“A people proscribed by opinion, and doomed by the laws to opprobrium and ignominy; a race which, driven from all liberal professions, has been for ages, and still is, robbed of its right to hold landed property; which, subjected to special and severe regula­tions, has learned at once to obey and yet to preserve a manner of independence; which, despite the contempt that it inspires and the hate that it awakes and the prejudices wherewith it is received and judged, still resists this contempt, this hatred, and finally all those causes which ought to disunite, loosen, and annihilate the family, the race, the nation;—such a people, I say, deserves the observer’s attention, if only from the fact of its existence.”
























OF general works upon the subject of the Gypsies we have perhaps enough, and more than enough; this objection, however, cannot be urged against specialities, which still are highly desirable in every department of “Chinganology.” I use the latter term in preference to the French Tsiganologie, of which more presently, and the “Romanology,” a term of dubious import, lately introduced into English.


I wish to place in extenso before the public the following conclusions which the study of some years has, it is hoped, justified me in drawing with regard to the relation of the Gypsies and the Jats


1. The medićval Gypsies of Europe were the last wave of Aryan emigration that flowed westward during the early fifteenth century; and this wave was possibly preceded by more than one similar exodus.


2. The medićval Gypsies show family resem­blances, physical and moral, ethnological and linguistic, with the modern Jats, a highly important race, which extends from the mouth of the Indus





[p. 134]


to the head of the great Valley, thence ramifying over Turkistan and the far North.


3. There are solid reasons for believing the Jats and the Jin‑tchi of Tatary to be the modern repre­sentatives of the classical Getć and the Goths of later days.


4. The language of both tribes (Jat and Gypsy) is of Indo‑Persian type, the Indian ingredient not being so much decomposed as in the modern varieties of Prakrit. An absolute isolation of speech, especial reasons for secrecy, and the fact of being oral and never written have preserved its purity among the Gypsies; while the Jats, in close contact with alien tongues, have made those secular linguistic changes which are familiar even to English and French.


5. The most ancient name of the race is Chingáneh, a term still used in Persia and Turkey, and neces­sarily corrupted by the Arabs, who have no ch, to Jingáneh.


6. Concerning the origin of the Gypsy article (o—os, a—as, etc.), which is unknown to both Sanskrit, and Prakrit, the suit is still pending. Possibly it is original and peculiar to the dialect; more probably it is an European and especially a Greek innovation. Briefly, until we have gram­matical and vocabularian sketches of the Central Asian and the Turkoman‑Gypsy tongues, we are not in a position to draw conclusions.


I propose to discuss the Indian affinities of the



[p. 135]


Gypsies. I begin with a detailed critique of the various reviews proceeding from the prolific pen of M. Paul Bataillard, who claims the merit, such as it is, of having first identified the Gypsies and the Jats. I end with topographical notes on both tribes throughout their extension from the Indus to Morocco and even to the Brazil.

Part  I









THE following letter to the Academy (March 27, 1875), which opened the discussion between M. Paul Bataillard and its author, speaks for itself1:


“In the Academy of February 27, 1875, I had these words:


“‘Professor de Goeje, of Leyden, has printed some interesting Contributions to the History of the Gipsies (sic). He accepts the view propounded by Pott,2 as early as 1853, that the Gipsies are closely


1 In this reprint of the original letter the only changes are a few verbal corrections and suppressions of the parts elsewhere enlarged upon.

2 The famous work Die Zigeuner in Europa und Asien, 2 vols. 8vo (Halle, 1844‑5). It was followed by two Nachtrags (which I have not seen). The first contains a Syro‑Gypsy vocabulary; and the second, notices of their manners and customs in Turkey and other countries. See Zeitschrift d. Deut. Morgen. Gesell., III., pp. 321—335, of 1849; and Ibid., Vol. VII., p. 393.




[p. 137]

related to the Indian Jatt (a name which the Arab historians transform into Zott). . . . Dr. Trumpp1 has already pointed out the close resemblance between the European Gipsies and the Jatt of the banks of the Indus.’


“I venture to hope that you will permit me to show the part taken by myself in this question.2 Sindh and the Races that inhabit the Valley of the Indus (London: Allen), my volume written between 1845 and 1849, and published in 1851, thus treats of the peoples of the plains:


“‘The Jat, or as others write the word, Jath, Juth, or Jutt, was, in the time of the Kalhorá dynasty, one of the ruling classes in Sindh. It was probably for this reason that the author of the Tohfat el Kirám. (a well‑known book of Sindhi Annals) made them of kindred origin with the Belochis, who now repudiate such an idea with disdain. The Jat’s account of his own descent gives to Ukayl, the companion of Muhammad, the high houour of being his progenitor; but what class of Muslim people, however vile, do not claim some equally high origin?


“‘As Játaki, the dialect peculiar to the people, proves, they (i.e. the Sindh division of this exten­sive race) must have come from the Panjáb,


1 Dr. Ernest Trumpp's Sindhi Grammar. (Trübner, 1872).

2 The literati of Europe form a guild into which none but members are admitted. At times their absolute disregard of meum and tuum, especially when they plunder an obscure name, is a fine study of trade morality—or its reverse.


[p. 138]


and the other districts Ubho or Báládasht, Jhang­-Siyál, Multán, and other regions dependent upon the great Country of the Five Rivers. Driven by war or famine from their own lands, they migrated southwards to Sebi (Sibi or Siwi, Upper Sindh) and to the hills around it. They are supposed to have entered Sindh a little before the accession of the Kalhorá Princes, and shortly afterwards to have risen to distinction by their superior courage and personal strength. At present they have lost all that distinguished them, and of their multitude of Jágírdárs, Zemindárs, and Sardárs now not a single descendant possesses anything like wealth or rank. The principal settlements are in the provinces of Kakrálo, Játi, Chediyo, Maniyár, Phuláji, and Johí. [Those of Umarkot speak, it is said, a different dialect from the Indine Jats, and not a few migrating tribes graze their herds on the great Delta.1] They are generally agriculturists or breeders of camels, and appear to be a quiet, inoffensive race. Throughout the eastern parts of Central Asia, the name Jat is synonymous with thief and scoundrel.


“‘The Sindhi Jats have many different Kamus or clans, the principal of which are the following: Babbur, Bháti, Jiskáni, Kalaru, Magási, Mir‑jat, Parhiyár, Sanjaráni, Siyál, and Solángi.’


“To this text were appended the following notes:


“Jatu in the Sindhi dialect means: 1. A camel‑driver


1 These words were afterwards added to my MS. copy.



[p. 139]


or breeder. 2. The name of a Beloch clan. Generally in the lower Indus Valley it is written Jatu, and pronounced Dyatu. It has three significations: 1. The name of a tribe, the Jats. 2. A Sindhi, as opposed to a Beloch; it is in this sense an insulting expression, and so the Beloch and Brahins of the hills call the Sindhi language Játhki. 3. A word of insult, a ‘barbarian,’ as in the expression do‑dasto Jatu, ‘an utter savage.’


“Lt. Wood's work shows that the Jats are still found in the Panjáb and all along the banks of the Indus.


“Under the name Jat no less than four races are comprised.


“I continued:


“‘it appears probable from the appearance and other peculiarities of the race that the Jats are connected by consanguinity with that peculiar race the Gypsies. Of 130 words used by the Gypsies in Syria, no less than 104 belong to the Indo‑Persian class of language. The rest may be either the remains of the barbarous tongues spoken by the aboriginal mountaineers who inhabited the tract between the Indus and Eastern Persia, or the in­vention of a subsequent age, when their dispersion among hostile tribes rendered a “thieves’ language” necessary. The numerals are almost all pure Persian. There are two words, “kuri” (a house) and “psih” (a cat), probably corrupted from the Pushtu “kor” and “pishu.” Two other words are Sindhí “mánna” for “máni,” bread, and “húi” for “hú,” he. As might be expected from a tribe inhabiting Syria,



[p. 140]


Arabic and Turkish words occasionally occur, but they form no part of the groundwork of the language.’


“It was my fortune to wander far and wide, during four years of staff service, about the Valley of the Indus; and to make personal acquaintance with many, if not all, its wild tribes. I saw much of the Jats, lodged in their huts and tents, and studied the camel under their tuition. They are the best ‘Vets.’ and breeders known to that part of the Indian Empire. My kind friend, now no more, then Colonel, and afterwards General, Walter Scott, of the Bombay Engineers, had a Jat in his service; and the rough old man’s peculiarities afforded us abundant diver­sion. Thus I was able to publish in 1849 the first known notice of Játakí and its literature. The author of the famous ‘Dabistan’1 applies the term ‘Jat tongue’ to that in which Nánah Sháh, the Apostle of the Sikhs, composed his Grauth2 and other works. Throughout the Panjáb Jatki bút (‘Jat tongue’) is synonymous with the Gunwár ki boli or ‘peasants’ jargon’ of Hindustan.


“I wrote the word Játaki with two italics. The first denotes the peculiar Sindhi sound, a blending of j and t; the second is the familiar cerebral of Sanskrit and Prakrit, which survives to a certain


1 The full title is Dabistán‑i‑Mazáhib, or School of Faiths (not “of Manners”): there is a translation by David Shea and Anthony Troyer for the Oriental Trans. Fund, 3 vols. 8vo (Paris, 1843).


2 Adi Grauth: the Sacred Book of the Sikhs.



[p. 141]


extent in our modern English tongue, though unknown to the Latin and the Teutonic languages. The tribal name is Jatu, with the short terminal vowel which in Sindhi, as in Sanskrit, follows the consonant; its plural, Jatán, ends with a well-­marked nasal.


“At that time I divided this rude race of semi-Bedawin into four great tribes; namely:


“‘The Panjábí Jat, who is neither a Hindu nor a Hindi (Muslim). He first appears in Indian history as a nomad, alternately shepherd, robber, and tem­porary tiller of the ground. Many became Sikhs, and did good service to Nánah Shah’s faith by their zealous opposition to Muhammadan bigotry. As this was their sole occupation for many years, they gradually grew more and more warlike, and at one time they were as fighting a race as any in India. They have been identified by Colonel Sleeman and others with the ancient Getć and their descendants the Goths.1


“‘The Jat or Dyat of the Hazárah country, Jhang-­Siyál, Kach (Kutch) Gandáva, and Sindh gene­rally, where they may number two hundred and fifty thousand out of a total population of one million. They are all Muslims, and are supposed to have


1 Jornandes, “De Getarum sive Gothorum Origine et rebus Gestis.” The learned Abbate Fortis (Dalmatia, I. 1, § 1) includes among the Slav peoples the Scythians, Getć or Goths, Slavini (Slovenes), Croats, Avars, and Vandals. Our grandfathers derived the term “Goths” from Gog (and Magog).


[p. 142]


emigrated from the north during or shortly after the Kalhorá accession; hence their dialect is commonly called Belochki. In those days the Belochís were very little known to Sindh, whose aristocracy, the Amírs, Jágírdárs, and opulent Zemindárs, was either Sindhi or Jats. About Pesháwur “Jat” is still synonymous with Zemindár or landed proprietor; at times, how­ever, it is used as a term of reproach.


“‘The third is a clan of Belochís, who spell their name with the Arabo‑Persian, not the Sindhi j. In the lower Indine Valley they hold the province of Játi, and other parts to the south‑east. The head of the tribe is entitled Malik (literally “King”), e.g. Malik Hammál Jat.1


“‘The next is a wandering tribe, many of whom are partially settled in Candahár, Herát, Meshhed, and other cities of the Persico‑Afghan frontier. They are found in Meckran; and they sometimes travel as far as Maskat, Sindh, and even Central India. They are held to be notorious thieves, occupying a low place in the scale of creation. No good account of this tribe has as yet appeared; and the smallest contributions upon the subject would be right thankfully received.’


“The fifth which must now be added is the Jin‑tchi of Central Asia. These people are not, as Mr. Schuyler2 seems to think, ‘Káfirs from Káfiristan’;


1 The account given by Mr. Hughes of the Jat in Belochistan will be found in a future page (215).

2 Turkistan. (Sampson Low & Co., 1876.)



[p. 143]


they are apparently true Jats—an idea once advanced by Mr. Andrew Wilson of the Abode of Snow.1


“These tribes are looked upon as aborigines, which simply means that their predecessors are unknown.2


“Such were the notices collected by me in manu­script some years before 1849. At that time the Orientalists of Europe were almost unanimous in identifying the Gypsies with the Nat’h, a scattered Indian tribe of itinerant tinkers and musicians, the ‘poor players’ of the great Peninsula, utterly ignorant of horse‑couping, cattle‑breeding, and even poultry‑snatching. And the conviction still holds its ground; only lately my erudite correspondent, Dr. J. Burnard Davis, reminded me of it.


“Of course the humble linguistic labours of a perpetual explorer can hardly be familiar to the professionally learned world; but I cherish a hope that you will aid me in resurrecting my buried and forgotten work.”


1 Academy, October 14, 1876.

2 The letter here contains a sketch of Játaki literature in Sindh. I have also suppressed a paragraph noticing their migration and tribal name; both these subjects will be discussed with more detail.