CHAPTER 2
India, Multan, Afghanistan, and Central Asia:
Reconstructing the Lyuli/Mughat Migration Problem
From the proposed monograph:
The Lyuli (Mughat) of Uzbekistan: Language, Ritual, Memory and the Search for South Asian Connections
Dr. Manish Kumar C. Mishra
2.1 Introduction: from origin stories to an evidence-based migration history
The question of where the Lyuli/Mughat came from is among the most frequently repeated—and least securely resolved—questions in the literature on Central Asian minority communities. South Asian origin has often been asserted through a combination of names, legends, linguistic fragments, occupational parallels, and analogies with Roma or other mobile populations. Yet these forms of evidence do not possess equal historical value. A migration history must therefore be reconstructed in layers rather than narrated as a single uninterrupted journey from “India” to Uzbekistan.
This chapter asks a narrower question: what can presently be established about the geographical and chronological stages that may connect the Mughat/Lyuli with the Indian subcontinent? It separates four evidentiary fields: medieval Persian literary traditions concerning Luri/Luli; the ethnonym Multoni/Multani and its apparent reference to Multan; direct evidence for Lyuli in Central Asia; and linguistic classifications linking some Central Asian in-group varieties with South Asian or Afghan populations.
The conclusion is deliberately provisional. The available scholarship makes a South Asian connection historically plausible and, in some formulations, highly probable. But it does not yet permit a single precise route, date, or homogeneous ancestral population to be reconstructed for all communities subsequently called Lyuli.
2.2 The oldest narrative: Bahram Gur and musicians from India
One of the most influential origin narratives appears in the Persian historical-literary tradition surrounding the Sasanian ruler Bahram Gur. Later accounts tell of musicians or entertainers brought from India and subsequently dispersed. Marushiakova and Popov review the long academic tradition that connected the Luri/Luli of these narratives with later Central Asian Lyuli, while also warning that the narrative was recorded centuries after the events it purportedly describes.
This warning is methodologically decisive. A medieval literary tradition may preserve a historical memory, but it cannot be treated as a contemporary migration register. The Bahram Gur narrative is therefore important evidence for the existence of a Persianate memory linking groups called Luri/Luli with India. It is not, by itself, proof that every modern Mughat family descends directly from the people described in that story.
Later Persian and Arabic writers repeated or elaborated the narrative. The persistence of the India–Luli association is historically significant because it shows that the connection was not invented in modern nationalist discourse. Yet repetition can also reproduce literary convention. The task is therefore to distinguish the history of the tradition from the history of the population.
2.3 Luli in Persianate memory: literature as evidence and as problem
Marushiakova and Popov note references to Luli in medieval Persian and Arabic sources and in Persian poetry. These references establish that Luli was a recognisable social-cultural designation within the Persianate world. They also demonstrate that the category carried associations with performance, mobility, marginality, and particular occupations.
However, literary representation is not equivalent to ethnographic description. Poets may employ a social label metaphorically; chroniclers may reproduce inherited stereotypes; and a name may shift meaning across centuries. Consequently, the historical method adopted here asks three questions of every reference: Does the source describe an identifiable population? Does it provide a location? Does it contain information independent of literary convention?
This distinction becomes crucial when tracing a path toward Central Asia. The fact that Luli occur in Persian literature strengthens the historical depth of the ethnonym, but it does not yet demonstrate continuity between all medieval Luli and modern Mughat.
2.4 Multoni/Multani: the most geographically specific clue
The designation Multoni or Multani deserves special attention because it is geographically more specific than the general term Lyuli. Marushiakova and Popov record its use in parts of Uzbekistan and historically in Samarkand and Surkhandarya, explaining it with reference to the city of Multan in medieval India, now Pakistan. They also describe an older social division between sedentary craft-associated Kasib/Kosib and more mobile Multoni/Multani.
Multan was a major urban, commercial, religious, and strategic centre linking the north-western subcontinent with routes toward Afghanistan, Iran, and Central Asia. For that reason, an ethnonym meaning “from Multan” would fit a historically plausible geography of transregional mobility. But plausibility is not proof.
The strongest test will be convergence. If Multoni is shown in early documents to designate the same population later called Mughat; if oral traditions independently remember Multan or neighbouring regions; and if the in-group lexicon contains systematic correspondences with Punjabi, Saraiki, Sindhi, or other languages of the north-western subcontinent, then the Multan hypothesis would gain substantial evidentiary force. Until that comparison is completed, Multoni should be classified as a strong geographical clue rather than a proven point of origin.
2.5 Afghanistan as corridor, contact zone, or source?
Any serious reconstruction of South Asia–Central Asia migration must consider Afghanistan not merely as a line on a map but as a historical contact zone. Routes linking Punjab, Sindh, Multan, Kabul, Balkh, Badakhshan, and Transoxiana carried merchants, soldiers, pilgrims, craftsmen, religious specialists, entertainers, and mobile communities in multiple directions.
The linguistic literature strengthens the importance of this zone. Encyclopaedia Iranica’s survey of so-called Gypsy dialects distinguishes several Central Asian groups and explicitly relates some to populations in Afghanistan. Jugi/Mugat in the Hissar Valley is described as Tajik-based and related to Jogi populations of Afghanistan and northern Iran; other groups show Afghan-Persian, Pashto, or Indic-based features. This does not establish that all such groups shared one ancestry. Rather, it demonstrates that Afghanistan formed part of a larger network within which mobile populations, argots, and social labels circulated.
The book will therefore avoid a simplistic arrow—India → Afghanistan → Uzbekistan—as though a single migration occurred once. Afghanistan may have functioned at different times as corridor, temporary homeland, linguistic contact zone, and source of later migration.
2.6 The first direct Central Asian evidence and the Baburnama problem
Specialist scholarship identifies the Baburnama of Zahir al-Din Muhammad Babur as containing the first direct evidence for Lyuli in Central Asia. This is potentially a major chronological anchor because Babur’s own world connected Fergana, Samarkand, Kabul, and northern India.
At the present stage, however, this book will not reproduce an exact Baburnama quotation until the passage has been checked against an authoritative edition and translation. Secondary scholarship is sufficient to establish that Marushiakova and Popov treat the Baburnama as the first direct Central Asian evidence, but a monograph of this kind must verify the exact wording, context, personal name, and location before drawing further conclusions.
This is an example of the source-critical principle that will govern the book: a secondary claim may identify where evidence exists, but the primary text should be consulted before it becomes the foundation of a historical argument.
2.7 From medieval presence to Central Asian rootedness
By the period for which the Mughat/Lyuli are more securely documented, they are not simply travellers passing through Central Asia. Scholarship describes them as an integral part of Central Asian life, living in towns and villages primarily in Uzbekistan and Tajikistan and also in neighbouring states. This long regional presence complicates any language of “foreignness.”
A community may have a migration history extending beyond Central Asia and still be deeply Central Asian by historical experience. Indeed, the very persistence of Tajik-based speech, Uzbek bilingualism, Sunni Muslim practice, local clan structures, and regionally specific occupations demonstrates adaptation across generations.
Thus the historical question is not whether Lyuli are “Indian or Uzbek/Tajik.” Such a binary is analytically misleading. The relevant question is how a population with possible South Asian historical connections became socially, linguistically, and culturally embedded in Central Asia.
2.8 Migration is rarely a single event
One of the central weaknesses in popular accounts of Lyuli origins is the assumption of a single departure date from India. Mobile communities can form through repeated movements, fission, incorporation of outsiders, occupational alliances, marriage, and language shift. A name can travel farther than a lineage, while a professional argot can circulate between communities without large-scale population movement.
The evidence therefore permits several migration models. One is a single-origin model in which an ancestral South Asian population migrated through Iran or Afghanistan and diversified in Central Asia. A second is a multiple-stream model in which different South Asian-connected populations entered Central Asia at different times and were later grouped under similar labels. A third is a convergence model in which populations of partly different origins acquired common social classifications and overlapping argot traditions within the Persianate and Central Asian environment.
At present, the multiple-stream and convergence models deserve serious consideration because specialist linguistic classifications reveal considerable heterogeneity among groups historically placed under broad “Gypsy” categories.
2.9 What language can contribute to migration history
Historical linguistics can test migration narratives more rigorously than visual resemblance or occupational analogy. If the Mughat in-group vocabulary contains an inherited Indic substrate, its geographical affinities may help identify the South Asian zone from which some ancestors came.
The key is systematic correspondence. A handful of words resembling Hindi would be weak evidence. A cluster of basic vocabulary showing regular phonological correspondences with Punjabi, Saraiki, Sindhi, Rajasthani, Haryanvi, or another Indo-Aryan group would be much stronger. Equally important would be negative evidence: if supposedly “Indian” terms prove to be Persian, Tajik, Uzbek, Abdoltili, or widely circulating argot words, the South Asian linguistic case would need to be narrowed.
The next major linguistic chapters will therefore test migration hypotheses rather than merely illustrate them.
2.10 Parya/Tajuzbeki as a comparative control
Parya is especially useful because its South Asian linguistic ancestry is demonstrable. Encyclopaedia Iranica separates Parya-type Indic-based varieties from Tajik-based Jugi/Mugat and other Central Asian argots. Bholanath Tiwari’s Tajuzbeki corpus provides an Indian scholarly documentation of the Parya language in the Tajik-Uzbek zone.
This comparison prevents circular reasoning. If a lexical method identifies Indo-Aryan inheritance clearly in Parya but finds only a small or specialised Indic layer in Mughat speech, the two communities likely experienced different linguistic histories. If, however, Mughat basic vocabulary reveals systematic correspondences with Parya or a specific north-western Indo-Aryan cluster, the possibility of deeper historical connection would deserve renewed investigation.
Parya is therefore not evidence that Lyuli are Indo-Aryan speakers; it is a control against which the strength of Lyuli linguistic evidence can be measured.
2.11 An evidence-weighted migration model
The present evidence can be arranged by confidence rather than forced into a single narrative. The long Persianate association of Luli with India is historically old but partly literary. The Multoni/Multani ethnonym is geographically suggestive and potentially important. Direct Central Asian presence is securely established by the early modern period in specialist scholarship. Long residence in the Tajik-Uzbek cultural zone is unquestionable. The exact South Asian homeland, departure date, number of migration streams, and relationship among Mughat, Parya, Roma, Dom, Jogi, and other mobile communities remain open.
This produces a cautious model:
Possible South Asian source region(s)
→ Persianate literary memory and/or movement through Iran-Afghanistan
→ differentiated migrations and contacts
→ established Central Asian communities
→ Tajik/Uzbek linguistic and cultural adaptation
→ modern Mughat/Lyuli identities.
The arrows are not equal. Some represent documented historical presence; others represent hypotheses that require linguistic or archival corroboration.
2.12 Research questions generated by the migration evidence
The migration problem can now be converted into a series of testable questions. Does the name Multoni occur in records early enough to predate modern speculation about Indian origins? Do Mughat oral histories name Multan, Punjab, Sindh, Afghanistan, or particular Central Asian staging points? Does clan nomenclature preserve geographical memory? Are there regional differences between Samarkand, Bukhara, Surkhandarya, Fergana, and Tajik communities that correspond to different migration histories? Does in-group vocabulary point toward one South Asian linguistic zone or several? Are words regarded as Indic actually inherited, or did they circulate through professional argots?
These questions require different sources. No single archive can answer them all. Historical texts, Soviet ethnography, Uzbek and Tajik scholarship, linguistic corpora, oral history, and contemporary fieldwork must be brought into dialogue.
2.13 Conclusion: what can be said before the lexical evidence is tested
The migration history of the Lyuli/Mughat cannot responsibly be reduced to a single sentence such as “they came from India.” The available evidence is both richer and more complicated.
A South Asian connection has deep roots in Persianate traditions concerning Luri/Luli and is reinforced by the geographically suggestive Multoni/Multani designation. Specialist scholarship places Mughat/Lyuli firmly within the long history of Central Asia and identifies direct evidence by the early modern period. Linguistic surveys further show that the region contained several mobile or marginal communities with different language bases—Tajik, Persian, Afghan-Persian, Pashto-influenced, and Indic—making a single homogeneous migration model unlikely.
The most defensible conclusion at this stage is therefore that the Mughat/Lyuli history probably belongs to a wider field of South Asia–Afghanistan–Iran–Central Asia mobility, but the precise ancestry, route, and chronology of the focal community remain to be demonstrated. The name Multoni may preserve an especially important geographical memory; Afghanistan was likely a crucial zone of movement and contact; and Central Asia became not merely a destination but the setting in which modern Mughat identity was historically formed.
The decisive next step is linguistic. Chapter 3 will therefore turn to the public and private linguistic repertoire of the Lyuli/Mughat, asking what is meant by “Mughat language,” “secret language,” “argot,” and Tajik-based community speech. Only after establishing those categories can the book undertake the word-by-word Indic lexical analysis that will test the migration hypotheses developed here.
Table 2.1. Migration claims and their present evidentiary status
Figure 2.1. Evidence-weighted migration model
Possible South Asian source region(s)
↓
Multan / north-western subcontinent? [hypothesis]
↓
Afghanistan–Iranian contact corridor [plausible regional context]
↓
Central Asian presence by early modern period [strong]
↓
Long-term Tajik–Uzbek adaptation [established]
↓
Modern Mughat/Lyuli communities
Note: The diagram distinguishes levels of evidence; it is not intended as a proven single migration route.
References cited in Chapter 2
Marushiakova, Elena, and Vesselin Popov. 2016. Gypsies of Central Asia and the Caucasus. Cham: Palgrave Macmillan. DOI: 10.1007/978-3-319-41057-9.
Marushiakova, Elena, and Vesselin Popov. 2015. ‘Central Asian Gypsies: Identities and Migrations’. Sprawy Narodowościowe / Nationalities Affairs, no. 47. DOI: 10.11649/sn.2015.031.
Windfuhr, Gernot L. 2002. ‘Gypsy ii. Gypsy Dialects’. Encyclopaedia Iranica. Comparative classification of Jugi/Mugat, Luli/Multani and other Central Asian varieties.
Ozierski, Przemysław. 2020. ‘Central Asian Gypsies – Lyuli: The Overview of Current Socio-Economic Problems’. Review of Nationalities, no. 3.
Tiwari, Bholanath. 1970. Tajuzbeki: Soviet Sangh mein Boli Jane Vali Hindi Boli: Aitihasik aur Tulnatmak Adhyayan tatha Sankshipt Shabdkosh. Delhi: National Publishing House.
Babur, Zahir al-Din Muhammad. Baburnama. Primary passage concerning Lyuli to be verified against an authoritative edition before quotation in the final monograph.
Source-critical note
This chapter deliberately distinguishes between primary historical evidence and claims presently available through specialist secondary scholarship. The Bahram Gur/Luri tradition is treated as a historical literary tradition rather than as a literal migration record. The Baburnama reference is not quoted because the exact passage has not yet been verified against an authoritative primary edition. The Multoni–Multan relationship is treated as a research hypothesis requiring independent linguistic and historical corroboration. These distinctions will be retained in the final book.
No comments:
Post a Comment
Share Your Views on this..