅ xeil¯ı p¯ıšraft=¯ı kort-a
IrBal. all-EZ PN-NOM much progress=PRON.3SG do-PERF
(Zahedan) “The whole of Sistan and Balochistan has progressed a lot.” (JAHANI 2003:125)
26 al¯ı-∅ hasan-∅=¯ı zat
(Lashari) “Ali hit Hasan.”24
21 Elicited by the author from D¯od¯a Mahm¯udzah¯ı, Iranshahr (January 2005).
22 This sentence is the translation of the Persian sentence (i.e. elicited)
ii t¯a do s¯al-e d¯ıgar x¯ane xar¯ıd-e-am
NP until two year-EZ next house buy-PERF-1SG
“I will have bought a house by the next two years.”
23 The clitic -¯e is usually termed “indefinite article”, but this does not seem quite adequate: its cooccurrence
with the oblique ending shows that it rather denotes e.g. “one (specific)”, not “a (any)” (see also fn. 43).
27 go-˜e al¯ı-∅ ˇian¯ı=¯ı košt-a o ˇiest-a-∅
IrBal. say.PRES-3PL PN-DIR wife=PRON.3SG kill-PERF and jump-PERF-3SG
“They say that Ali has killed his wife and run away.” (MAHMOODI BAKHTIARI
In sentences like these, the use of the pronominal clitic disambiguates a sentence that otherwise would be open to two different analyses: as the pronominal clitic may not be suffixed to the agent, the noun that carries the clitic must be the patient, so Ali is the agent.

3.2 Personal pronouns as agent
As shown in table 4, the 1st and 2nd person pronouns have the same form in the direct and oblique case in Western and Eastern Balochi. In Southern Balochi, the form of the direct case is used for the agent in ergative constructions. The only exception is the EBal. 1sg. pronoun, which has a separate form for the oblique case that is also used for the agent (see ex. 3b).26 Except for the EBal. 1sg., a 1st or 2nd person agent expressed by a full pronoun is in (what is also) the direct case. Such sentences with a 3rd person patient show neutral marking:27
28 man-∅ wat¯ı l¯og-∅ pr¯ošt-ag
SBal. I-DIR own house-DIR break-PERF
(Kech) “I have broken my own house.” (MOCKLER 1877:86)
29 man-∅ xat-∅ likit-ã
SBal. I-DIR letter-DIR write.PAST-3PL
(Karachi) “I wrote letters.” (FARRELL 1990:40)
Examples of this type are not so common, though, because the agent seems to be particularly liable to be indexed by a pronominal clitic.

4. Marking of the patient
In addition to the neutral constructions, there are other patterns in Balochi that have the arguments of sentences in the PAST domain in something else than the ergative pattern. These do not show the patient in the direct case, but the patient is marked as it would be in a nominative construction, i.e. it is in the oblique or in the object case. Examples of this type 24 Elicited by the author from Mohammad Y¯usef Parvareš (Ra’¯ıs¯ı), Espake (January 2005). 25 This sentence is the translation of the Persian sentence
iii m¯ı-g-an al¯ı zan=eš=o košt-e-∅ o far¯ar kard-e-∅
NP PRES-say-3PL NP wife=PRON.3SG=DO kill-PERF-3SG and escape do-PERF-3SG
“They say that Ali has killed his wife and run away.”
26 This form is likely to have been introduced secondarily to match the pattern of agent marking in the oblique
(see KORN, fthc. 1).
27 For 1st and 2nd person patients, see 4.3.

have been considered as incorrect by some authors.28 However, they are rather common, so it seems more adequate to describe them as patterns in their own right, i.e. specific types of mixed constructions.

4.1 Patient in the oblique case
In sentences that have the patient in the oblique case and the agent not a 1st or 2nd person pronoun, the agent and the patient are marked in the same way. The difference to a neutral construction is that the subject of intransitive verbs is in a different case, and that agent and patient are in the oblique. This pattern may be termed “double oblique”.29

7.: Marking of arguments
in Balochi double oblique constructions
The existence of a pattern which has both the agent and the patient marked as oblique is noteworthy since it has been noted that such sentences do not occur.30
30 bacˇakk-a¯ wat¯ı danta¯n-㯠pro¯št
WBal. boy-OBL own tooth-OBL.PL break.PAST
(Pakistan) “The boy broke his teeth.” (BARKER/MENGAL 1969/I:348)
This construction existed already in the 1820s (see also ex. 38):
31 n¯am-¯a har kas-¯a z¯ant
SBal. name-OBL every person-OBL know.PAST
(19th c.) “Everyone knew the name.” (CodOrAdd 24048: f. 13b, l. 13)31
32 ¯e haps-¯a ¯o ¯e zahm-¯a kill¯ah-¯a paˇc=¯ı gipt
SBal. DEM horse-OBL and DEM sword-OBL fort-OBL open=PRON.3SG take.PAST
(19th c.) “He got hold of this horse and this sword [and] the fort.” (CodOrAdd 24048: f. 5b, l. 1-
28 See e.g. COLLETT 1983:21 (who says these constructions “should not” be used) and ELFENBEIN 1983:7.
29 Following HARRIS/CAMPBELL 1995:241. Such constructions are termed n¯ader “singular, uncommon” by
KALBA¯ S¯I 1988:73.
30 FARRELL 1995:222, 224. However, their occurrence is also noted by RZEHAK 1998:178. As Collett does not differentiate between what is oblique and object case here, and as he does not give examples, it is not clear whether the note about the existence of unusual ergative constructions (COLLETT 1983:21) refers to the patterns classed here as double oblique or to tripartite constructions, or to both.
31 ELFENBEIN 1983:30.

33 t¯ıng-¯a k¯az¯ı[-¯e] k¯er-¯a =¯e b¯un-¯a burrit
SBal. slave.girl-OBL officer[-GEN] penis-OBL from base-OBL cut.PAST
(19th c.) “The slave girl cut the officer’s penis from its base.” (CodOrAdd 24048: f. 4a, l. 5-
This pattern is ergativoid in that the verb does not agree with the agent, but may show agreement with the patient:
34 ma¯ zahm-㯠a¯r\-ag-ant
EBal. I.OBL sword-OBL.PL bring-PERF-3PL
“I brought the swords.” (GILBERTSON 1923:113) In all examples of a patient marked in the oblique in the PAST domain that I have found so far, the patient is definite: it seems that definiteness is a necessary condition for the patient being marked this way. However, definiteness does not imply that the patient needs to be in the oblique as is shown, for instance, by examples 1-3. So oblique marking of the patient is obviously optional and is likely to depend on pragmatic factors.

4.2 Patient in the object case
Other examples from the PAST domain have the patient in the object case (with double underlining):
35 kuˇcik-¯a ham¯a ˇiinik-¯ar¯a d¯ıst
SBal. dog-OBL that.very girl-OBJ see.PAST
(Karachi) “The dog saw this girl.” (FARRELL 1995:221)
36 ma¯ mard-ã¯ra¯ ˇia\-a
EBal. I.OBL man-OBJ.PL hit-PERF
“I have struck the men.” (GILBERTSON 1923:197)
32 ELFENBEIN 1983:16. Elfenbein reads zahm (against the photo of the manuscript). For what I assume to be kill¯ah-¯a, the photo indicates kul¯ah¯a, which Elfenbein transcribes as kull¯ah¯a and translates “entirely”, but it is not clear how kull¯ah¯a might be derived from kull “whole”, and in several other places in the story (cf. f. 4b, l. 2 and f. 6b, l. 9-10), zahm, haps and kill¯ah are enumerated as the possessions that are taken away first and given back later. Maybe the copyist mistook a šadda sign in the original for ˙damma.
33 ELFENBEIN 1983:14. I apologise for this example.
The manuscript, which often confuses vowel length, writes burr¯ıt, which is surely an error. Elfenbein transcribes t¯ınga (probably only a misprint), k¯ır¯ae (but the word is k¯er in all other Bal. sources) and b¯on, which is not known to me from other sources. The usual word is bun, so maybe it is a writing error. However, as the word is written throughout in this manuscript, it might perhaps be an existing variant, cf. Persian bon besides b¯un, which might be different developments from Proto-Ir. *budna-.
The genitive ending on k¯az¯ı is not present in this sentence, maybe due to some uncertainty how to write word-final -¯ı-¯e, but it is there in a variant of the same sentence occurring later on in the story: k¯az¯ı-¯e k¯er-¯a-¯e b¯un-¯a burritag “someone has cut…” (f. 6a, l. 12, ELFENBEIN 1983:18). The parallel in f. 6b, l. 3 (ELFENBEIN 1983:18) has drust ¯e (Elfenbein reads ¯ay¯ı) b¯un¯a burritag-ant “someone cut everything from the base”, -ant agreeing with the patient, makes clear that drust here and k¯er-¯a in the other sentences is the patient and that
b¯un¯a has locative function.
This construction is likewise already present in the 1820 manuscript:
37 d¯ıt=iš mard-¯ar¯a
SBal. see.PAST=PRON.3PL man-OBJ
(19th c.) “They saw the man.” (CodOrAdd 24048: f. 4b, l. 3-4)34
38 wat¯ı mardum-¯an¯a l¯o ˙t-¯a¯ent wat¯ı huštir-¯an=¯ı
SBal. own man-OBJ.PL want-CAUS.PAST own camel-OBL.PL=PRON.3SG
(19th c.) “He let ask for his men [and] his camels.” (CodOrAdd 24048: f. 2a, l. 3-4)35
This pattern has agent, patient and subject each in different cases:
8.: Marking of arguments in Balochi tripartite constructions
The difference between these examples and those in the preceding section is that a patient marked with object case ending is human while things, body parts and animals would have the oblique ending (see 4.1). This statement seems to be contradicted by one example in
39 t¯o-∅ ¯ay-r¯a ¯art
(Oman) “You brought it.” (COLLETT 1983:10, Collett’s translation)
Similarly, MOCKLER 1877:18 states that any noun has the endings -∅ or -¯ar¯a when
functioning as patient of an ergative construction, e.g.
40 mard-¯a aps-∅ /
SBal. man-OBL horse-DIR /
(Kech) “The man has killed the horse.” (MOCKLER 1877:21) It is not quite clear how this should be interpreted. The data adduced here by Collett and Mockler here are clearly not derived from free speech, but appear to be elicited, if not even
34 ELFENBEIN 1983:16.
35 ELFENBEIN 1983:12. Elfenbein translates “he asked for his man (sic), his drivers”, and edits l¯o ˙t¯aint and mardum¯an. The photo of the manuscript shows £” x£*HdÆ* (sic); this seems to indicate mardum¯an¯a which to the scribe or the copyist (who probably were not Baloch according to ELFENBEIN 1983:3-4) was not clear:
in the same way, b¯agp¯an¯a (OBL of b¯agp¯an “gardener”, usually spelled £”£C u£†) is written £” x£C u£† in f. 2b, last line – f. 3a, l. 1.
constructed by the authors themselves. It is not excluded, though, that some SBal. dialect(s?) patterns somewhat differently than the others. At any rate, one might say that human patients (and maybe in some SBal. varieties also other patients) may be marked with the object case ending if they are definite. Again, this marking is clearly optional, since sentences like example 4 show a definite human patient in the direct case. According to FARRELL 1995:224,
the marking depends on the presence of a specific emphasis on the patient. It remains to be seen, however, if more specific conditions can be found.
4.3 Personal pronouns
As pronouns of the 1st and 2nd persons are by definition human and definite, it is to be expected that they can appear in the object case when functioning as patient in the PAST domain as well. Indeed, nowadays they apparently have to be in the object case. In Southern Balochi, the use of the oblique is also possible.
41 ta-∅ be ˇc¯akar-∅ man-¯a baxšet
IrBal. you.SG-NOM to PN-NOM I-OBJ give.PAST
(Khash) “You gave me to Chakar.” (JAHANI 2003:126)
42 man-∅ ta-r¯a gušt
(Pakistan) “I told you.” (ELFENBEIN 1990/I:104 no. 100)36
43 r¯ah-¯a mn-¯a tunn-¯a ˇiat-a
WBal. way-OBL I-OBJ thirst-OBL strike-PERF (Afghanistan) “On the way, thirst has struck me.” (RZEHAK 1998:178)37
44 man-∅ ta-r¯a gitt
SBal. I-DIR you.SG-OBL take.PAST
(Karachi) “I caught you.” (FARRELL 1995:224)
45 ba¯dša¯h-a¯ man-㯠khušth-a
EBal. king-OBL I-OBJ kill-PERF
“The king has killed me” (GRIERSON 1921:352)
Again, the 19th century manuscript shows the same structures:
46 ¯e man-∅ b¯ıt-ag-¯an ki
(19th c.) ta-r¯a=un ¯awurt-ag y¯a digar=¯e b¯ıt-∅
you.SG-OBL=PRON.1SG bring-PERF or other=one be.PAST-3SG
“Was it me who (lit.: that I) has brought you, or was it another one (= someone else)?”
(CodOrAdd 24048: f. 8a, l. 5)38
36 The text is a story in the dialect of Kharan (Pakistan).
37 The Bal. dialect of Afghanistan is otherwise entirely nominative. However, epic poetry shows ergative and
other patterns as well.
38 The reading tar¯a-un cautiously suggested by ELFENBEIN 1983:20 seems to be the only solution that makes
sense for the manuscript’s HxÆ° .
47 man-¯ar¯a ¯e kamuk-¯a na-ˇiat=¯ı
(19th c.) “She did not hit me this bit (= not even a bit).” (CodOrAdd 24048: f. 3a, l. 13)39
However, in this manuscript, the pronoun also appears in the direct case when functioning as
48 man-∅=¯ı ˇiat
(19th c.) “She hit me.” (CodOrAdd 24048: f. 3a, l. 7)40
This manuscript seems to indicate a language change within the last 200 years, starting with an optional object case marking of human definite patients in general and ending with the 1st and 2nd person pronouns being always in the object case. The explanation may be that the form of these pronoun is (identical to) the direct case when it functions as agent, there is a strong motivation to mark it differently when occurring as patient. The reason for the marking being oblique or object in Southern Balochi instead of object case throughout is likely to be that the object case marking seems to be a recent system.41 5. Summary of case use in ergative constructions Table 9 presents the result of a counting of case uses in ergative and mixed constructions in
the first story in the British Library manuscript (ELFENBEIN 1983:10-21). The numbers here are not to be taken too literally, as some sentences may be open to different interpretations.
However, the table should give an idea of the relative frequency of sentence patterns.42 We may conclude that in Bal. sentence patterns of the PAST domain, the choice of the case of the patients seems to be governed by criteria of definiteness and animacy (table 10): if the patient is definite, things and animals are optionally in the oblique, thus identical with the agent, while humans may show a specific patient marking, which is not used for inanimate patients or agents, but also for animate patients in the PRESENT domain.
39 ELFENBEIN 1983:14. Elfenbein transcribes man¯ar¯a-¯ı, so that the sentence would contain two pronominal
clitics of the 3sg. The photo seems to indicate ¯e as it is marked with a diacritic sign which in other places of the manuscript is used to differentiate ¯e from ¯ı.
40 ELFENBEIN 1983:14.
41 The function of tar¯a etc. as oblique is due to a rearrangement of the SBal. pronominal system, which uses a doubly marked form (tar¯ar¯a etc.) for the object case (see table 4). The use of tar¯a etc. here may be said to reflect the stage prior to this adjustment.
42 “Transitive verb forms” here include compound verbs which have an additional patient. Multiple patients of one verb are counted as one if they are in the same case.
9.: Marking of agent and patient in the first story in CodOrAdd 24048
transitive verb forms in the PAST domain: 221. plurality of patient marked on these: 11 noun pronoun of the 1st, 2nd person
oblique: 118
“one” + oblique: 243
agent = pronominal clitic: 18
direct case: 2 direct case: 15
direct case: 64
direct + “one”: 12
+ 17 possible nominal parts of
compound verbs
OBL/OBJ case:
sg.: -¯a: 5, -¯ar¯a: 1
pl.: -¯an: 1, -¯an¯a: 1
direct case: 1 man¯ar¯a: 1
tar¯a: 1
10.: Marking of patients in Balochi ergative and mixed constructions
indefinite definite
non-human DIR OBL (optional)
human DIR OBJ (optional)
pronoun 1st, 2nd person – OBL/OBJ (1820 optional, today regular)
To the extent that Bal. neutral, double oblique and tripartite constructions have been noted at all, they have been explained as mixtures of the nominative and the active construction, i.e. by a mixing of the structures seen in tables 1 and 2.44 Such language-internal factors may certainly play a role, but it seems worthwhile to check for additional factors that might have influenced the Bal. sentence patterns.
6. Definiteness and animacy in ergative constructions of neighbouring languages
6.1 Urdu
Indic languages likewise display split ergativity, and the marking of the patient depends on criteria of definiteness and animacy. One might thus wonder whether influence from Urdu 43 These cases are:
yak r¯oˇc-¯e mardum-¯e-¯a ˇc¯arit “one day, a man looked” (f. 4a, l. 10-11); ELFENBEIN 1983:14 reads mardumiy¯a,
which would be morphologically unclear, and ˇc¯ar¯ıt, which is indeed what the manuscript has and would be the 3sg. present tense, but the past stem suffix -it is frequently written -¯ıt in this manuscript (cf. fn. 33);
yakk-¯e-¯a gušt “someone said” (f. 5b, l. 4); ELFENBEIN 1983:16 reads yakkay¯a, but translates “somebody said”, in which function his form would not be clear. In Balochi, the suffix -¯e “one” somes before the OBL ending in all dialects that allow this combination (see KORN, fthc. 2).
44 Cf. e.g. MOŠKALO 1985:121, who uses the term “contamination”. might have caused the Bal. ergative system.45
11.: Marking of patients in Urdu/Hindi ergative constructions
indefinite definite
inanimate NOM NOM
animate (humans and animals) NOM ACC However, comparing the Urdu system to the Balochi one, it emerges that they are not parallel: inanimate patients are not marked in Urdu, no matter whether they are definite or not, while they may be marked in Balochi if they are definite (cf. KLAIMAN 1987:76).
Conversely, marking of definite animate patients is regular in Urdu while it is optional in Balochi even if the patient is animate and definite. The animacy split is also different: while in Urdu, it is animates vs. inanimates, it is humans vs. the rest in Balochi. So the Bal. system of marking of patients is not likely to have been influenced by the Urdu system.
5.2 Bactrian
The animacy split of humans vs. not-humans recalls a phenomenon observed in Bactrian, a Middle Ir. language which was spoken in Northern Afghanistan and beyond and in several respects occupies an intermediary position between East and Western Iranian. Bactrian shows split ergativity with agreement of the verb wiht the patient in person and number:
49 oto=mo to … azado … uirt-hio
Bactr. and=PRON.1SG you.SG.DIR free release.PAST-2SG
“I released you.”46
The preposition abo, which has directional function, is also used to mark patients in the PRESENT and PAST domain if these are human and definite (SIMS-WILLIAMS 1998:86, 2004:2). In this example from the PRESENT domain, the first abo marks the patient, the second and third have local function: 50 od=aldo abo twmaxo abo lado od=abo razogolo oihl-amo Bactr. and=or to you.PL.OBL to court and=to royal tribunal bring.SBJ-1PL
“…or we should take you to court and to the royal tribunal.”47
The same marking is found for definite human patients in the PAST domain:
51 aggit=ido amaxo mano babo odo pidoko abo raliko olo
Bactr. receive.PAST=PTC we I.OBL PN and PN to PN wife
“We received – I, Bab, and [I], Piduk – Ralik [as our] wife.”48
45 For Urdu influence on Balochi, see e.g. FARRELL 2003, KORN 2005:48-50, for more on Urdu cases, see e.g.
46 From a deed of manumission (ed. SIMS-WILLIAMS 2000:45, document F, l. 8, maybe 480 AD).
47 From a contract for the purchase of an estate (SIMS-WILLIAMS 2000:59, document J, l. 24, possibly from 528
AD). twmaxo is only attested in oblique function (SIMS-WILLIAMS 2000:227).
The Bactrian constructions might indicate that criteria of animacy and definiteness were relevant in Iranian languages of the region already in Middle Iranian times.
6.3 Parthian
If this is the case, this might open an interesting aspect for Parthian, which is particularly relevant here since it is the Middle Iranian language that is most closely related to Balochi.
Like Bactrian, Parthian shows split ergativity with verbal agreement with the patient in person and number:
52 u=t az hišt h-¯em s¯ewag
Parth. and=PRON.2SG I.OBL leave.PAST COP-1SG orphan
“… and you have left me as an orphan.”49
53 u=š¯an ¯o murd¯an ¯edw¯ast h-¯em
Parth. and=PRON.3PL to dead-OBL.PL lead.PAST COP-1SG
“… and they have lead me to the dead.”50
In many Parthian examples from the PAST domain, a plural patient51 is not in the direct, but in the oblique case, thus marked identically with the agent. In example 53, the agent is expressed by the pronominal clitic -um, the patients, which are definite and human, are marked with the oblique ending, and the verb agrees with them: 54 ab¯aw=um harw-¯ın br¯adar-¯an ud wx¯ar-¯ın
Parth. there=PRON.1SG all-OBL.PL brother-OBL.PL and sister-OBL.PL
pad kirb¯ag wind¯ad ah-¯end
in piety find.PAST COP-3PL
“There, I found all brothers and sisters in piety.”52 Such examples have been interpreted as showing the ending being generalised as plural marker. This process is well-known to have happened in Middle Persian.53 It remains to be investigated, however, to which degree it has operated in Parthian, i.e. how many of the instances of an unexpected Prth. suffix involve the marking of a patient in an otherwise ergative sentence, and whether animacy and definiteness might play a role here as well.
48 From a marriage contract (SIMS-WILLIAMS 2000:33, document A, l. 15-16, maybe from 333 AD). amaxo serves both as direct and as oblique case of the 1pl. pronoun (SIMS-WILLIAMS 2000:179).
49 Fragment M 42 R i l. 15-16, quoted from DURKIN-MEISTERERNST p. 282.
50 Fragment M 7 II V ii, l. 1-3 (transliteration and German translation in ANDREAS/HENNING 1934:29.
51 In the singular, nouns (including family terms, cf. SIMS-WILLIAMS 1981:170) are not differentiated for case.
52 Transliteration and German translation in ANDREAS/HENNING 1934:858. Part of the example is also cited in RASTORGUEVA/MOLˇCANOVA 1981:223.
53 Cf. e.g. SUNDERMANN 1989:155. The same process also takes place in IrBal. dialects (see 3.1).

7. Conclusion
7.1 Balochi sentence patterns
The discussion above has revealed the existence of a large variation of sentence patterns in Balochi: in addition to nominative and ergative patterns, there are neutral, double oblique and tripartite patterns. Bal. neutral double oblique and tripartite patterns are charaterised by the verb optionally agreeing with 3pl. patients. These patterns interact in complex ways: Balochi as a whole patterns nominatively in the PRESENT domain, and in sentences of the PAST domain that have a pronoun of the 1st or 2nd person54 both as agent and as patient. Some WBal. varieties pattern nominatively also in all other contexts. The remaining dialects show neutral patterning for 1st and 2nd pronoun agents in sentences with a 3rd person patient. For other constructions, the dialects diverge considerably. For Iranian Balochi, neutral patterning is the general pattern in the PAST domain, while the remaining dialects have ergative constructions. Instead of the ergative, double oblique may be used for definite non-human patients and tripartite patterning for definite human ones.

12.: Patterns of argument marking in Balochi dialects
Western Balochi Southern
PRESENT domain nominative pattern (table 1)
PAST domain:
agent and patient 1st, 2nd pronoun
PAST domain:
agent 1st, 2nd pronoun, patient 3rd nominative
(table 1)
neutral pattern (table 5)
PAST domain: agent and patient 3rd
ergative pattern (table 2) neutral pattern
(table 5)
PAST domain:
patient 3rd definite non-human ergative pattern (table 2) or double oblique pattern (table 7)
PAST domain:
patient 3rd definite human ergative pattern (table 2) or tripartite pattern (table 8)
PAST domain:
patient 1st, 2nd pronoun tripartite pattern (table 8)
54 In this table, “pronoun” denotes “full pronoun” (to the exclusion of pronominal clitics). For the EBal. 1sg. pronoun, see 3.2. The nowadays regular marking of 1st and 2nd person pronouns in a way that is different from that of 3rd persons may be described as a Identified Object Marking (IOM) or Differential Case Marking phenomenon (see FARRELL 1995:222, MIRDEHGHAN, fthc.). FARRELL 1995:224 argues that the optional marking of patients (in Farrell’s view only with object case endings) is not a candidate for IOM as it does not depend on identification, but on emphasis. However, the data suggest that only identified objects may be marked (albeit additional factors are also necessary) while unidentified may not, so the oblique and object case marking of patients may also be interpreted within an IOM framework.

Thence some WBal. dialects only show one pattern, Iranian Balochi shows nominative and neutral pattern and the remaining dialects show all five patterns that have been observed in language typology. Indeed, it appears that no Bal. dialect has (only) the two patterns shown in table 1 and 2.

7.2 The context of Iranian ergative constructions
It seems rather plausible that a similar statement might apply to other Iranian languages as well, as can indeed be inferred from the Bactrian examples given above. For instance, STILO 2004:243 notes nominative, ergative and double oblique constructions for Vafsi. However, the Vafsi double oblique constructions differ from the Bal. ones in that the verb agrees rather with the subject. Tripartite constructions are not uncommon in Western Iranian languages either: in Middle Persian and Parthian, patients (and indirect objects) in the PRESENT and PAST domain may be marked by the preposition ¯o,55 so that in the PAST domain, there is tripartite marking besides ergative. There is also a certain tendency to neutral marking in sentences where ¯o is not used, as direct and oblique case are in many instances not distinguished (see 6.3).
These data taken together might tend to speak against the terms in which Iranian neutral, double oblique and tripartite constructions in the PAST domain have been described. So far, these have been thought to show a “decay” of ergativity and “transition” between ergative and nominative constructions.56 It goes without saying that from a diachronic point of view, this is certainly correct in that the starting point are ergative constructions, and it is possible that the end point is a consistent nominative patterning as it is in the case of New Persian and some WBal. varieties. However, the presence of neutral, double oblique and tripartite constructions in such a wide range of languages from the Middle and New Iranian period would seem to indicate that such constructions may indeed be rather stable,57 so in this sense, their labeling as transitory or decaying is somewhat misleading. The mixed constructions may indeed have persisted for quite a long time (and continue to do so), and coexist with other patterns in one and the same language. Abbreviations:
1sg., 1SG 1st person sg. (other persons accordingly)
A agent (of transitive verbs)
ACC accusative case
ADJ adjective suffix
Bal. Balochi
CAUS causative
CodOr 24048 = ed. ELFENBEIN 1983
COP copula
DEM demonstrative pronoun
55 Unlike Bactrian, this marking seems to be independent of animacy, e.g.
iv nidraxt ¯o haw-¯ın panˇi ahrewar
Parth. oppress.PAST to that-OBL.PL five pit.of.death
“(The Prince of Darkness) subdued those five pits of destruction.” (Fragment M 507 V l. 14,
transliteration and translation in BOYCE 1952:441)
For the uses of ¯o, see also BRUNNER 1977:132-140 and DURKIN-MEISTERERNST p. 230-238.
56 Cf. e.g. FARRELL 1995:218, 240 and SIMS-WILLIAMS/CRIBB 1996:87, 90.
57 See also WENDTLAND 2005 for data that seems to point in the same direction.
DIR direct case
DO marker of direct object
EBal. Eastern Balochi
EZ e ˙z¯afe
f. folio
GEN genitive case
IPF imperfective aspect
Ir. Iranian
IrBal. Iranian Balochi (= Balochi spoken in Iran)
l. line
NEG negation
NOM nominative
NP New Persian
OBJ object case
obl., OBL oblique case
P patient
PAST past tense; domain of ergativity (see 1.1)
PERF present perfect
pl., PL plural
PN personal name
PPERF past perfect
PRES present tense
PRESENT domain of nominative constructions (see
PRON pronominal clitic
PTC particle
S subject (of intransitive verbs)
SBal. Southern Balochi
SBJ subjunctive mood
sg., SG singular
SUB subordinating particle
WBal. Western Balochi


ANDREAS, Friedrich C., and Walter B. HENNING 1934: “Mitteliranische Manichaica aus Chinesisch Turkestan
III”. In: Sitzungsberichte der Preußischen Akademie der Wissenschaften, pp. 846-912 (= Walter B. HENNING 1977: Selected Papers I [Acta Iranica 14], pp. 275-339)
BARANZEHI, Adam Nader 2003: “The Sarawani dialect of Balochi and the Persian influence on it”. In: JAHANI/KORN, pp. 75-111
BARKER, Muhammad A., and Aqil Khan MENGAL 1969: A Course in Baluchi. Montreal: McGill University, 2 vol.
BASHIR, Elena 1991: A Contrastive Analysis of Balochi and Urdu. Washington: Academy for Educational Development
BOYCE, Mary 1952: “Some Parthian Abecedarian Hymns”. In: Bulletin of the School of Oriental and African Studies 14, pp. 435-450
BRUNNER, Christopher 1977: A Syntax of Western Middle Iranian [Persian Studies Series 3]. Delmar, NY: Caravan Books
BUTT, Miriam, and Tracy Holloway KING 2004: “The Status of Case”. In: Veneeta DAYAL, Anoop MAHAJAN (eds.): Clause Structures in South Asian Languages [Studies in Natural Language & Linguistic Theory 61]. Boston etc.: Kluwer, pp. 153-198\ COLLETT, Nigel A. 1983: A Grammar, Phrase Book and Vocabulary of Baluchi. Abingdon: Burgess
DABIR-MOGHADDAM, Mohammad (fthc.): “On Agent Clitics in Balochi in Comparison with Other Iranian
DURKIN-MEISTERERNST, Desmond (no year): Grammatik des Westmitteliranischen (Parthisch und Mittelpersisch) auf Grund manichäischer Texte, des Inschriftenmaterials und auszugsweise der Pahlav¯ı-
Literatur. Münster (unpublished manuscript)
ELFENBEIN, Josef 1983: A Baluchi Miscellany of Erotica and Poetry: Codex Oriental Additional 24048 of the British Library. Naples: Istituto Universitario Orientale
––– 1990: An Anthology of Classical and Modern Balochi Literature. Wiesbaden: Harrassowitz, 2 vol.
FARRELL, Tim 1990: Basic Balochi. An Introductory Course [Baluchistan Monograph Series 1]. Naples: Istituto Universitario Orientale
––– 1995: “Fading Ergativity? A Study of Ergativity in Balochi”. In: David C. BENNETT, Theodora BYNON, B.
George HEWITT (eds.): Subject, Voice and Ergativity: Selected Essays. London: School of Oriental and
African Studies, pp. 218-243
GILBERTSON, George W. 1923: The Balochi Language. A grammar and manual. Hertford: Austin & Sons
GRIERSON, George A. 1921: “Bal¯och¯ı”. In: Linguistic Survey of India X: Specimens of Languages of the Eranian
Family. Calcutta: Superintendent Gov. Print., pp. 327-451
HARRIS, Alice C., and Lyle CAMPBELL 1995: Historical Syntax in Cross-Linguistic Perspective [Cambridge
Studies in Linguistics 74]. Cambridge: Cambridge University Press
JAHANI, Carina 1994: “Notes on the use of genitive construction versus i ˙z¯afa construction in Iranian Balochi”.
In: Studia Iranica 23, pp. 285-298
––– (ed.) 2000: Language in Society – Eight Sociolinguistic Essays on Balochi [Studia Iranica Upsaliensia 3].
Uppsala: Uppsala University
––– 2003: “The Case System in Balochi in a Contact Linguistic Perspective”. In: JAHANI/KORN, pp. 113-132
JAHANI, Carina, and Agnes KORN (eds.) 2003: The Baloch and Their Neighbours: Ethnic and Linguistic Contact
in Balochistan in Historical and Modern Times. Wiesbaden: Reichert
JAHANI, Carina, Agnes KORN, and Paul TITUS (eds.) (fthc.): Pluralism in Balochistan. Wiesbaden: Reichert
KALBA¯ S¯I, I¯ra¯n 1988: “Ergat¯ıv dar zaba¯nha¯ va gu¯yešha-ye I¯ra¯n”. In: Zaba¯n-šena¯s¯ı 5/2 (1367 h.š.), pp. 70-87
KLAIMAN, Miriam 1987: “Mechanisms of Ergativity in South Asia”. In: Robert DIXON (ed.): Lingua 71, pp. 61-
KORN, Agnes 2005: Towards a Historical Grammar of Balochi. Studies in Balochi Historical Phonology and
Vocabulary [Beiträge zur Iranistik 26]. Wiesbaden: Reichert
––– (fthc. 1): “The Ergative System in Balochi from a Typological perspective”. In: Behrooz MAHMOODI
BAKHTIARI (ed.): Studies on the Typology of the Iranian Languages.
––– (fthc. 2): “Das Nominalsystem des Balochi, mitteliranisch betrachtet”. In: Günter SCHWEIGER (ed.):
Festschrift für Gert Klingenschmitt
––– (fthc. 3): “The Nominal Systems of Balochi: How Many Grammars?” In: JAHANI/KORN/TITUS
MAHMOODI BAKHTIARI, Behrooz 2003: “Notes on the Tense System in Balochi and Standard Persian”. In:
JAHANI/KORN, pp. 133-145
MIRDEHGHAN, Mahinnaz (fthc.): “Split-Ergative Morphology in Hindi/Urdu, Pashto and Balochi”. In:
MOCKLER, E. 1877: A Grammar of the Baloochee Language, as it is Spoken in Makr¯an (ancient Gedrosia), in
the Persi-Arabic Character. London: Henry S. King
MOŠKALO, Vyaˇceslav V. 1985: “Sprjaženie perexodnyx glagolov prošedšego vremeni v beludžskom jazyke”. In:
Iranskoje jazykoznanije. Ežegodnik 1981. Moscow, pp. 113-125
PAYNE, John R. 1988: “Ergative construction”. In: Encyclopædia Iranica 8, pp. 555-558
RASTORGUEVA, Vera S., and E. K. MOLˇCANOVA 1981: “Parfjanskij jazyk”. In: Osnovy iranskogo jazykoznanija:
Sredneiranskie jazyki. Moscow: Nauka, pp. 147-232
RZEHAK, Lutz 1998: “Das belutschische Epos von Šay Muru¯d und A¯ no (Materialien aus dem Nachlaß von
Alexander L. Grünberg)”. In: V. V. KUSHEV et al. (eds.) 1998: Strany i narody vostoka XXX: Central’naja
Azija. Vostoˇcnyj Gindukuš. St. Peterburg: Peterburgskoe Vostokovedenie, pp. 149-186
SIMS-WILLIAMS, Nicholas 1981: “Notes on Manichaean Middle Persian Morphology”. In: Studia Iranica 10, pp.
––– 1998: “Further notes on the inscription of Rabatak, with an Appendix on the names of Kujula Kadphises
and Vima Taktu in Chinese (pl. 9-12)”. In: id. (ed.): Proceedings of the Third European Conference of
Iranian Studies held in Cambridge, 11th to 15th September 1995. Part I: Old and Middle Iranian Studies.
Wiesbaden: Reichert, pp. 79-92
––– 2000: Bactrian Documents from Northern Afghanistan I: Legal and Economic Documents [Corpus
Inscriptionum Iranicarum II/6]. Oxford etc.: Oxford University Press
––– 2004: The Bactrian language (handout of a lecture at Bonn university, June 2004)
SIMS-WILLIAMS, Nicholas, and Joe Cribb 1996: “A New Bactrian Inscription of Kanishka the Great”. In: Silk
Road Art and Archeology 4, pp. 75-142
STILO, Donald 2004: Vafsi Folk Tales. Twenty Four Folk Tales in the Gurchani Dialect of Vafsi as Narrated
by Ghazanfar Mahmudi and Mashdi Mahdi and Collected by Lawrence P. Ellwell-Sutton [Beiträge zur
Iranistik 25]. Wiesbaden: Reichert
SUNDERMANN, Werner 1989: “Mittelpersisch”. In: Rüdiger SCHMITT (ed.): Compendium Linguarum Iranicarum.
Wiesbaden: Reichert, pp. 138-164
WENDTLAND, Antje 2005: “On Ergativity in Pamir Languages” (paper read at the First International Conference
of Iranian Linguistics, June 2005, Leipzig)
WINDFUHR, Gernot L. 1992: “Cases”. In: Encyclopædia Iranica 5, pp. 25-37


State Control and its Impact on Language in Balochistan

Prof Dr. Carina Jahani

By: Prof Dr. Carina Jahani

The purpose of the present article will be to present certain aspects of the sociolinguistic situation among one of the least studied ethnic groups in the  Middle East, the Baloch, who inhabit the south-eastern corner of the Iranian linguistic area. It is an area where the dominance of the state is relatively recent, and where modern society with a monetary economy, a settled lifestyle, mass education, state administration etc. is just being established. It is particularly interesting to study language-related decisions of the state, and the implementation of these decisions in a region like Balochistan, where until recently there were no such phenomena as e.g. language planning, education, mass media, newspapers or administrative language. However, in Iran the Persian language and in Pakistan Urdu and English have started to play a constantly growing role in Balochistan, something which is by many Baloch felt as a threat to both their language and their distinct ethnic identity. It must be stressed that modernity is not regarded as negative, but the Baloch intellectuals face the dilemma of how to retain their ethnic and linguistic diversity at the same time as they seek active participation in an increasingly globalised world.

The Historical Background

The border between Iran and Pakistan, which cuts through the traditional land of the Baloch, has since the time of its demarcation in the late nineteenth century been constantly questioned and frequently ignored by the Baloch living on both sides of it. It is called the Goldsmid line, and was drawn by a border commission headed by the British general Goldsmid, which also held representatives from Tehran and the Balochi Khanate of Kalat (see below) (Breseeg 2001: 133-134, Hosseinbor 2000: 73-80).1 However, it has had a considerable impact on linguistic issues, and it is therefore interesting to study the position of Balochi on both sides of this border.
There is very little known about the early history of the Baloch, but two main theories prevail as to when they arrived in their present habitat, which includes south-eastern Iran, south-western Pakistan and southern Afghanistan.
The ‘native theory’ argues that the core of the Baloch settled in Balochistan and mixed with other local peoples as early as 2000 years ago, as a continued movement of the Aryan tribes that had already invaded the Iranian plateau from the north. The ‘migration theory’, supported by the indigenous epic tradition as related in the epic poetry on genealogies and the wanderings of the Baloch tribes, suggests that the Baloch arrived in Balochistan from the northwest considerably more recently, some time around the tenth century A.D. In fact, the ballads suggest a Semitic origin for the Baloch and a close relation to the prophet Muhammed. This could, however, be seen as a pseudohistoric way of legitimising the Baloch as good and orthodox Muslims. Other origins, such as Turkic or Indian, have also been suggested for the Baloch (Dames 1904: 7). It may well be that the Baloch earlier in their history were ‘a series of tribal communities not sharing any feelings of common ethnicity’ (Spooner 1989: 607), and that even though linguistic evidence suggests the likelihood that at least a core group were of Indo-European origin2 who had migrated from the north-west, ‘Arab groups could have found their way into the heterogeneous tribal population that eventually assimilated Baluch identity east of Kerman’ (Spooner 1989: 609). Arab historians from the ninth and tenth centuries A.D. associate them with the area between Kerman, Khorasan, Sistan and Makran (Spooner 1989: 606). It is also possible that they assimilated a major part of the local inhabitants in Balochistan when they settled there.
It is not possible to talk about ‘a Baloch race’ (cf. Dames 1904) in order to distinguish them from neighbouring peoples, but there are other factors which bind them together and separate them from others in the region. Anthony D. Smith (1986: 21) finds that the term ethnos ‘would appear to be more suited to cultural rather than biological or kinship differences’. Among such cultural differences, he enumerates ‘a collective name’, ‘a common myth of descent’, ‘a shared history’, ‘a distinctive shared culture’, ‘an association with a specific territory’ and ‘a sense of solidarity’ as crucial components of ethnic affiliation (Smith 1986: 22-31). All these factors are applicable in the case of the Baloch.
Among the components of a shared culture, those of language and religion are particularly important, and the Balochi language as well as the Sunni creed are distinguishing factors in relation to neighbouring ethnic groups.3 It is important to be able to distinguish the ‘self-group’ from other surrounding ethnic groups. In fact, it is only in an interactive relation to other groups that are perceived as different that a delimitation of the ‘own-group’ versus the others becomes meaningful (see e.g. Eriksen 1993). In Iran the Sunni creed is crucial in that respect, since the Balochi language is closely related to Persian and is normally in the official discourse described as a ‘dialect’ (guyeš) of the Persian language (zabān), whereas the majority in Iran, contrary to most Baloch, profess Shi’a Islam. In Pakistan, on the other hand, the language, which is not closely related to Sindhi, Lahnda, Punjabi, Urdu or other Indian languages and very distinct from the eastern Iranian language Pashto, is more crucial, since the majority of the Muslims in Pakistan, including a vast majority of the Baloch, profess Sunni Islam.
The traditional socio-economic systems in Balochistan divide the land into a northern part and a southern part.4 In the north, pastoral nomadism has been the predominant lifestyle, whereas in the south agriculture, with few landowners and landless workers or slaves, has been more common. The tribal structure has, however, historically been a uniting factor among free-born Baloch in all Balochistan, and it has been easy for originally non-Baloch tribes and clans to associate with and incorporate themselves into the Balochi tribal system.5 Nowadays the de-tribalisation process is strong, especially in those parts of Balochistan where the traditional economy is based on settled agriculture rather than on pastoral nomadism.6 Tribal loyalties are also often felt to hamper a strong nationalist movement, and many intellectual Baloch nowadays try to propagate the replacement of tribal (sub-national) loyalties with loyalty to the entire Balochi ethnie (see Smith 1986: 21).
In the seventeenth century the Baloch allied themselves with another tribal people, the Brahuis,7 against other forces in the region, and this Balochi- Brahui Khanate, with its centre in Kalat (in present-day Pakistan) continued to exist until 1947. It was especially powerful during the second half of the eighteenth century, under Nasir Khan I, who ‘was the only khan who successfully transcended tribal loyalties’ (Spooner 1989: 611), but it was later weakened and incorporated into the British administration in 1839. The language of administration in Kalat was from the beginning Persian (Baloch 1987: 120),8 but English later replaced Persian for official purposes.
In the nineteenth century the Qajar shahs, ruling from Tehran, made several attempts to subdue the western parts of Balochistan. Likewise, British India had intentions of expanding westwards in Balochistan. This is the background of the Goldsmid border commission, and the demarcation that resulted from it divided most of the Balochi mainland between British India and Iran.9 Even so, the Qajars never succeeded in establishing their power in Balochistan, and it was only in 1928 that the newly established Pahlavi monarchy was successful in imposing direct control over the province.

Official Language Policy and the Impact of Education and Mass Media

Education in Balochistan prior to the modern era was comprised of the traditional Islamic madrasa-education. There was no possibility of studying the Balochi language and literature within this system, since there was no written literary tradition in Balochi to refer to. The languages employed were Arabic, which was the language of religion and science, and Persian, the language of a long and elevated literary tradition.
Along with modernisation came a secular education system and a nationalist discourse as well, first in British India, and later in Iran. Hosseinbor finds several reasons as to why a nationalist movement demanding political, linguistic and cultural rights for the Baloch was much slower to emerge in western (Iranian) Balochistan than in eastern (Pakistani) Balochistan, e.g. the ‘extremely slow pace of urbanization, the absence of social and economic modernization, and the very limited modern education introduced in western Baluchistan prior to the 1960s’ as well as the ‘repressive Pahlavi rule’ (Hosseinbor 2000: 150).
When the demands for freedom from the colonial yoke grew stronger and stronger in British India, several of the educated young Baloch started to demand a free Balochistan as well. Some of these nationalists were also poets and writers and played a major part in the Balochi literary movement. The British had already paid considerable attention to the Balochi language. A number of grammatical descriptions and dictionaries of various Balochi dialects were produced in the late nineteenth and early twentieth century.10 Balochi oral epic poetry, folktales, stories etc. were collected and published by M. Longworth Dames, who as a result made a great contribution to the preservation and study of Balochi oral literature.11 Examinations were also held in Balochi, and the colonial officials were encouraged to learn the language (Bruce 1900: 69). When the young nationalists from the 1930s onwards started to write poetry, stories and other literary pieces in Balochi, they thus had a certain\ tradition to fall back on. The importance of books and periodicals in Balochi published from the mid-1950s onwards must not be underestimated. These publications, however, are also very limited in many ways. Firstly, their distribution is geographically limited, largely to Pakistan, and secondly, their readership is very limited. All educated Baloch have received their education in English/Urdu (in Pakistan), Persian (in Iran and Afghanistan) or another language, e.g. Arabic in the Gulf States. In addition, the Balochi literary movement is founded solely on personal initiatives, with next to no official support. Therefore, only a small literary elite takes an interest in reading books and magazines in Balochi, which also places financial constraints on publishing in Balochi (see e.g. Dashtyari 2003).There is no official use of Balochi as a language of administration or education in Pakistan, even if voices have been raised in particular for introducing it as a language of education. The official support the language receives in Pakistan is that it is taught at the University of Balochistan in Quetta, that it has an official academy, the Balochi Academy, founded in 1961 and that it is used as a language for radio and TV transmissions. There is also a periodical, Ulus, published in Quetta by the Ministry of Information and Broadcasting.12
In the early 1990s measures were taken to introduce Balochi as a language of primary education in those areas of Pakistani Balochistan where it is spoken by a majority of the population. This experiment was not very successful, however, due to several reasons, including disagreement among the Baloch on orthographic issues and the lack of qualified teachers. What was perhaps more important was the speakers’ attitude towards Balochi, which many Baloch see as a ‘backward and rural’ language, knowledge of which offers no improvement in terms of social or economic status in today’s Pakistan. Likewise, people felt threatened by the fact that the Brahuis and the Baloch would be taught in different languages, even though as a result of their political alliance in the Khanate of Kalat, they strongly identify as one people. There is also widespread bilingualism (Balochi – Brahui) among the Brahuis (Farrell 2000: 24-25).
Although the cultural climate in British India, and later in Pakistan, was not totally negative to a Balochi cultural and literary movement, things were quite different on the other side of the Goldsmid line. The linguistic and cultural policy of the Pahlavi monarchs was that of strict conformity to the majority. All attempts at strengthening local customs, traditions and cultures were viewed as opposition against the nation and as threats to the territorial integrity of Iran. Especially those languages spoken within the borders of Iran that are related to Persian (the Iranian languages)13 were regarded as local dialects of Persian. Under such circumstances there was, of course, no provision made by the government for mother tongue education or even cultural activities or publication in the minority languages. Mojab and Hassanpour describe this cultural and linguistic hegemony as the propagation of ‘racist and national chauvinistic myths in the state-controlled media, in educational institutions (all state owned), and in government organs’, denying the national, linguistic and cultural diversity of Iran (Mojab and Hassanpour 1995: 231-232).
According to the Constitution of the Islamic Republic of Iran, chapter 1, article 15, in addition to the official language Persian, ‘[t]he use of regional and national…languages in the press and mass media…as well as for teaching in schools the literatures written in them, is permitted’ (Algar 1980: 34). This means that it is in principle permitted to publish books and newspapers in Balochi, but at present there is hardly any such publication taking place in Iran. When it comes to teaching Balochi literature in the schools, there is, of course, no provision being made for such a subject, due to the almost total lack of Balochi literature in written form. The first time publication in Balochi was possible in Iran was directly after the Islamic Revolution (1979-1980). A number of magazines appeared for some months, but this publication was soon forced to cease. In the late 1990s publishing in Balochi was resumed, and two bilingual magazines (Persian- Balochi) are at present being published, one from Iranshahr and one from  Zahedan. As for radio programmes, the situation is different, and Radio Zahedan has daily broadcasts in Balochi. In fact, these broadcasts date back at least to the 1960s, thus to the time of the Pahlavi monarchy.14 The contents of these broadcasts are usually viewed with suspicion by the Baloch, being regarded as ‘official propaganda’ rather than as genuine concern for the Balochi language.
There is no provision being made for TV transmissions in Balochi in Iran. State control in Iran expresses itself as control over education, administration, media, publication etc., and thus there is an exclusive or nearly exclusive use of Persian in these language domains. There is no interest on the part of the state to stimulate the development of a vigorous Balochi language to be used in media, education and administration. The reason  for this is obvious. There is a strong fear that a movement in support of cultural autonomy would soon develop into a political movement with demands for independence. It should, however, be said in favour of the present regime that it has allowed much more cultural plurality than the Pahlavi monarchy. TV programmes showing regional variations in e.g. lifestyle, dress, dance etc. are frequently broadcast. Permission has been given to arrange ‘poetry evenings’ with recital of Balochi traditional and modern poetry e.g. in Chabahar where many culturally active Baloch live. The bilingual magazines in Persian- Balochi (see above) are also a positive feature. There is, in fact, a considerable publication (books, newspapers etc.) taking place in the two largest minority languages Azerbaijani and Kurdish, and in the academic year 2004-05 B.A. programmes in the Azerbaijani language and literature (in Tabriz) and in the Kurdish language and literature (in Sanandaj) are offered in Iran for the very first time.15 There is also a Department of Gilan Studies at the University of Rasht.16 In Pakistan the use of Balochi in education, administration and media is also very limited, although not as restricted as it still is in Iran. Balochi is mainly spoken at home, within one’s immediate community (with relatives, neighbours, friends etc.), and sometimes at work as well. When it comes to religion, Arabic is the language of recitation and worship, whereas sermons are normally delivered in Balochi for the sake of comprehension in Balochi speaking communities both in Iran17 and in Pakistan. On the whole, however, Balochi may be regarded as the language of the traditional domains, which carry no particular status in today’s society. Prestigious domains such as administration and education, which hold the opportunities for advancing in society, are non-Balochi domains. Here the state language(s) is/are totally predominant. Therefore, many parents who are eager for their children to advance in society prefer them to learn these languages instead of Balochi.18 The survival of Balochi would under such circumstances only be possible at the cost of education and progress in society. Farrell (2000: 20) finds that ‘[a]t present it is partly lack of education that is ensuring the strength of Balochi’, a situation that is, of course, both impossible and undesirable to perpetuate. The state, i.e. the ruling elite, may desire to keep the Baloch uneducated and unable to participate in modern political discourse, but that is definitely no longer the desire of a majority of the Baloch themselves. Even though the literacy rate is still low in Balochistan, it is gradually rising. I have made my own observations and also interviewed several people who live in Iranian Balochistan or who have visited the region recently, and it is quite clear that most children there, both boys and girls, nowadays receive at least primary education. The traditional lifestyle is more and more giving way to modern life, which makes school attendance easier19 and more attractive. Thus, nearly the entire younger generation becomes acquainted with Persian at least from the age of six, when they begin school. The socio-cultural hindrances against girls’ education were also to a certain degree weakened after the establishment of the Islamic Republic, with separate boys’ and girls’ schools and the compulsory veil. On the other hand, reports also indicate that in small villages with only a few children, boys and girls are taught together in one class, something which causes most girls to drop out after elementary
In the census carried out in 1996 (Ābān 1375 in the Iranian Anno Hijra solar calendar) the literacy rate for the province of Sistan and Balochistan was 57 per cent for persons six years of age and older. Looking at the literacy rate by gender, the breakdown was 49 per cent for females and 65 per cent for males (Iran Statistical Yearbook 1377 2000: 603). The definition of literacy was very generally stated as ‘all individuals who can read and write a simple text in Farsi or any other language’ and even those who had studied only the first year in primary school or the equivalent were counted as literate (Iran Statistical Yearbook 1377 2000: 595).
The figures for Sistan and Balochistan may be compared to the average literacy rate for the entire country, which was 80 per cent for the same year,20 and to the province with the lowest literacy rate after Sistan and Balochistan, namely Kurdistan, where 68 per cent of the population was literate (Iran Statistical Yearbook 1377 2000: 603).21 Ten years earlier, in 1986 (Mehr 1365 in the Iranian Anno Hijra solar calendar), the literacy rate in Sistan and Balochistan was 36 per cent22 compared to 61 per cent in the whole country (Iran Statistical Yearbook 1370 1993: 123).
Another factor that has greatly strengthened the impact of Persian in Iranian Balochistan is the electrification of the province, which, although it had already started, was speeded up and almost totally completed shortly after the Islamic Revolution. Along with electricity came television. Many Baloch men, especially in the southern parts of Balochistan, spend at least some time as guest workers in the Gulf States, where they often purchase electronic goods, such as radio- and TV-sets, for their family in Balochistan. Television was first introduced in the provincial capital, Zahedan, in the 1960s, but nowadays television has reached even the most remote areas of Iranian Balochistan.
Television has been a major breakthrough in the introduction of Persian in Balochistan. By watching Persian programmes at an early age, often even before going to school, the children get acquainted with this language and learn to pronounce it with a Tehrani accent, something which is not true of most educated Baloch in the older generation, who generally speak Persian with a ‘Balochi accent’. One reason for that is that they were introduced to Persian at school, generally by local teachers, who themselves spoke Persian with a Sistani or Balochi accent. Several Baloch friends of mine from Iran remember that they felt ashamed to speak Persian at school, especially in front of Persian-speaking classmates who made fun of their accent.23

Structural Influence from Persian on Iranian Balochi 24

In Iranian Balochistan two major dialects are spoken, the Western (or Rakhshani) dialect and the Southern (or Makrani) dialect. Both these dialects are spoken on the other side of the Goldsmid line as well. There is also an area in Iranian Balochistan, Sarawan, where a very particular dialect, more profoundly influenced by Persian than other Balochi dialects, is spoken (see also Baranzehi 2003).
Nowadays it is quite obvious that the national language, Persian, is the socially and culturally dominant language, and that Balochi is the low-status vernacular. However, this has not always been the case, and the example of Sarawan proves that clearly. Within this area one or two centuries ago, Baloch tribesmen of high status in the local society lived side by side with immigrant peasants of Afghan or other Persian-speaking origins, who had come to Sarawan more recently than the Baloch (Spooner 1967: 56). Languages in contact can affect each other in different ways. Much depends on the relative status of the languages. Two or more languages of more or less equal status may be spoken side by side and mutually affect each other in terms of structure and lexicon without eradicating either one or the other language. This is called adstrate influence.
Another setting is when a dominant language, e.g. the language of a conquering group or the political elite, exercises influence on a dominated language, e.g. the language of a minority group. This type of influence is often called superstrate. Sometimes this term also implies that the final outcome of language contact is that the prestigious language is abandoned by the conquerors in favour of the local language, which, however, has been considerably influenced by that language. Such an outcome is more likely when a small number of conquerors seize political power in an area where a language other than their own is spoken, e.g. at the Norman conquest of Britain.
However, the term superstrate is also used in a broader sense to describe the influence on a low-prestige language when ‘another and more prestigious language which is imposed upon the speakers of the first, usually by conquest or political absorption…exercises an identifiable effect upon that first language’ (Trask 2000: 330). By this definition, Persian structural and lexical influence on minority languages in Iran could be termed superstrate influence. It is, however, very unlikely that local languages would replace Persian in
present-day Iran.
The term substrate is normally used for a language already spoken in an area or by a group of people ‘which has had a detectable effect upon the newly arrived one’ (Trask 2000: 329). As with superstrate, this term generally refers to a difference in status between the substrate language and the newly arrived language, where the substrate language is the low-status language. Likewise, it is often used to describe settings where this language has been replaced by the new language, in which it has left structural and lexical traits. As an example, Celtic traits in English could be mentioned.
The dialect of the central valley of Sarawan mentioned above is especially interesting to study from a contact linguistic perspective. In Sarawan, the non pastoral economy, mainly based on settled agriculture, has a longer tradition than in other parts of Iranian Balochistan. The same applies to education. Since education is in Persian, it considerably strengthens the Persian influence.
This, together with the immigration of Persian speakers to Sarawan in the past centuries, has made this dialect a very interesting object for studying linguistic contact. It seems that in former times, Balochi was the high-status language in Sarawan, since the immigrants gave up Persian for Balochi.25 However, a significant number of substrate phenomena26 from Persian can be found in Sarawani Balochi. These features are not encountered in any other dialect of Balochi, either in Iran or in Pakistan. Examples of such phenomena are the replacement of the genitive with the izāfa-construction for genitive attributes, and, on the whole, a case system very similar to that of Persian. Another structural reshaping is that all postpositions have changed into prepositions.
Today Persian is the unquestioned high-status language, and nowadays superstrate effects from Persian on Sarawani as well as on other Balochi dialects in Iran is heavy. Among superstrate phenomena can be mentioned syntactic constructions instead of morphological case marking (e.g. for the indirect object), adjectives placed after their nouns with the izāfa as the linker and Tense-Aspect-Mood (TAM) forms of the verb modelled on Persian constructions (e.g. the progressive present and past). Extensive use of Persian lexical items can also be observed in most dialects.

Is there a Future for the Balochi Language within the State of Iran?

As long as a language is used in all the domains represented in everyone’s daily life, or at least in the life of a majority of its speakers, it cannot be seen as threatened. But as soon as socio-economic changes take place, the traditional lifestyle and culture of an ethnie, and with that possibly its whole basis of identification, is threatened.27 When this happens among a minority group, like the Baloch, there are several courses the group, or members of the group, can take. They may seek full acceptance in the majority society and be ready to give up what is distinct and exclusive to their group, including the language, or they may see a need to defend their identity by demanding selfdetermination in some form or another. They may also strive to retain their identity as a separate ethnie within the present political framework. This approach is a realistic alternative only if adequate opportunities for the development of distinct languages, forms of worship and other components of the ‘ethnic’ cultures are given by the state authorities. In Iranian Balochistan there is a deeply rooted suspicion against the Persians,28 who are often derogatorily called gajar.29 The Balochi way of living, including its social organisation and its honour codes, differs considerably from that of the Persians. There have been several insurgencies against the direct rule from Tehran over Balochistan after 1928 (see e.g. Hosseinbor 2000: 141-164). After 1979, the different religious creeds have also been accentuated, and there is a strong feeling among the Baloch that due to their Sunni creed, they are not given adequate opportunities to advance in
the Islamic Republic, which is based on the Shi’a creed.
The Iranian Baloch also have strong links with the Baloch in Pakistan, Afghanistan and the Gulf States, e.g. by massive immigration to Karachi from Iran after the 1928 events, by tribal links to the Baloch in Afghanistan and by Iranian Baloch seeking employment as guest workers in the Gulf States. The choice to be incorporated into the majority society would mean giving up these links. It is therefore unlikely that the Baloch will seek full acceptance in the majority society, and if adequate measures are not taken for cultural (including linguistic) development within the present political framework, there will most likely be increased demands for political self-determination. Indeed, many Baloch intellectuals are genuinely concerned about the future of the Balochi language in Iran. With the introduction of education and modern socio-economic structures in Balochistan, the basis of the traditional society is being eradicated slowly but surely. However, there is also an awakening to the fact that if Balochi is to continue to be a vigorous language in Iranian Balochistan, the Baloch themselves must adopt a positive attitude towards the language and work for its development in the new domains as well, i.e. as a language of writing, and ultimately of education. Some steps have already been taken, such as the publication of bilingual (Persian-Balochi) journals and the more and more frequent arrangements of Balochi literary and cultural gatherings (see above). I have also been told that there is some lobbying for the introduction of Balochi as a subject at university level in Iranian Balochistan.30 These are all steps that have been taken several years ago in Pakistani Balochistan.
It is the state that forms policies towards language and culture and controls the administrative and educational systems. The demands of the people within the state have, of course, a strong bearing on the decisions of the ruling elite.
The present paper has tried to show that the more favourable conditions for the Balochi language in British India than those seen in Iran from the beginning of the twentieth century prepared the path for a new generation of educated Baloch to also use their language in reading and writing.31 It must, however, be stressed that even in Pakistan the Balochi cultural movement is facing enormous problems, including low readership, financial constraints and official suspicion. The unsuccessful launching of Balochi as a language of primary education in Pakistani Balochistan can also be seen as a serious backlash.
On the other hand, it was noted above that political strength is a changeable factor, and that at one time the Baloch tribes in Iranian Sarawan were a mighty force, who assimilated other ethnic groups such as Afghans and Sistanis, whereas now the Shi’a Muslims32 constitute the ruling elite in Iran. What will happen on the political scene in Iran and Pakistan in the future is hard to foresee, and one can do nothing but speculate about whether there will be new shifts in the political strengths between the Baloch and the present ruling elites.


1 The Baloch delegates opposed the demarcation already at the meeting of the border commission, arguing that both the western and the eastern parts of Balochistan ought to belong to the Balochi Khanate of Kalat.
2 Balochi is an Indo-European language of the Iranian branch, most closely related to Kurdish, Gilaki, Mazandarani, Talyshi and other north-western Iranian languages.
3 However, a small number of Baloch in Iran profess Shi’a Islam. There are also Shi’a Muslims among the Baloch in Pakistan (Breseeg 2001: 58). There are, furthermore, communities who profess the so-called zikri religion which developed out of Sunni Islam around 1500 A.D. As for language, many people who ethnically regard themselves as Baloch, especially in Punjab and Sindh, do not any longer speak Balochi, which means that there is a process of language shift under way among the Baloch. There are strong reasons to believe that this process will continue and probably even be speeded up as modern society with education etc. more and more strongly penetrates Balochistan.
4 For a thorough discussion of socio-economic structures in Balochistan, see e.g. Fabietti (1996).
5 See Titus (1998: 668), who discusses the differences in social structure between the Pashtuns and the Baloch, and finds that among the Baloch a hierarchical structure predominates and facilitates the incorporation of new elements into the tribe.
6 Cf. Orywal (1996), who describes the de-tribalisation process among the Baloch in Afghanistan.
7 The Brahuis speak a Dravidian language, and therefore are generally considered to have migrated to their present habitat from the south-east.
8 Persian had a long tradition as a language of writing, literature and administration. The Samanids, ruling from Bokhara over large areas of Central Asia and Khorasan in the ninth and tenth centuries, began using New Persian in addition to Arabic, the language of the Muslim conquerors, as an administrative language.
9 A smaller part is found within the borders of present-day Afghanistan.
10 E.g. Gilbertson, George W., The Balochi Language, A Grammar and Manual, Hertford 1923; Gilbertson, George W., English-Balochi Colloquial Dictionary, I-II, Hertford 1925; Marston, E. W., Grammar and Vocabulary of the Mekranee Beloochee Dialect, Bombay 1877; Mockler, E., A Grammar of the Baloochee Language, Henry S. King & Co., London 1877; Pierce, E., ‘A
Description of the Mekranee-Beloochee Dialect’, Journal of the Bombay Branch of the Royal
Asiatic Society, 11(1875): 31, Bombay 1876, pp 1-98.
11 In Dames, M. Longworth, Popular Poetry of the Baloches, I-II, published for the Folk-Lore Society by David Nutt, London 1907, and Dames, M. Longworth, A Text Book of the Balochi Language, Punjab Government Press, Lahore 1891.
12 Ulus was published regularly between 1961 and 1991. It was re-started in 1996, but the issues appear very irregularly.
13 These languages include e.g. Kurdish, Lori, Gilaki, Mazandarani and Balochi. Languages totally unrelated to Persian that are spoken in Iran are e.g. Arabic (a Semitic language) and Azerbaijani Turkish (a Turkic language).
14 Elfenbein (1966: 1) refers to broadcasts in Balochi from Zahedan.
15 For Azerbaijani, see Information about Kurdish was given
by Hashem Ahmadzadeh, Uppsala.
16 Oral communication with Padideh Pakpour, Uppsala, who spent the spring semester 2004 as a guest student at the University of Rasht.
17 This occurs within the Sunni mosques throughout Balochistan, which are not frequented by the Shi’ites. However, in the Masjed-e Makki ‘The Meccan Mosque’ in Zahedan the sermon is in Persian due to the fact that this mosque is also frequented by other Sunnis than the Baloch (recent information obtained from Iran).
18 It was, in fact, observed in Zahedan, the provincial capital in the Iranian province of Sistan and Balochistan, that some families chose to speak Persian to their children rather than Balochi. They argued that this would prepare their children better for starting school, since the child would otherwise in school meet a new language that he/she was not familiar with, something that could hamper the learning process and put the child at a disadvantage compared to classmates who already knew Persian. See also Farrell (2000: 25), who notes that in Balochi tuition centres in Karachi, the aim of which is to provide supplementary education to the children of their community, Urdu is used as the teaching language, since the goal is ‘academic advancement of their pupils rather than any concern for language issues’. For a comprehensive study on language attitudes among university students in Pakistan, see Mahboob (2002). Only
10 per cent of the respondents to the question ‘Should your first language (other than Urdu) be the medium of instruction for primary education’ give a positive answer, whereas the same figure for Urdu is 63.1 per cent and for English as high as 76 per cent (Mahboob 2002: 30).
19 E.g. when the nomadic lifestyle is replaced by a settled way of life.
20 74 per cent for females and 85 per cent for males.
21 57 per cent for females and 79 per cent for males.
22 25 per cent for females and 46 per cent for males. Compare this to Kurdistan, with a 39 per cent literacy rate (23 per cent for females and 54 per cent for males).
23 The power of television exceeds by far the power of radio. Since it is both audial and visual, it makes a much greater impact than only audial media. The power of television in making a whole generation acquainted with a language can also clearly be seen in Sweden, where many children in the younger generation who grow up with American serials subtitled in Swedish start speaking English before they are even able to read and write Swedish. Procházka also stresses the impact of television as one of the key factors behind the ongoing language shift from Arabic to Turkish among speakers of Arabic dialects in Turkey (Procházka 1999: 124).
24 The purpose of a research project carried out by the present author, funded by the Swedish Research Council between 1998 and 2002, was to describe the linguistic interaction between Persian and Balochi in Iranian Balochistan. For further information on phonological, morphosyntactic and lexical influences from Persian on Balochi both in historical and modern times, see Jahani and Korn (2003), particularly the articles written by Baranzehi, Farrell, Jahani, Korn, Mahmoodi Bakhtiari and Mahmoodzahi.
25 This was my assumption at the time when I wrote the article. However, new information on Sarawan (obtained during a field journey to Iran in February-March 2005) gives a different picture. The Baranzahi/Barakzahi khans of Sarawan were of Afghan origin, thus not speakers of Balochi. However, in order to be able to communicate with their subjects, they acquired Balochi, retaining the Persian grammatical structures referred to in the article as substrate phenomena.
26 I am broadening the use of this term here. Normally it is used to describe the influence from an indigenous low-prestige language on a newly arrived language. Here, I use it to denote the influence of any language of low status (in this case Persian spoken by an immigrant group who had settled in the central valley of Sarawan) on a more prestigious language (here Balochi, the language of the ruling elite, which was also gradually adopted by the immigrants).
27 Religious beliefs may be re-evaluated (secularisation), and the language may be threatened, especially in a minority setting, based on the fact that new language domains (e.g. education and official administration in a language other than the minority language) enter into the life of the members of the ethnie.
28 This is also the case with other Shi’ites, e.g. the Azerbaijani Turks.
29 This term comes from the Qajar dynasty, who ruled Iran between 1796 and 1925. It was during this dynasty that Tehran became the capital. Military campaigns from this new capital were launched towards Balochistan to subdue the region, something which created strong hatred towards the ‘gajars’ among the Baloch.


Algar, Hamid. (1980). Constitution of the Islamic Republic of Iran (Berkeley, Mizan Press).
Baloch, Inayatullah. (1987). The Problem of ‘Greater Baluchistan’: A Study of Baluch Nationalism (Stuttgart, Steiner Verlag Wiesbaden).
Baranzehi, Adam Nader. (2003). ‘The Sarawani Dialect of Balochi and Persian Influence on It’, in C. Jahani and A. Korn (eds), The Baloch and Their Neighbours: Ethnic and Linguistic
Contact in Balochistan in Historical and Modern Times (Wiesbaden, Reichert Verlag), pp 75-
Breseeg, Taj Mohammad. (2001). Baloch Nationalism: Its Origin and Development up to 1980 (London, unpublished Ph.D. thesis, School of Oriental and African Studies).
30 The two state universities in Iranian Balochistan are located in Zahedan and Iranshahr. There are also ‘free’ (i.e. private) universities in Balochistan. In view of the fact that such permissions have been granted for Azerbaijani and Kurdish (see above), it is likely that this will happen also for Balochi in a not too distant future.
31 This is not necessarily due to a genuine concern on the part of the British administration for the Balochi language. They just applied a different policy towards local languages in order to maintain and strengthen their rule than that employed by the Pahlavi monarchs.
32 These are mainly the ethnic Persians and Azerbaijanis. Among the Azerbaijanis, there is an ambivalent attitude towards the Persian cultural dominance. Many seek full acceptance in the Persian culture even to the degree that they feel ashamed of their background which they associate with a rural and ‘backward’ way of living. Others are proud of being Azerbaijanis, i.e. speakers of a Turkic language, and want to retain this language and transfer it to their children.
For them, there is the choice of trying to acquire cultural rights within the present system or to struggle for independence. The fact that there is a fear of Azerbaijani attempts at selfdetermination can be seen in, e.g., Iran’s official support for the Christian Republic of Armenia in its conflict with Azerbaijan.
Bruce, Richard. (1900). The Forward Policy (London etc., Longmans, Green, and Co.).
Dames, M. Longworth. (1904). The Baloch Race: A Historical and Ethnological Sketch (London, Royal Asiatic Society).
Dashtyari, Saba. (2003). ‘Periodicals in Balochi: A Brief Description of Balochi Printed Media’, in C. Jahani and A. Korn (eds), The Baloch and Their Neighbours: Ethnic and Linguistic Contact in Balochistan in Historical and Modern Times (Wiesbaden, Reichert Verlag), pp 321-
Elfenbein, Josef. (1966). The Baluchi Language: A Dialectology with Texts (London, Royal Asiatic Society of Great Britain and Ireland).
Eriksen, Thomas H. (1993). Ethnicity and Nationalism: Anthropological Perspectives (London, Pluto Press Limited).
Fabietti, Ugo. (1996). ‘Equality versus Hierarchy: Conceptualizing Change in Southern Balochistan’, in P. Titus (ed), Marginality and Modernity: Ethnicity and Change in Post- Colonial Balochistan (Karachi etc., Oxford University Press), pp 3-27.
Farrell, Tim. (2000). ‘Mother Tongue Education and the Health and Survival of the Balochi Language’, in C. Jahani (ed), Language in Society – Eight Sociolinguistic Essays on Balochi
(Uppsala, Acta Universitatis Upsaliensis), pp 19-32. Hosseinbor, Mohammad Hassan. (2000). Iran and its Nationalities: the Case of Baluch
Nationalism (Karachi, Pakistani Adab Publications).
Iran Statistical Yearbook 1370. (1993). (Tehran, Statistical Center of Iran).
Iran Statistical Yearbook 1377. (2000). (Tehran, Statistical Center of Iran).
Jahani, Carina and Korn, Agnes (eds). (2003). The Baloch and Their Neighbours: Ethnic and
Linguistic Contact in Balochistan in Historical and Modern Times (Wiesbaden, Reichert
Mahboob, Ahmar. (2002). ‘“No English, No Future!”: Language Policy in Pakistan’, in S. G.
Obeng and B. Hartford (eds), Political Independence with Linguistic Servitude: The Politics about Languages in the Developing World (New York, Nova Science Publishers, Inc.), pp 15-
Mojab, Shahrzad, and Hassanpour, Amir. (1995). ‘The Politics of Nationality and Ethnic Diversity’, in S. Rahnema and S. Behdad (eds), Iran after the Revolution (London and New York, I.B. Tauris), pp 229-250.
Orywal, Erwin. (1996). ‘Periphery and Identity: Process of Detribalization among the Baloch of
Afghanistan’, in P. Titus (ed), Marginality and Modernity: Ethnicity and Change in Post- Colonial Balochistan (Karachi etc., Oxford University Press), pp 78-102.
Procházka, Stephan. (1999). ‘From Language Contact to Language Death: The Example of the Arabic Spoken in Cilicia (Southern Turkey)’, Orientalia Suecana 48, pp 115-125.
Smith, Anthony D. (1986). The Ethnic Origins of Nations (Oxford, Basil Blackwell).
Spooner, Brian. (1967). ‘Notes on the Baluchī Spoken in Persian Baluchistan’, Iran, Journal of the British Institute of Persian Studies 5, pp 51-71.
Spooner, Brian. (1989). ‘Baluchistan, I: Geography, History, and Ethnography’, in E. Yarshater (ed), Encyclopaedia Iranica, III (London and New York, Routledge and Kegan Paul), pp 598- 632.
Titus, Paul. (1998). ‘Honor the Baloch, Buy the Pushtun: Stereotypes, Social Organization and History in Western Pakistan’, Modern Asian Studies 32/3 (Cambridge, Cambridge University Press), pp 657-687.
Trask, Robert L. (2000). The Dictionary of Historical and Comparative Linguistics (Edinburgh, Edinburgh University Press).

%d bloggers like this: