Tajuzbeki/Parya and the Lyuli of Central Asia:
Language, Migration, Identity, and the Problem of Indic Origins in the Tajik-Uzbek Cultural Zone
Dr. Manish Kumar C. Mishra
Associate Professor, Department of Hindi
K. M. Agrawal College of Arts, Commerce and Science, Kalyan (West), Maharashtra, India
Abstract. Bholanath Tiwari’s Tajuzbeki (1970) is a rare Indian-language monograph on an Indo-Aryan speech community in Soviet Central Asia. Later linguistic scholarship identifies the same language as Parya, concentrated historically in the Hissar Valley of Tajikistan and adjoining Surkhandarya districts of Uzbekistan. The region is also home to communities described in ethnographic literature as Lyuli/Luli, Jugi, Multoni and, in some scholarship, by the self-designations Mugat or Mugati Tubjoni. Their geographical proximity and traditions of southern or Indian-connected origins invite comparison, but also create a danger of collapsing distinct peoples into a single “Central Asian Gypsy” category. This article combines close reading of the supplied scan of Tiwari’s primary text with later Parya linguistics and Lyuli/Mugat ethnography. It concludes that Tajuzbeki is securely identifiable with Parya and that Parya’s Indo-Aryan genealogy is demonstrable. By contrast, current evidence does not establish that Lyuli/Mugat and Parya are the same population, nor that one is simply a language-shifted branch of the other. The more defensible hypothesis is that Central Asia received multiple South Asian-connected migration streams that later converged socially and geographically within Tajik-Uzbek environments. The article proposes an interdisciplinary research programme combining comparative linguistics, archival history, oral traditions, GIS settlement mapping and community-centred ethnography.
Keywords: Tajuzbeki; Parya; Lyuli; Luli; Mugat; Bholanath Tiwari; Indo-Aryan; Uzbekistan; Tajikistan; Surkhandarya; Hissar Valley; migration; language contact.
Figure 1. Title page of Bholanath Tiwari’s Tajuzbeki, defining the work as a historical and comparative study with a concise dictionary. Source: Tiwari (1970), preliminary title page; supplied scan.
1. Introduction: replacing civilizational rhetoric with testable evidence
India-Central Asia relations are often described through the language of “ancient” or “civilizational” connections. Such formulations may be diplomatically useful, but historical scholarship requires narrower propositions: which people moved, when, through what routes, what evidence survives, and how confidently can it be interpreted? Minority languages offer an unusually powerful archive because inherited grammar may preserve population history even after political boundaries and collective memories change.
The problem examined here begins with two bodies of evidence. The first is linguistic: Bholanath Tiwari’s Tajuzbeki, published in Delhi in 1970, documents a speech form encountered in Soviet Central Asia that he regarded as closely related to Hindi and neighbouring Indo-Aryan varieties. Later scholarship calls the language Parya. The second body is ethnographic: communities known in Uzbekistan, Tajikistan and Kyrgyzstan as Lyuli/Luli and by other names, including Jugi and Multoni, while some scholarship records Mugat or Mugati Tubjoni as self-designations (Ozierski 2014/2020).
The central question is therefore not whether both can be called “Indian-origin communities,” but whether the available linguistic, historical and ethnographic evidence demonstrates a genealogical relationship between them. The method adopted here is deliberately asymmetric: strong conclusions are permitted only where independent evidence converges; attractive but unsupported connections are retained as hypotheses.
2. Tiwari’s Tajuzbeki as a primary source and a history of field research
Tiwari’s preface records that he arrived in Tashkent in 1962 as a professor of Hindi and learned of a local speech form that appeared unexpectedly close to Indian languages. His first comparisons involved numerals and everyday vocabulary; he also noticed strong Tajik influence (Tiwari 1970: 9-12). The inquiry therefore began from observed linguistic correspondences rather than from a predetermined ethnic theory.
The book reproduces official correspondence that gives the research an unusually verifiable institutional history. Letters involving the Embassy of India in Moscow and Soviet authorities concern permission and arrangements for visits to relevant districts of the Tajik SSR (Tiwari 1970: 13-15). Tiwari also acknowledges I. M. Oransky and other Soviet scholars, informants and students. Tajuzbeki is consequently a product of Indo-Soviet academic exchange as well as an individual work of Hindi linguistics.
The book’s architecture is empirical: historical introduction, sound and grammar, a Tajuzbeki-Hindi dictionary and oral/folkloric materials. The supplied scan is incomplete, ending during the dictionary at printed p. 134, while library catalogues record the original volume as xvii + 201 pages. This article therefore uses page-specific claims only where visible in the supplied primary source and does not reconstruct missing material.
Figure 2. Embassy of India/Moscow correspondence reproduced by Tiwari, documenting official facilitation of research access. Source: Tiwari (1970: 15; supplied scan).
3. Tajuzbeki = Parya: the strongest identification in the evidence
The identification of Tiwari’s Tajuzbeki with Parya is supported by converging evidence. Tiwari places his speakers in Tajikistan and Uzbekistan; Tatiana Oranskaia’s Great Russian Encyclopedia entry places Parya in the Hissar Valley and in districts of Uzbekistan’s Surkhandarya region bordering Tajikistan. Tiwari finds an Indo-Aryan grammatical and lexical base with substantial Tajik influence; later scholarship independently classifies Parya as Indo-Aryan and treats Tajik bilingualism as structurally consequential (Oranskaia 2014; Oranskaia 2025).
The nomenclature differs because Tiwari coined “Tajuzbeki” from Tajik + Uzbek to reflect the geographical distribution known to him, whereas later international scholarship standardised Parya. The difference is historiographical, not evidence for two separate languages. In this article “Tajuzbeki” is retained when discussing Tiwari’s corpus and “Parya” for the broader linguistic literature.
This conclusion is strengthened by later sociolinguistic work. Abbess et al. (2010) describe Parya communities in Tajikistan living among Tajik and Uzbek populations and identify the language as a central marker of Parya ethnic identity. Oranskaia (2025) demonstrates how long-term Tajik/Dari bilingualism affected the formation of Parya pronominal clitics. The later research thus confirms both inheritance and contact: Parya is Indo-Aryan, but it is also historically Central Asian.
4. Linguistic ancestry and the limits of the label “Hindi dialect”
Tiwari’s subtitle describes Tajuzbeki as a Hindi speech variety. Read through contemporary standard-language ideology, the phrase can be misleading. His actual comparisons range across Hindi, Haryanvi, Punjabi, Braj, Rajasthani, Urdu, Tajik and Uzbek. He is locating the language within a wider Indo-Aryan continuum, not demonstrating that the speakers used codified Modern Standard Hindi.
The strongest modern formulation is therefore genealogical: Parya/Tajuzbeki is an Indo-Aryan language with significant affinities to north-western and central Indo-Aryan varieties historically discussed in relation to Western Hindi, and it has been transformed by prolonged Tajik and Uzbek contact. The distinction matters because shared inherited vocabulary is not the same as borrowing from Hindi, and structural ancestry is stronger evidence than isolated lexical resemblance.
This reinterpretation does not diminish Tiwari. It clarifies what was genuinely remarkable about his discovery: an Indo-Aryan linguistic system had survived far outside the principal South Asian Indo-Aryan geographical continuum.
5. Migration history: what is established and what remains conjectural
Tiwari records traditions connecting the community with Laghman in Afghanistan and reproduces his attempt to investigate the claim through Afghan diplomatic channels (Tiwari 1970: 16-17). He then proposes a deeper South Asian origin, relating the language to an area near the Rajasthan-Haryana-Punjab interface and suggesting movement through Punjab and Afghanistan before settlement in Central Asia (Tiwari 1970: 17-18). His approximate fifteenth-century chronology is explicitly tentative.
Oranskaia’s encyclopedia entry independently records local names for Parya meaning “Afghan language” and “language of the Laghmanis,” and states that the Parya came from Laghman to their present territory. This makes the Afghan intermediate stage considerably more plausible than a precise date of departure from India. The deeper homeland remains a comparative-linguistic hypothesis requiring renewed testing.
The evidence should therefore be ranked: Indo-Aryan ancestry is strong; the Hissar-Surkhandarya settlement zone is strong; an Afghan/Laghman stage is plausible and multiply attested; a precise north Indian locality and c. 1400 migration date remain unproven.
Figure 3. Evidence-led migration/contact model for Parya/Tajuzbeki. Source: author’s analytical diagram based on Tiwari (1970: 16-18) and Oranskaia (2014). The deeper homeland and chronology remain hypotheses.
6. Who are the Lyuli/Luli/Mugat?
The Lyuli evidence is structurally different from the Parya evidence. Ozierski’s study describes a Central Asian ethnic group “known by many names,” recording Mugat or Mugati Tubjoni as self-designations and Lyuli, Jugi, Multoni, Mazang and Tavoktarosh among names used by others. The communities are found primarily in Tajikistan, Uzbekistan and Kyrgyzstan and have experienced persistent social marginalisation as well as Soviet-era integration and post-Soviet migration pressures (Ozierski 2014/2020).
This multiplicity of names is analytically important. “Lyuli” is not simply the name of a single documented Indo-Aryan language in the way that Parya is. It is an ethnographic and social category whose historical usage may encompass internally differentiated communities. Scholarship and journalism have often translated or compared such groups to “Gypsies,” but this outsider category can obscure local identities and should not be treated as proof of Romani identity or of common ancestry with Parya.
Recent Uzbek ethnographic work continues to study Lyuli family ritual, migration and cultural adaptation as a distinctive field of inquiry (Ruziyeva 2026). Such research reinforces the need to analyse Lyuli history on its own terms rather than to make it an appendix to Parya linguistics.
7. Visual ethnography and the ethics of representation
Photographs of Lyuli families in Uzbekistan are valuable as records of contemporary social life, but they cannot prove ancestry. The photograph included here, credited by its published source to Aleksandr Barkovsky, shows women and children in Uzbekistan and has been circulated in reporting on the social margins of Lyuli life. It should be read as visual ethnography of a contemporary community, not as an illustration of what medieval migrants looked like.
This distinction is essential because marginalised communities are often represented through poverty, mobility or exoticised dress. A scholarly article should avoid turning such images into ethnic “types.” The evidential function of the photograph here is limited and explicit: it establishes the contemporary human presence behind the ethnographic category and reminds the reader that questions of naming, identity and historical origin concern living communities.
Figure 4. Lyuli women and children in Uzbekistan. Photograph credited by the published source to Aleksandr Barkovsky. Source: Global Voices/Adinkra, “Life on the margins: The Lyuli people of Uzbekistan” (image retrieved 12 August 2026). Used here as contemporary visual ethnography, not as evidence of ancestry.
8. Geography: overlap is significant, but not genealogical proof
Parya is securely associated with the Hissar Valley and adjoining Surkhandarya districts. Lyuli/Mugat communities occur across the same wider Tajik-Uzbek cultural space. This overlap makes comparison legitimate, especially because the borderlands have long been multilingual and socially mobile. Yet the same geography can host unrelated minorities, and shared residence can result from separate migrations.
Four historical mechanisms could produce the observed proximity: common ancestry followed by differentiation; related but separate South Asian migrations; interaction between already distinct communities after arrival; or outsider classification of socially comparable minorities under overlapping labels. Geography alone cannot choose among these mechanisms.
The crucial research task is therefore to connect settlement history to linguistic and archival evidence. If specific Parya and Lyuli settlements can be shown to have shared marriage networks, occupational associations, migration memories or historical administrative classifications, the relationship would become testable rather than impressionistic.
9. The decisive asymmetry: language
The strongest present argument against simply identifying Parya with Lyuli is linguistic asymmetry. Parya preserves a demonstrable Indo-Aryan language. Contemporary Lyuli communities are generally described through Tajik- and Uzbek-speaking environments; the sources examined here do not demonstrate that “Lyuli” as a whole preserves the Parya linguistic system.
This does not prove that no deeper relationship exists. Language shift can erase an ancestral language while social identity persists. A branch of a population might adopt Tajik or Uzbek over generations. But that possibility is precisely a hypothesis requiring evidence. It cannot be used circularly: one cannot first assume that Lyuli are language-shifted Parya and then cite their lack of Parya speech as evidence of language shift.
For this reason, comparative kinship vocabulary, inherited numerals, core verbs, phonological correspondences and historical word lists would be especially valuable. If Lyuli-specific lexical material reveals systematic correspondences with Parya beyond shared Tajik or Uzbek vocabulary, the common-origin hypothesis would become substantially stronger.
10. Three historical models
Model A, the common-origin model, proposes that Parya and Lyuli descend from one South Asian ancestral population but underwent different linguistic histories. Current evidence is insufficient to establish this.
Model B, the parallel-migration model, proposes multiple South Asian-connected populations moving through Afghanistan or related corridors into Central Asia at different periods. This is compatible with the evidence and does not require Parya and Lyuli to be identical.
Model C, the convergence model, proposes separate origins followed by social proximity, occupational parallels, inter-community contact and outsider classification within Central Asia. This is also compatible with current evidence. At present Models B and C require fewer unsupported assumptions than Model A. They are not mutually exclusive: distinct migrant streams could later converge socially.
Figure 5. Three competing models for the Parya-Lyuli relationship. Source: author’s synthesis of the evidence reviewed in this article.
11. Why “Central Asian Gypsy” can obscure more than it explains
The broad label “Central Asian Gypsy” has often been applied to Lyuli and related peripatetic or marginal communities. Parya too has sometimes been placed near such categories in reference databases. Yet typological social resemblance does not equal genealogical identity. Similar livelihoods, endogamy, marginality or outsider stereotypes can produce the same classificatory label across unrelated populations.
This is especially important in comparative work with Romani, Domari and other Indo-Aryan diasporic languages. Indo-Aryan genealogy may demonstrate a very deep South Asian linguistic connection, but different communities can represent different migration dates, routes and histories. Parya’s relatively transparent relationship with north-western/central Indo-Aryan and its remembered Afghan connection may reflect a migration history distinct from that of European Romani.
Accordingly, the present study uses Lyuli, Mugat and Parya as historically situated terms rather than as interchangeable synonyms. Where a source uses an outsider term, that usage should be attributed to the source rather than silently adopted as the community’s preferred identity.
12. Evidence matrix and provisional conclusion
The accumulated evidence produces a clear hierarchy. The identification Tajuzbeki = Parya is strong. Parya’s Indo-Aryan genealogy is strong. Its Tajik-Uzbek location and long-term multilingual contact are strong. An Afghan/Laghman migration stage is plausible. Tiwari’s exact deeper homeland and chronology remain hypotheses. Lyuli/Mugat communities are well documented in the same wider region, but the evidence examined does not establish that they are Parya, descendants of Parya, or speakers of the same inherited language.
The most defensible historical interpretation is therefore not “Parya and Lyuli are the same people,” but a broader proposition: Central Asia appears to preserve more than one history of populations with real or claimed South Asian connections. These histories may have crossed, converged or been socially conflated after arrival. The Parya case is unusually transparent because language preserves the Indo-Aryan connection; the Lyuli case requires a different evidential pathway through ethnography, archival history and community traditions.
13. A research programme capable of resolving the question
The first requirement is a digital critical edition of Tiwari’s Tajuzbeki. Every lexical item in the surviving dictionary should be transcribed with printed-page coordinates, transliteration and language-of-comparison tags. Tiwari’s Haryanvi, Punjabi, Braj, Rajasthani, Tajik and Uzbek comparisons can then be tested quantitatively against modern datasets.
The second requirement is paired fieldwork. Parya and Lyuli/Mugat communities should not merely be studied in separate projects; comparable questionnaires should document kinship vocabulary, numerals, basic verbs, ritual terminology, migration narratives, marriage networks, settlement histories and self-designations. The objective is not to force similarity but to discover whether independent evidence converges.
The third requirement is historical GIS. Tiwari’s settlements, Oransky’s Hissar data, Surkhandarya locations and documented Lyuli settlements can be mapped with evidential confidence levels. Attested residence, remembered residence and hypothesised origin should use different symbols. Such a map would prevent a common scholarly error: visually presenting a speculative migration route as though it were documented fact.
The fourth requirement is archival research in Soviet ethnographic, census and administrative records. If Parya and Lyuli were categorised together or separately at particular times, those records could reveal when social labels began to overlap. Finally, all contemporary research should be community-centred: self-identification, consent, language ownership and the return of digital materials to communities should be built into the research design.
14. Conclusion
The question whether the Lyuli/Luli of Uzbekistan are related to the Parya/Tajuzbeki community cannot presently be answered with a simple affirmative. The evidence is nevertheless sufficiently rich to produce a meaningful conclusion.
Tiwari’s Tajuzbeki is securely identifiable with the language now generally called Parya. Its Indo-Aryan character is demonstrated by structural and lexical evidence, while later research confirms its Hissar-Surkhandarya distribution and the profound effects of Tajik bilingualism. The language is therefore a genuine historical archive of South Asian ancestry transformed by Central Asian residence.
Lyuli/Mugat communities form a second and more ethnographically complex history. Their multiple names, regional distribution and traditions of origin make comparison with Parya legitimate, but present sources do not establish that the two are identical or directly genealogically related. Shared geography is not descent; social marginality is not descent; and an “Indian origin” tradition is not, by itself, descent from Parya.
The most economical interpretation is that the Tajik-Uzbek cultural zone may have received multiple South Asian-connected migration streams, some retaining Indo-Aryan speech and others undergoing more extensive language shift and social reclassification. Later contact and outsider naming could then produce apparent convergence. This parallel-migration-plus-convergence model fits the available evidence better than a simple common-origin assertion, while leaving open the possibility that future linguistic or archival evidence may reveal deeper links.
The importance of the problem extends beyond the identification of one minority. It changes how India-Central Asia connections can be studied. Instead of treating “Indian influence” as a one-way diffusion of culture, Parya and Lyuli histories direct attention to people who moved, settled, adapted, changed language, negotiated names and created new social worlds. Tiwari’s 1970 monograph is therefore not merely evidence for Hindi abroad. Read critically, it is an entry point into the historical anthropology of South Asian mobility across Afghanistan and Central Asia.
Table 1. Comparative evidence: Parya/Tajuzbeki and Lyuli/Mugat
Table 2. Evidential status of major propositions
References
Abbess, Elisabeth, Katja Müller, Daniel Paul, Calvin Tiessen, and Gabriela Tiessen. 2010. “Language Maintenance Among the Parya of Tajikistan.” SIL Electronic Survey Reports 2010-014, pp. 1-32.
Cardona, George, and Dhanesh Jain (eds). 2003. The Indo-Aryan Languages. London and New York: Routledge.
Masica, Colin P. 1991. The Indo-Aryan Languages. Cambridge: Cambridge University Press.
Oranskiy, I. M. 1977. Folklor i yazyk Gissarskikh parya [Folklore and Language of the Hissar Parya]. Moscow: Nauka, Glavnaia redaktsiia vostochnoi literatury.
Oranskiy, I. M. 1983. Tadzhikoyazychnye etnograficheskie gruppy Gissarskoy doliny (Srednyaya Aziya): etnolingvisticheskoe issledovanie. Moscow: Nauka.
Oranskaia, Tatiana I. 2014. “Parya.” Great Russian Encyclopedia, vol. 25, pp. 405-406.
Oranskaia, Tatiana I. 2025. “Influence of the Tajik Language on the Formation of the System of Pronominal Clitics in the Paryá Language in Tajikistan.” Sociolingvistika 2(22): 136-157. DOI 10.37892/2713-2951-2-22-136-157.
Ozierski, Przemysław. 2014. “Central Asian Gypsies - Lyuli: The Overview of Current Socio-Economic Problems.” Przegląd Narodowościowy / Review of Nationalities, no. 3. Online journal version published 28 January 2020.
Payne, John R. 1997. “Indic Languages Outside India: Romani, Parya and Dumaki.” In Shirin Akiner and Nicholas Sims-Williams (eds), Languages and Scripts of Central Asia. London.
Ruziyeva, Mashhura Abdumuminovna. 2026. “Historical Roots and Evolution of Family Rituals among the Lyuli.” Oriental Journal of History, Politics and Law, pp. 689-697.
Tiwari, Bholanath. 1970. Tajuzbeki: Soviet Sangh mein boli jane vali Hindi boli: aitihasik aur tulnatmak adhyayan tatha sankshipt shabdkosh. 1st edn. Delhi: National Publishing House.
Source-critical note
The supplied PDF of Tiwari is treated as the primary source for page-specific claims. It contains 150 scan pages and ends at printed p. 134 during the Tajuzbeki-Hindi dictionary; external bibliographic records indicate that the original volume extends to 201 pages. The article therefore does not attribute unseen content to the supplied scan. Web-based sources were used to verify later Parya scholarship and Lyuli ethnography; journal and institutional sources were preferred over unsourced summaries. The Lyuli photograph is separately credited and is not used as historical proof of ancestry.
No comments:
Post a Comment
Share Your Views on this..