8. Volume 2 — part 8

which I have examined, two are in conformity with the extract given above, with the exception of reading Katoriyan for Katorman, and a third has —

“ after Basdeo from among their rulers (i.e., of the Indians), one was Kanak, and he was the last of the Kayorman kings.”

The omission of all notice of the Kabul Turkish dynasty, and the making Kanak succeed Basdeo, and the Brahmans succeed Kanak, without any notice or allusion to there being intermediate kings, is a culpable omission on the part of Bashidu-d din and Binakiti. The making Kanak the last of the Turkish dynasty does not seem au- thorized by the only original of A1 Birum’s Tdr’ikliu-l Hind which we possess, and Bashidu-d din must have had other copies or other works to have authorized him to make this statement. M. Eeinaud (Mem. 30) considers that he has used some other work of A1 Birum’s which has not come down to us, but this may reasonably be doubted.

M. Eeinaud altogether ignores these readings of the manuscripts consulted by Mr. Thomas, and merely observes upon them, “ On a vu ci-devant, que le vizir de Perse Baschid-eddin, avait, dans son Histoire des Mongols, mis a contribution un ecrit d’Albyrouny autre que celui-ci, et que ne nous est point parvenu. Malheureuse- ment, les manuscripts de l’ouvrage de Baschid-ed din different entre eux : au lieu de Laldouzeman, ils portent Katourman, et on ne dis- tingue pas bien s’il s’agit la d’un prince ou d’un pays.” Notwith- standing this, I have been given to understand by those who have seen the original manuscript of the Tarikliu-l Hind, that even that bears a closer resemblance to Katourman than Laldouzeman } Taking all circumstances into consideration, I am disposed to get rid of the name of Laktouzeman from the Tdr'iklm-l Hind, and to substitute for it, by two slight changes in the original, al Katorman, which repre-

1 [The name occurs only twice in Reinaud’s printed extract. In the first instance, it is given as quoted above, but in the second it is ; ,y.O Lakturzaman. See Fragments, p. 135.J

sents the name of a tribe, or prince of that tribe, as well as the name of the country in which that tribe resided. I have therefore trans- lated the disputed line, “ The last of them was a Katorman.”

Let us now enter upon some of the considerations which this name suggests.

The Katormans, or Kators, have hitherto been better known to modem than ancient history. We are informed that it was the name of one of the tribes of Kafiristan,1 and that the ruler of Chitral to this day bears the title of Shah Kator,3 and I have heard the same designation given to the chief of Gilgit. The country of Kator is also spoken of by Sadik Isfahani, as being the country of the Siyah- poshes, or black-vested, on the borders of Kabul.3

These Kators boast still of their Grecian lineage, and their claim to this honour is by no means, as many have supposed, of modern origin, attributable to our own enquiries after the descendants of the followers of the Macedonian conqueror.4

We find at the period of Timur’s invasion of India, the Katorians making themselves conspicuous for their opposition to that monarch. After leaving Inderab he entered their difficult country by way of Khawah, and after an expedition of eighteen days reduced them to submission. As we thus have proof that this country and people were called by the name of Kator at so early a period, it seems pro- bable that the Kators whom we read of in Abu-1 Fazl Baihaki are no other than the descendants of the dynasty we have been consider- ing, and that the Ghaznivide sovereigns organized them among their troops, as we know from the Tdr'tkh-i Tarnini that Mahmud was in the practice of doing with conquered nations, as exemplified in his treat- ment of the Khiljis, Afghans, and Indians. It is evident from the extracts given in this work from the TabaJcat-i Akbari and the Tar’ikh-i Mas’udi, that a body of Kator troops was kept in pay, and that the Tilak mentioned therein was the commander of these foreign troops,

1 Elphinstone’s Kabul , vol. ii. pp, 376, 387.

3 Bume’s Bokhara, vol. ii. p. 209 ; and Journal A. S. Bengal, vol. vii. p. 331.

4 [For other references to the Kators, see Thomas’s Prinsep, I. 314. Lassen, Ind. Alt. III. 890, 1176. Masson's Narrative, I. 193. Vigne, Ghazni, etc., p. 235. Trumpp, in Journ. E. A. S. xix. 1. Jour, des Sav. Vol. V., 1855, where M. Viv. de St. Martin attempts to identify them with the Cudrusii of Pliny VI. xxiii.]

which were rated as Indian, he being hi one passage spoken of as commander of the Indians, in another of the Kator troops. It opens a very interesting subject of investigation to enquire if these Kators have no memorials of themselves in India. The identity of name and the period of the establishment of the Kators in Kumaun appear to render it probable that we have in them the descendants of those Kators who fought under the banners of the first Muhammadan conquerors.

A curious coincidence of names seems worth noticing in this place. It will he observed that A1 Biruni makes the Turk kings of Kabul come from the mountains of Tibet, and Grecian and Chinese authors concur in saying that in the first years of the Christian era the valley of the Indus and some of the neighbouring countries were occupied by a race from Tartary. Ptolemy, Dionysius, and the author of the Periplus of the Erythrean Sea, give to the country watered by the Lower Indus the name of Indo-Scytliia, and Ptolemy applies the same name to a country at the bottom of the Gulf of Cambay. The Chinese writers inform us that a people of Tatar race named Yue-chi or Yue-tchi crossed the Hindu-kush, and established themselves in Afghanistan. Fa-Hian speaks of these barbarians having occupied, long before his visit to India, the province of Peshawar.

De Guignes has informed us, after Chinese authors, that the nomade race of Yue-tchi, being driven about the year 160 before Christ from its original seat in the western provinces of China, by another race called Hioung-non, established themselves in Transoxiana, and spread over the countries in that neighbourhood. Abel-Remusat and Klaproth have also furnished us with further particulars from the same sources. We learn that the Yue-tchi took part in the struggle which took place between the Greek princes of Bactvia and the Arsacidan monarchs of Persia, and that they contributed to the down- fall of the former. A few years before Christ, the Yue-tchi chief, named Khieou-tsieou-hy, after subjugating the other independent rulers of his own tribe, proclaimed himself king, and conquered the countries situated between the Oxus, Hindu-kush and Little Tibet. His successor, Yan-kao-tchin, penetrated as far as India.

Some time after, the monarch of the Yue-tchi, whom the Chinese

call Ki-to-lo, which Klaproth has converted into Ghidor, descended to the south of the Hindu-kush “ in following the valley of the Indus” (?), and invaded India on the north. Among other regions he reduced the province of Peshawar ; but being himself compelled to return westward, left the government of the conquered country to his son.1 M. Reinaud is of opinion2 that it is to this Ki-to-lo that Fa-Hian alludes, when he says, “ Formerly the king of the Yue-tchi, levied a powerful army, and came to attack the country he was anxious to obtain.”

The conquerors, who remained in the valley of Kabul, received the name of the “ Little Yue-tchi,” while the mass of the nation was designated the “ Great Yue-tchi.” In these Little Yue-tchi we have the ancestors of our modem Jats, a subject which I may, perhaps, discuss at further length hereafter.

It is impossible not to be struck here with the coincidence of the name of Ki-to-lo with Kitor or Kator, the l and the r being as usual convertible. Here we seem to have the origin of the name Kitor, the establishment of a prince of that name between Kabul and the Hindu-kusli, on the very site of the modem Kafiristan, or land of Siyah-poshes and the country of Kitor, according to the authorities given above. It is probable that we are to look to one of his de- scendants for the Katorman, who was the last of the Turkish dynasty ; and these united considerations have combined to induce me to adopt the readings to which I have given the preference above.

It is to be observed that A1 Biruni asserts the Turkish dynasty of Kabul to have lasted for sixty generations ; but we are not to sup- pose that the crown continued in the same family or tribe, but that they were members of the great Turkish stem of nations, which conveys no more definite notion than the Scythians of the ancients, or the Tartars of the moderns. There may have been Turks of other tribes who ruled in the kingdom, who, whether Sakas, Turushkas, Huraris, Yue-tchis, or Kators, would still be classed under the generic designation of Turks, as the last of the Turks appears to have reigned about a.d. 850. If we allow fourteen years as the

1 Nouveaux Melanges Asiatiques. Tom. i. p. 223. LaicLlay’s Translation of Fa- Hian. Foe-koue-ki, p. 81. Tableaux Historiques d’ I’Asie. p. 134.

2 Memoire sur I'lnde, p. 83, from which work the preceding abstract of Yue-tchi history is taken.

average duration of their reigns, we shall find the period of the conquest occurring about the first year of the era of Our Saviour ; and if we allow sixteen years as the average duration, we shall exactly bring it to the period of the downfall of the Greco-Bactrian Empire in 125 before Christ.

Here, then, there is reason to suppose that the first monarch of the Turkish dynasty must have been the subverter of the Grecian Empire in the East. He is called by A1 Biruni “Barlitigm;” tigin being a common Turkish affix, signifying “ the brave,” as Alp-tigin, Subuk-tigin. M. Reinaud conjectures that Barh or Barha answers, probably, to the word pharahatassa, which Lassen and Wil- son have read on certain Greco-Barbarian coins, and to be the same name which the Greeks have converted into Phraates and Phraoites.1 A1 Biruni informs us that the names of these princes were recorded on a piece of silk, which was found in the fort of Nagarkot, when it was taken by the Muhammadans ; but that circumstances prevented his fulfilling his anxious desire to examine it.

A1 Biruni mentions that Kanak was of the number of these kings, and that he founded the Vihar, or Buddhist monastery at Peshawar, called after his name even in A1 Biruni’s time, and which, probably, occupied the site of the present conspicuous building, called the Gor-khattrf, at the eastern entrance of that town. The romantic anecdote which he relates of him, and which, probably, has little foundation in truth, will be found among the extracts translated from the Tar'ilchu-l Hind, in this volume.

M. Reinaud considers this Kanak to have reigned a little prior to the commencement of our era, and to be the same as the Kanika or Nika of Fa-Hian; the Kanishka of Hiuen-thsang and the Raja- tarangini and the Kanerkes of the Greco-Barbarian coins ; and General A. Cunningham has formed the same opinion independently with reference to the two first identifications, considering the same monarch to be the Kanika of the Chinese, and the Kanaksen from whom many Rajput families trace their lineage.2

According to Hiuen-thsang, Kanika or Kanishka reigned over

2 Mem. sur l’ hide, p. 73; Thomas’ Prinsep, Index “Kanishka;” Jour. Peng. As. Soe., Yol. xxiii.

the whole valley of Kabul, the province of Peshawar, the Panjab, and Kashmir. He crossed the Hindu-kush and Himalaya, and subjected Tukharistan and Little Tibet. He received the title of the Lord of Jambu-dwipa, which is equivalent to “ The Paramount of all India.” He was a long time a stranger to the dogmas of Buddhism, and despised the law ; until, by chance, he was con- verted to that faith, and became one of its most zealous disciples and promoters.

The same Chinese author states that he reigned four hundred years after the death of Buddha, which, as it occurred 544 years before our era, would bring it to more than a century before Christ ; but as he expresses his dates in round numbers, we cannot rely much upon his precision. We may with more probability look for it a century later, if, at least, he be the same as Kanerkes, for among the coins and other objects bearing his name, which were found in the tope of Manikyala, and which would appear to indicate that that monument was constructed under the reign of that prince, certain Roman medals were also found of the period of Octavius and Antony extending to as low as 33 b.c.1

The Yue-tchi evidently established themselves in Kabul subse- quent to the reign of Kanishka, and probably not long after, for Fa-Hian, about the year 400 a.d., speaks of their occupation of that valley, as if it were a transaction of no recent date. If we assign to Ki-to-lo the date of a.d. 200, we shall have nearly seven hundred years from the first to the last of the Katorman dynasty, during which, probably, other families and other tribes may have inter- mediately occupied the throne, without entirely subverting the right of the Yue-tchi conquerors of the valley.

The statement of A1 Birum, respecting the occupation of Kabul by the Turks, is in strict conformity with Biladuri and Tabari, and with the brief notices which the other early Arabic historians and geographers have given us respecting that city. They couple it, however, with the curious announcement of an occupation divided between the dominant Turks and subject Hindus. Mr. E. Thomas

1 M. Raoul-Rockette, Journal des Savants, ann. 1836, p. 70. [Thomas’ Frinsep. I. 150, and Index, v. Manikyala.]

has considered this subject at considerable length in another excellent paper by him, on the Coins of the Ghaznivides.1

The first in order is Mas’udf, who visited the valley of the Indus in 303 a.h.= 915 a.d. He says nothing of the political and religious revolution which we have been considering, by which Brahmans had been substituted for Buddhist Turks. On the contrary, he designates the prince who reigned at Kabul by the same title as he held when the Arabs penetrated for the first time into those regions.

Istahkri, who wrote within six years after Mas’udi travelled in India, says : —

“ Kabul has a castle celebrated for its strength, accessible only by one road. In it there are Musulmans, and it has a town, in which are infidels from Hind.”

Ibn Haukal began his travels in 33 1 a.h. = 942 a.d., and wrote an account of them thirty-five years later. He follows his predecessor implicitly in the main points, but respecting the occupants of the town, the Bodleian copy varies2 from the Lucknow one, which bears the name of Ashkdlu-l Bilad. In the former, “Hindu infidels” is converted into “ Infidels and Jews.” The latter reads : —

Tlie statement of A1 Birum, in his Kdnun-i Ma'sud'i, written less than a century after this, is : —

Here there is no specification respecting the different occupancy of the castle and town, but nothing to impugn the correctness of what is asserted by Istakhri and Ibn Haukal. There is no occasion to

1 Journal of the Royal Asiatic Society , vol. is. p. 267.

quote any of the later geographers, who add nothing to our informa- tion, and are careless as well as confused in their statements.

Before concluding this subject of the Turkish occupation of Kabul, the statement of Ibn Khallikan should be noticed, who states in his article on “ Ya’kub bin Lais,” that Kabul, in the times of that prince, was inhabited by a Turkish race who appertained to a tribe called Durar'i. This name is new, and the assertion would authorise us to conclude that in his time the Turks were still pre- dominant, though that fact would scarcely seem consistent with what we shall have to advance under Kamlka. It is possible that the term Duran may have connection with Darra, a hill pass, and that allusion may be to the country to the north of Kabul, just in the same way as in modern times the inhabitants of those same tracts are styled in Kabul “ Kohistanis,” or liill-men.

It does not appear when the city was either first or finally subdued by the Muhammadans. It is evident, however, that the first inroads were not followed by permanent occupation, and that there was no entire subversion of the native dynasty till the Ghaznivide dynasty rose to power.

The first invasion we read of was in the time of ’Abdu-llah, governor of ’Irak, on the part of the Khalif ’Usman. He was directed by the Khalif to send an emissary to explore the provinces of Hind ; and notwithstanding a discouraging report, ’Abdu-lla ordered the country of Sijistan to be invaded by one of his cousins, ’Abdu-r Rahman, son of Sarara. ’Abdu-r Rahman advanced to the city of Zaranj, and besieged the Marzaban, or Persian governor, in his palace, on the festival of the ’I'd. The governor solicited peace, and submitted to pay a tribute of two millions of dirhams and two thousand slaves. After that, ’Abdu-r Rahman subdued the country between Zaranj and Kish, which was then styled Indian territory, and the tract between Ar-Rukhaj (Arachosia) and the province of Dawar — in which latter country he attacked the idolaters in the mountain of Zur, who sued for peace ; and though he had with him 8,000 men, the booty acquired during this incursion was so great, that each man received four thousand pieces of silver as his share. Their idol of Zur was of gold, and its eyes were two rubies. The zealous Musulmans cut off its hand and plucked out its eyes, and then

remarked to the Marzaban how powerless was his idol “ to do either good or evil.” In the same expedition, Bust was taken. After this, ’Abdu-r Bahman advanced to Zabul, and afterwards, in the time of Mu’awiya, to Kabul.1 The year in which this inroad was made is not mentioned, but as ’Abd-ulla was removed from his government in 36 a.h., we may consider it to have taken place about the year 35.

In the year 44 a.h. Muhallab ibn Abu Sufra, whose army chiefly consisted of the tribe of Azd, which was very powerful in Khurasan, and contributed largely to the downfall of the Ummayides — advanced on the Indian frontier as far as Banna (Banu) and Alahwaz [or “Alahwar” = Lahore?] two places situated between Kabul and Multan. Firishta makes him penetrate as far as Multan, and opens his history by saying he was the first chieftain who spread the banners of the true faith on the plains of Hind. He says he plundered the country and brought back to the head-quarters of the army at Khurasan many prisoners who were compelled to become converts to the faith, Muhallab had been detached from the main army which had invaded

1 Biladuri, quoted in Memoire, p. 173, and in Geschichten der Chalifen, vol, i. Anhang, p. x. Tarjv.ma-i Futuhdt of Ahmad bin ’Asmi Kufi. — [I have found two Persian extracts from the Futuhdt of Ahmad among the papers. They are short and important, so I give translations.— Ed.]

Conquest of Sijistan by ’Abdu-r Rahman Samrat under the Khalif ’Usman. — ’Abdu-llah, son of ’Amir, wrote for his nephew on the father’s side, ’Abdu-r Rahman Samrat bin Jandab bin ’Abd Shamsh bin ’Abd Sinaf, and having fitted out an army for him, sent him to Sijistan. 'Abdu-r Rahman led his forces to Zaranj. The people of the city offered battle, and a fierce fight ensued between the opposing parties. The city was taken, and the Musulmans obtained great spoil, carrying off many captives from Sijistan, and incalculable wealth. ’Abdu-r Rahman then marched to subdue Kfihul.

Conquest of Kabul. — When ’Abdu-r Rahman came in sight of Kabul, the ruler of the place (Kabul Shah), who was lame, was in the city. lie came out and fought several engagements with the Musulmans, hut retreated into the city, and came forth no more. ’Abdu-r Rahman besieged it, and remained seated before it, fighting with the garrison for a whole year. He and his soldiers had to endure many hardships during the siege, but at length they carried the place by assault ; and when they entered it, they put the fighting men to the sword, and made the women and children prisoners. Kabul Shah was taken captive, and brought before ’Abdu-r Rahman ; but when he was ordered to be beheaded he turned Muhammadan, and repeated the creed. ’Abdu-r Rahman treated him with honour and kindness. The plunder and the captives which had been taken in Kabul, Zaranj, and Sijistan, was collected, and a fifth portion was set apart and sent to ’Abdu-llah bin ’Amir, with a report of the conquest of Sijistan and Kabul.]

Kabul from Merv, under ’Abdu-r Rahman bin Shimar, and had made converts of twelve thousand persons. Muhallab subsequently made himself conspicuous as governor of Alahwar, and exterminator of the Azrakian insurgents, and as a traitor to his master, ’Abdu-llah ibn Zubair, the Khalif of Mecca. He was the ancestor of those chiefs, who, under the name of Muhallabis, often occur in the history of the later members of the Ummaya family, until they were nearly exter- minated at Kandabil in 101 h.1 Gildemeister doubts the truth of this expedition, as Sijistan had not yet been conquered ; but he forgets that the Musulmans did not penetrate to India through Sijistan, but through Kabul.

In Biladuri’s account of this interesting expedition, there is a curious relation which must not be altogether omitted. He informs us that in the country of Kikan, Muhallab encountered eighteen Turks, mounted on horses with their tails cut. As they were all killed fighting, Muhallab attributed the activity and valour of “ the barbarians ” to the fact of their horses’ tails being cut. “ Upon which he ordered his own horses’ tails to be docked ; and he was the first amongst the Musulmans who adopted the practice.”3

About the same time, ’Abbad, the son of Ziyad, made an incursion on the frontier of India, by way of Sijistan. He went through Rudbar to the Hindmand (Helmand), and after staying at Kish, he crossed the desert, and reached Kandahar. Although the country was conquered, many Musulmans lost their lives in this expedition. 3

Biladuri informs us that under the Khilafat of Mu’awiya, ’Abdu-r Rahman, son of Samrah, penetrated to the city of Kabul, and obtained possession of it after a month’s siege. He conquered also the circum- jacent countries, especially Ar-Rukhaj (Arachosia). The king of Kabul made an appeal to the warriors of India, and the Musulmans were driven out of Kabul. He recovered all the other conquered countries, and advanced as far as Bust, but on the approach of another

1 Erpenii Elmaein Historia Saracenic a, ann. 101.

J Biladuri, see Vol. i. p. 116. Briggs, Firishta, vol. i. p. 4. The Chinese authorities seem to allude to this expedition. Memoires concernant les Chinois, Tom. xv. p. 474. See also Tom. xvi. p. 372-5. Hammer, Gemaldesaal der Lebensbeschre- ibungen, vol. ii. p. 9.

3 Biladuri, ut supra. "Weil, Geschichte der Chalifen , vol. i. p. 292.

Musulman army, lie submitted, and engaged to pay an annual tribute.1

Tlie Kabuli's subsequently profited by the contests which dis- tracted the Khilafat, and the tribute was withheld ; but in 64 a.h. = 683-4 a.d. ’Abdu-1 ’aziz, the governor of Sistan, declaimed war against the king of Kabul, and in the combat which took place, that king was defeated and killed. The war continued under his successor, and he was compelled to submit to the payment of tribute, but whenever opportunity offered, renewed efforts were made by the Kabulis to recover their lost independence.2

Amongst the earliest attempts against Kabul may be noticed that of ’Abdu-llah, governor of Sistan, in 78 a.h.=697-8 a.d., or accord- ing to some, in the following year. When he arrived at Ximroz, Hajjaj desired him not to linger in Sistan, but to march without delay towards Kabul to enforce the payment of the tribute from Ranbal, to which that chief had agreed ; and ordered him peremptorily not to return until he had subjugated the whole province. Ranbal retiring before his assailant, detached troops to their rear and blocking up the defiles, entirely intercepted their retreat, and in this situation exposed to the danger of perishing by famine, ’Abdu-llah was com- pelled to purchase the liberation of himself and followers for a ransom of seven hundred thousand dirhams.3

To wipe out the disgrace which the Muhammadan arms had sus- tained, ’Abdu-r Rahman bin Muhammad bin Asha’s, was despatched to Kabul by the famous Hajjaj in 81 a.h.=700-1 a.d. ; 4 or in the preceding year, according to some authors, he was sent at the head of forty thou- sand men into Sistan, and having there united to his own troops the troops of the province, marched without delay against the prince of Kabul. ’Abdu-r Rahman returned to Sistan laden with booty, but incurred the displeasure of Hajjaj by not remaining to secure his conquest. Exasperated by a threat of supersession, he determined to cany his arms against his master, and, in order to strengthen his power, concluded a treaty with the enemies of his faith, in which it was

1 Memoire sur V Inde, p. 179. 2 Memoire sur V hide, p. 178.

3 Tdrikh-i-Alfi, Ann. 68, p.m. Muhammad. See the extracts from that work in a subsequent volume of this compilation. Price’s Hahommedan Hist., Vol. i.. p. 454.

4 Mem. sur l' Inde, p. 179; Weil, Geschichte der Chalifen, Tom. I. p. 449; Ockley’s History of the Saracens. [82 a.h.] Bohn’s Edit. p. 490.

stipulated that if his expedition should be attended with success, JRanbal should be absolved from every species of tribute, provided the latter should agree to afford him an asylum in the event of failure. After many vicissitudes of fortune, ’Abdu-r Rahman was at last com- pelled to seek the protection of his ally, who, after treating him for some time with kindness and hospitality, was at last seduced by the promises or by the threats of Hajjaj to deliver up his guest. ’Abdu-r Rahman frustrated the vindictive designs of his enemy by throwing himself down from a precipice while he was on his way — a.h. 84.1

The interest which this contest excited throughout the Khilafat seems to have invested the Prince of Kabul with a fictitious celebrity, insomuch that he is the hero of many Arab stories of the holy wars on the frontiers of Hind. Nevertheless there is no certainty as to the proper mode of spelling the name. The various readings of the European authors who have noticed him show how little the orthography is settled. Ockley2 calls him “ Zentil Weil,3 “Zehbil;” Reinaud,4 “Ratbyl” and “Zenbyl.” Wilson,5 “Rated, Ratpeil, Ratbal, Rantal, Zantil — variations easily accounted for by the nature of the Persian letters.” E. Thomas,6 “Ratpil;” Price.7 “Reteil,” “Ratteil,” or “Retpeil.”8

Price observes that the name bespeaks him to be either a Tartar or Hindu, and that the real name might perhaps have been Yittel, still common among the Hindus. Wilson considers it as a genuine Indian appellation ; Ratna-pala or Rutun-pal.9

1 Price’s Mahommedan History , Yol. i. pp. 455-463.

2 History of the Saracens, Bohn’? Edit., p, 490.

3 Geschichte der Chalifen, i. pp. 449, 461.

4 Memoire surl’Inde, pp. 71, 72, and 178. 5 Ariana Antiqua, p. 133.

6 Journal of the Royal Asiatic Society, Yol. xii. p. 344.

7 Retrospect of Mahommedan History, Vol. i., pp. 454-5.

8 [The Mujmalu-t Tawarikh (Paris MS. p. 274), says: —

“The kings of Kabul and Sind are called Eathil.” Ihn Khurdadba (Oxford MS. p. 26), has which M. Barbier de

Meynard ( Joicrn . Asiatiqae, 1865, p. 251), renders “ Le roi de Sistan Rotbil." Mas’udi (Paris Ed. ii. p. 87), has “Zenbil qui est reste commun jusqu’ a ce jour.” The various readings of the Jarni'u-l Hikdyut have been noticed in a previous page, supra, 178.]

Mas’udi, in his chapter in the Muruj, which is consecrated to the kings of Syria, makes mention of a prince who reigned in the Talley of the Indus, and who after having subjugated Eastern Persia, advanced to the banks of the Tigris and Euphrates. The name of this prince was Eanbal, under one of its various modifications, and he adds that the name formed in his time the designation of the indigenous princes of the country, and he calls the Buddhist princes of Kabul by this epithet, which he makes common to all. In this he is borne out by Tabari, and M. Eeinaud is induced therefore to consider the word significative.1 But it is not improbable that this assertion arises from the ignorance of the Muhammadans, and that they were ready to apply all the stories relating to the border chiefs of India to that one who had obtained the greatest noto- riety with historians by his transactions with the generals of the Khilafat, just as the Had'ika Sanai speaks of Jaipal being the king of India in the time of Bahrain, and Hatifi speaks of Eai Pithaura as the same even in the time of Timur.

The Jami’u-l Jlikaydt ascribes the name to a contemporary of Ya’kub Lais, which would make him one hundred and sixty years later than the invader of Syria, a long time for a title to have remained attached to a succession of petty chiefs. Moreover, at one time we find him ruler in Sind, at another in Kabul, though at the period spoken of those countries were not united under one dominion.

Khaki Shirazi says : — “ In the year twenty-two the province of Sijistan was conquered for ’Umar-bin Khattab, by the hands of ’Amru bin al Tamimi ; and in the same year Makran was subdued by Abdu-llah bin ’Abdu-llak Anan, who marched against it from Kirman. The ruler of that province, whose name in the language of the country was Zambil, was also ruler of Sind, and was killed.”

In the opening of the history of Mas’ud the Ghaznivide, by Abu-1 Fazl Baihaki, reference is made to the Palace of Eanbal, where it certainly seems to apply to an individual rather than a class.2

The Eanbal of whom we have been speaking as the opponent,

2 [Tabari, the Mujmal, and Mas’udi are all clear as to the import of the name, and its use as a dynastic royal title. Weil says it is “ a general name for the king ot the Turkomans, but more especially for the prince of Kabul and the territories between Hir&t and Kabul. — Geschichte , p. 449. "J

ally, protector, and betrayer of ’Abdu-r Rahman, must Lave been one of the Turkish dynasty of Kabul, of the Buddhist persuasion. TVe find, from the Arabic histories of the period, that some of his relatives still held dominion in Transoxiana, though the relationship was pro- bably rather that of tribe than family. If the family had been Hindu rather than Turkish, Ran-bal, “ strong in battle,” would have been sufficiently significative to render that the most likely reading of this disputed name. The probable prevalence, however, of the language of the Hindus in these parts might still have encouraged the use of the terms, notwithstanding that the Brahmans had not yet attained their supremacy.

In 107 a.h.=725-6 a.d., under the Kliilafat of Hasham, part of the dominions of Kabul was taken, but the capture of the town itself is not noticed.1

The lieutenants of the Khalifs A1 Mahdi and Ar Rashid took tribute from the Ranbal of Sijistan, proportioned to the strength or weakness of that prince, and named governors to the countries where Islam prevailed — a.h. 158-193=a.d. 775-809. When A1 Mamun was made governor of Khurasan, he demanded double tribute. He took Kabul, and the king submitted, and professed Islam. An agent on the part of Mamun resided in that city, and a post was established which enabled A1 Mamun to procure from it fresh myrobalans.2

After this we read nothing of Kabul till the time of the Saffarides — a.h. 256=a.d. 868-9.3 In the succeeding year4 Ya’kub Lais took Kabul, and made its prince a prisoner. The king of Ar Rukhaj was put to death, and its inhabitants forced to embrace Islam. Ya’kub returned to his capital loaded with booty, and carrying with him the heads of three kings ; and many statues of Indian divinities, which were amongst the booty, were sent to Baghdad for presentation to the Khalif.3

This Muhammadan conquest appears to have been more durable

! Gladwin’s Ayin A Jcberi, Yol. ii. p. 209. Price’s Maliommedan History, Vol. i, p- 567. 2 Biladuri, quoted in the Mem. sur I’lnde, p. 196-7.

3 Historia priorum regum Persarum, e tc., p. 19. 4 Tabakat-i Nasiri.

5 Ibn Asir, Kitabu-l Jihrist, and Ibn Khallikan, quoted in Mem. sur I'Inde. p. 209.

than the preceding ones, for we find coins of Ya’kub struck at Panjshir, to the north-east of Kabul, in the years 260 and 261 h.1

By referring to the passages given above from the geographers, we shall leam the state of the occupancy of Kabul from the time of the Saffarides to that of the Ghaznivides, which commenced as early as the time of Alptigin, according to the statement of Abu-1 Fazl, and it is probably to his time that the story related by Al-Biruni refers, where he states that when the Espehbed, or general-in-chief, had the gates of Kabul opened to him, the inhabitants imposed upon him the condition not to eat cow’s flesh or indulge in unnatural crimes.2 Neither condition is strictly observed by the modern occupants.

We will now proceed to examine more particularly the attempted identification of the several names of this series of Kabul kings : —

Jaipal I. Anandpal. Jaipal II. Bhimpal. e 0 e

Barhtigin has been already sufficiently remarked upon.

Kanah — Katorman. — Both these names have also been the subject of extended remarks. It will be observed that all the authorities quoted above from the original, make Kanak the last of the Turks, excepting only the Tar'ikhu-l Hind, which makes him only one, and the most famous one of the middle series of the Turkish kings for sixty generations. Allowing that Kanak is Kanishka, for which ample ground has already been advanced, this becomes impossible, and we must fall back upon the better authority of the Tar'ikhu-l Hind, and consider the Katorman or Laktuzaman as the last. In the more modern narratives of Rashidu-d din and Binakiti we must place a full stop after “ Kanak returned to his country.” Then proceed, “the last of the kings was the Katorman.” This requires

1 Frselin Summarische Uebersicht, etc., and Bulletin de r Academie, Tom. x. p. 81.

no violent alteration of tlie text. Indeed the mere omission of from the Arabic, and .1 from the Persian reconciles everything, and this last omission is actually made in the British Museum MS.

The writers themselves knew little of the state of the case, and wished merely to translate A1 Birum, who knew well enough what he was writing. Por instance, Binakitf wishing to reduce the narrative of the JamV, makes it appear that Ujen was the predecessor of Kanak. Haidar Bazi, again, among the names of the illustrious kings of India who succeeded Basdeo (here meant not for him of Kanauj, but the great Krishna) mentions Arjun and Jasand (the former being manifestly the famous hero of the Maha-lhardta, and the latter Jarasandha), and “after him came Kanak, Chand.” This, thorough indifference to correct chronology, enables us to see that by Ujen is meant Arjun, the senior of Kanak by several centuries. Mr. Thomas is persuaded that to this Kanak, the last of the Turks, are to be ascribed the coins which bear the name of Sri Yanka Deva “of the elephant-and-lion type of coin, which preceded the bull- and-horseman money introduced by the Brahmans. The similitude of names and the needful correspondence of all available evidence are surely sufficient to authorise our indicating Yanka Deva” as the Kanak above mentioned. This is by no means admissible, and he has himself since found that the real reading on the coin is “Yarka,” and has, consequently, altogether abandoned this speculation.1

Kalar “ is, we have little doubt, the Syalapati of our coins. There is less difference in sound between Syala and Kalar than would at first be imagined ; so that if our translator, A1 Biruni, wrote his Arabic version from oral tradition, this slight change in the initial pronunciation of the name would be fairly probable.” This is carrying speculation to an extreme, and there is no warrant whatever for the presumed identification.

1 [Mr. Thomas, who might naturally desire to reply to these early criticisms on his confessedly initiatory essay on the coins in question, agrees with me in thinking that Sir H. Elliot’s text should be preserved intact in the present pub- lication, without comment or controversy on his part. This kind of knowledge is happily progressive, and many valid advances may be admitted to have been made between the theories of 1817 and 1868, without compromising the original author, or his censor of days gone by. Many of the objections here advanced have already been answered, in anticipation, by Mr. Thomas, in his edition of JPrinsep’s Essays (London, 1858), an extract from which will be found below (p. 428).]

It is to be observed that the Jdmi'u-t TawarVkh and its followers omit all notice of Kalar, making Samand the immediate successor of Kanak.

The Syala or Syal-pati (7 roTt? in Greek), of whom so many coins are found in Afghanistan, was probably a leader, and, perhaps, even the progenitor of the Syal Jats of Jhang Syal and other localities in the Panjab.

Samand. — Coins of Samanta, or Samanta Deva, are found in great profusion not only in Afghanistan, but throughout the Panjab and the whole of Northern India, and one has even been found in the province of Posen.1 Mr. Thomas is of opinion that this is owing to his having called in the coins of his Buddhist predecessors, in order to give prevalence to his own creed of Brahmanism by the substitution of the bull-and-liorseman type for that of the elephant- and-lion, which is considered emblematic of Buddhism ; 2 but this supposition seems defeated by the fact of our finding Samanta coins with the elephant also upon them. The name of this reviver of the old faith became so celebrated, that we find it upon the coins of his successors, extending even down to the Muhammadan conquest of Dehli, in 1192 a.d., and the coins of Rai Pithaura.

Professor Wilson attributed these coins to a Rajput prince, who lived many years afterwards. M. Reinaud never hesitated to recognize in these medals the name of the king of Kabul, and his opinion was confirmed by the examination which M. Adrien de Longperier made of them.3

It may be considered presumption to oppose such an array of authority in favour of this identification, but, nevertheless, I hesitate to concur in it without more cogent arguments than those that have yet been adduced. Putting aside the improbability that one man’s name should be stamped on a series of coins, extending through more than two centuries, sometimes in supercession, and sometimes in con- junction with, that of the reigning monarch — and that, too, even in the case of the later Ghazni vides — there seems so obvious a solution

1 M. Longperier in Fragments Arobes et Persons, p. 223,

• Journal Royal Asiatic Society, vol. ix. p. 181.

3 Memoire sur V Inde, p. 212. Journal Asiatiquc, Feb. 1845, p. 192, and Frag- ments Arabes et Persons, p. 219.

of this continuance of a single name, that it requires far less boldness to adopt this simple explanation, than to seek grounds for establishing a position which, from its many improbabilities, is always open to question. It may, perhaps, be admitted that the coins which bear the simple name of Sri Samant Deva are to he referred to the Samand of Abu Rihan ; but even that admission is open to objection, there being a double mis-spelling in the name, for in the former we have a short a instead of a broad one, and a t instead of a d.1

It appears to me, then, that Samanta, whenever it is found with another name, is throughout merely a title, meaning the warrior, the hero, the preux chevalier , the leader of an army, the Amir ; and that after being used concurrently with Sri Hann'r on the later Ghaznivide coins, it was by the early Ghorian monarchs altogether displaced by that more appropriate title.

At this latter period the prevalence of the title of Samant is obvious from its frequent use by the bard Chand, who has celebrated the exploits of Rai Pithaura, and his three himdred Samants, or stalwart knights.

Katnlua. — Mr. Thomas wishes to appropriate to this monarch a medal bearing the legend of Khvadavayaka or Khedavayaka, while he confesses that even to liberal ears these names are not quite accordant in sound. He then seeks to justify the appropriation by mutations, blots, or intermixture of letters.2 We must reject this, it being not worthy of the least credit ; and the discovery of the name of Kamliia in another history sets the question at rest, and establishes the correctness of A1 Biruni.

This discovery is in other respects important, as enabling us to fix a synchronism by which we may conjecture the periods of the other monarchs of this dynasty. In one of the stories translated from the Jdmi’u-l Hikaydt ,3 it will be found that he was a contem- porary of ’Amru Lais, who reigned between 265-287 a.h.— 878-900 a.d. Kamlua is there called the Rai of Hindustan, and he must have ruled sometime within this period.

If we admit that these names represent a continuous series of

1 [Longperier reads the name with a long a— Samanta. See Fragments Arabes et Fersans, 221-223.]

successive monarchs, and not rather those who alone were conspicuous, we shall have to place the commencement of Kamlua’s reign as late as possible within the twenty-two years above-named. For we must connect it with another synchronism which we obtain from the same Jami'u-l Hilidyat, wherein we learn that Mahmud was only fourteen years old when the defeat of Jaipal occurred near the miraculous fountain, which — as he died in a.h. 421, 1 when he was sixty-three years old — reduces that date to 372 a.h., or 982-3 a.d., fifteen years before the death of Subuktigin.

Jaipal died in 1002 a.d., and it is evident from the statement in the Tar'ikh-i Yamtni, that he was then a very old man. He had opposed Subuktigin, while yet that warrior was only general of Alptigin, and therefore before 976 a.d., making his reign at least a quarter of a century. If we assume that Kamlua’s reign commenced in 890 a.d., being about the middle of that of ’Amrii Lais, we shall have to divide the period extending from 890 to 1002 a.d., between the reigns of Kamlua, Bhfm and Jaipal, being an average of thirty-seven years for each, which seems much too long. But as there is no dis- puting the dates, we must admit the long duration of 112 years for only three reigns, or admit that the names of unimportant monarchs have been omitted ; just as in the case of the Turkish series, of which only Kanak is mentioned, between the first and last of the dynasty.

In the same way, between Kalar and Samand, and Samand and Kamlua — there may have been other omissions, and even long inter- regna of Muhammadan supremacy ; and we may thus throw back the period of the Brahmanical revolution to an earlier date than has yet been conjectured. It must be confessed this would relieve us of some difficulties, and enable us to dispose of other names of this series, of which we have incidental notice elsewhere : as, for instance, in the Sairu-l Mululc, where we meet with the name of Lomak.

Syala, Khedavayaka, Yarka, and even Ranbal may have been individuals of the Kabul series, either Turk or Hindu, though not honoured with distinct mention by Abu Rihan. Numismatists,

‘•April, 1030. See the inscription on his tomb in Thornton’s Gazetteer of the Countries adjacent to India , vol. i. p. 200, [and Journ, It. A. S., xvii. p. 161.]

indeed, are now so certain that these coins do belong to the Kabul series, and trace with such confidence the relative antiquity of each extant medal from the difference in devices and execution, that we may readily concede the point to such able and experienced enquirers. All that is required is that there should be no unnatural forcing to suit preconceived theories.

Mr. Thomas has conjectured on other grounds that the acces- sion of Samand occurred in 935 a.d.,1 but his computation does not rest on any such specific dates as the two mentioned above, and he considers that, under any circumstances, it is imperfect, and that “ the utmost the materials at our command enable us to assert with any degree of certainty is that Syala’s usurpation took place early in the tenth century but even this certainty is dispelled by the establishment of the fact that Kamlua was, unquestionably, a contemporary of ’Amru Lais. Altogether, we may consider the sub- version of the Turk by the Brahman dynasty to have occurred about 850 a.d., shortly before its capture by Ya’kub Lais ; and as it appears from the Arab geographers that Musulmans held the castle, it is evident that the Brahmans were only occasionally dominant, and did not hold their power without long and frequent interruptions.

Bhim. — The coins of Bhim are found in Kabulistan, but are seldom, if ever, met with in India. There is no reason to doubt that this is the same Bhim as the Sri Bhim Deva of the bull-and-horse- man series, and this is the only one of which the identification can be admitted without question.

M. Reinaud considers that this Bhim is the one mentioned by ’Utbi and Firishta as the founder of Nagarkot;2 but there is more reason to believe the hero of the Mahd-lhdrata to be the one indicated.

Jciipal 1. — It is strange that no coins of Jaipal are found. Firishta calls him the son of Ishtpal,3 and distinctly avers that he was a Brahman, and Birum also includes him in that dynasty ; but the introduction of the term Pal, which is now continued to the close of the dynasty, might incline us to suppose that a new family had com-

1 Journal Royal Asiatic Society, vol. ix. p. 179.

3 [Briggs’ translation says “ Ilutpal,” but tbe lithographed text has “Ishtpul.”]

menced. This seems in other respects not improbable, for in the opening of the Tar'ifch-i Yam'mi we find Jaipal’s western border extended no further than Lamghan, Kabul being already in posses- sion of Subuktigin. It seems probable, therefore, that the succession of the real Kabul sovereigns ceased with Blnm, and that the king of Northern India succeeded to the paramount sovereignty which, as far as the Muhammadans were concerned, had hitherto been held by the ruler of Kabul. It is a mistake to suppose that Jaipal was king of Dehli. It does not appear that any such place existed in his time, and Abu-1 Fida’s determination of its latitude and longitude on the authority of the Kanun-% Mus’udi is a misquotation, which it is of importance to correct, for there is nowhere mention of Dehli either in that work or in the Tar'ikhu-l Hind. The principal places of his residence appear to have been Lahore, Bhera. and Waihind ; and it may be doubted if any of these places, except perhaps the last, had been held by the kings of Kabul.

The assertion that he was a Brahman probably arises from ignorance on the part of Firishta. A1 Bfrum is not specific in his statement that he was a Brahman, but merely includes him in the dynasty which commenced with a Brahman, and he may no more have been really of that caste than were the Bahmam sovereigns of the Delchin, though they were called after one. The term Brahman, in the conception of a Musulman, might merely imply that he maintained the doctrines of that faith, and from his position was its staunchest defender and champion. There seems ground to suppose he must have been a Bajput, and some reasons have been assigned in the note on Mahmud’s invasion for considering him a Bhattf.

Anandpal. — Mr. Thomas observes1 that the coins of Anandpal are common, and are plentiful in the Panjab and the northern parts of the Ganges Duab. But these are evidently to be referred to the monarch of Delhi, who lived a century and a half later, and we have to deal with Anandpal not Anangpal. ’Utbi calls him Andpal.

Jaipal //.—This is not the name given by A1 Bi'runi, where it appears more like Tardijanbal, and in the other authors who men- tion him it goes through various forms. Tadan Jaipal, Nanduwa 1 Jour, E. A. S., ix. p. 121, [and later, Ptinsep’s Essays, i. 330.]

Jaipal, Turu Jaipal, Parou Jaipal, Nardajanpala, Niranjanpal, Tasdar Jaipal, and many more.1 The latest reading proposed by M. Reinaud is Trilochan Pal, after the “three-eyed” Siva. Persian authors generally call him Nabira Jaipal, or the grandson of Jaipal, and in that relationship no doubt he stood to the first Jaipal. Hence Dow calls him “ Pitterugepal.” The real name was, perhaps, Pur Jaipal, or Jaipal junior, Jaipal the son or grandson. A1 Biruni tells us that his father Anandpal was an inveterate enemy of the Musulmans from the time that Pur Jaipal was taken prisoner, but Pur Jaipal himself was well disposed towards them.

According to ’Utbi we find him holding dominion as far eastward as Kanauj and the Rahib, respecting which the note on the ninth and twelfth expeditions of Mahmud may be consulted. The same author mentions another son of Anandpal, by the name of Brahman Pal, who is probably a different one.

Abu Ri'han informs us that ho was killed in 412 a.h.=1021-2 A.n. It does not appear exactly when he began to reign, but he certainly opposed Mahmud during the Kanauj campaign in 409 n.

Bhim Pal. — In him we have the last of the dynasty of Kabul and Northern India. As he is mentioned by Abu Rihan, he must have succeeded to some remnant of his father’s domains ; but it does not appear that in his time he contested the advance of the Muham- madans, though before he ascended the throne we find him taking an active part in defending his father’s dominions, under the name of Nidar Bhim, “Bhim the Dauntless.”3

From his letter to Chand Rai, which is recorded by ’Utbi, it would appear that he was inclined to peaceful counsels, and that bitter experience had taught him the hopelessness of contending with his relentless and sanguinary rivals.3

From a statement in the Tar'ikhu-l Sind , we may infer that his capital was Bari, to the east of Kanauj.

Neither of Bhim Pal, nor of any other of the Pal family, are any coins extant.

Bhim Pal survived his father five years, and died, therefore, in 417 a.h., the eventful year of the capture and plunder of Somnat. Haidar Razi gives nine years as the period of his reign.

1 [See supra, pp. 45-47.] 2 [Supra, p. 38.] 3 [Supra, p. 4S.]

Extract of Mr. Thomas' Edition of Prinsep's Essays, (1858. Vol. I. p. 331), referred to in page 9 supra.

“ Before I leave the subject, I may be permitted to make some observations in reference to an original suggestion of my own, that the Sri Hamirab, on the reverse of the immediately succeeding Moslem coins, was designed to convey the title of the spiritual representative of the Arabian Prophet on earth, embodied for the time being in the Kbalif of Baghdad. Sir H. M. Elliot, placing himself under the guidance of Capt. Cunningham, has contested this inference. I am not only prepared to concede the fact that Muhammad bin Sam uses this term in connection with bis own name on the lower Kanauj coins, but I can supply further indepen- dent evidence, that my opponents could not then cite against me, in the association of this title with the name of the early Sultans of Debli in the Palam Inscription (1333 Vikramaditya) ; but, on the other band, I can claim a still more definite support in an item of testimony contributed by the consecutive suite of the selfsame fabric of coins, where the (, hamirah ) is replaced by the word

( [khalifa ). As far as I have yet been able to ascertain, this transition first takes place on the money of ’Alau-d din Mas’ud (639-644 a.h.) ; and here, again, I can afford, in all frankness, to cite further data that may eventually bear against myself, in recording that this reverse of Sri Khalifa is combined in other cases with a broken obverse legend of . . . • • • which, being interpreted to

stand for the Amiru-l Mumirnn of the Arabic system, may either be accepted as the Sanskrit counterpart legend of Altamsh’s anonymous coins in the Persian character,”1 or be converted into a possible argu- ment against my theory, if supposed to represent the independent spiritual supremacy claimed by subsequent Sultans of Dehli ; which last assignment, however, will scarcely carry weight in the present state of our knowledge. As regards the difficulty raised respecting the conventional acceptance of the Sri Samanta Dev a of the coins as an historical, rather than an individually titular, impress, I have always been fully prepared to recognize the linguistic value of the 1 Pathan Sultans of Dihli, by Ed. Thomas. London, Wertheimer, 1847 ; p. 17-

word Samanta, and yet claim to retain the Sri Samanta Deva — which comes down to us, in numismatic sequence, in the place of honour on so many mint issues — as an independent name or title, to which some speoial prestige attached, rather than to look upon it as an ordinary prefix to the designation of each potentate on whose money it appears. And such a decision, in parallel apposition to the succes- sion of the titles of Hamira and Khalifa , just noticed, would seem to be strikingly confirmed by the replacement of this same legend of Sri Samanta Deva on the local coins of Chahad Deva, by the style and title of the Moslem suzerain, to whom that raja had eventually to concede allegiance.

The two classes of coins to which I allude may, for the moment, be exemplified, the one in the type given in £ Ariana Antiqua,’ xix. 16 ; the other in pi. xxvi. fig. 31, Yol. i. fPrinsepJ.

The former, when corrected up and amplified from more perfect specimens, will be found to bear the legends : Obv.

Rev. -q — while the latter will be seen to

I understand this obverse legend to convey, in imperfect ortho- graphy, the name of Shamsu-d din Altamsh — whose other coins, of but little varied type, have a similarly outlined name, with the Moslem Sri ITamirah on the reverse.

The contents of this volume relate more especially to the history of the Ghaznivides. It therefore seems expedient to take a general review of the authors who have particularly treated of that dynasty.

First in order comes ’Utbf, who has already been sufficiently noticed. It may be remarked generally that he is deficient in dates, and, though the chief and earliest authority on all which relates to the early invasions of India, he evidently had no personal knowledge of that country, a circumstance which of course greatly detracts

from his value. He is fuller in the reijm of Subuktigdn and the transactions in Turkistan than any of his successors.

Thirty years later comes Abu-1 Fazl Baihaki, of whose voluminous and important work only a portion has come down to us.

After, an interval of more than two centuries follows the Nizamu-t Tawdr'ikh, composed in 674 h., about a century after the extinction of the dynasty. The short notice which this work devotes to the Ghaznivides has been translated as an extract from that work, but it is of little authority, and confuses dates irremediably towards the close of the dynasty, in which the transactions were carried on too far eastward to he within the foreign ken of ihe author. Indeed he confesses that he knows nothing of their successors, the Ghorians, beyond the names of three of their kings.

The next, hut after a period of two hundred years from ’Utbi is the Tabakat-i Ndsiri, the chief value of which is that it quotes the lost volumes of Abu-1 Fazl Baihaki. It is for this reason, however, greatly to be regretted, especially as he is one of the earliest Muhammadan authors who wrote in India, that his notice of Mahmud’s reign is so very curt ; for it is that in which we most feel the want of Baihaki’s familiar gossiping narrative. It is true he is quoted in the JamVu-l Uikaydt , Tdrikh-i Guz'ida , Rduzatu-s Safa, and Firishta; yet it may he doubted if any except the author of the first ever saw his Tar'ikh-i Ndsiri, which is mentioned by name in the Tabakat. In some of the other Ghaznivide reigns, this work differs from others, as will be seen from the passages which are extracted in the article Tabakat-i Nasiki in this volume.

The great copyist and extractor, Rashidu-d din, follows after the lapse of about twenty years. In his JamVu-t Tawdr'ikh, he follows ’Utbi implicitly, as far as the Yam'ini extends, taking out not only his facts, but giving a literal translation of that work, even to the images and similes. So little does he attempt to improve upon the Yamini, that he even leaves out the important expedition to Somnat, which was undertaken after the close of that work. This resource fails him altogether in the later reigns, which are conse- quently very unsatisfactorily disposed of in the Jami'u-t Tawdr'ikh }

About twenty years later follows the Tar'ikh-i Guz'ida of Hamdu-lla 1 [See an article by Major Lees, in Jour. It. A. S., Yol. iii. N.S., 1868.]

Mustaufi — although he mentions the Mahdmdt of Abu Nasr Miska’ti, and the Mujalladdt of Abu-1 Fazl Baihaki, he does not appear to have read them : at least he gives no information derived from them, and altogether his account of Mahmud’s reign is very meagre. He mentions the names of the towns taken by him, omitting, however, all notice of Somnat, and without stating the dates of their capture. He is so often quoted by Mirkhond, Khondamir, and Firishta, that he has had more credit than he deserves in this portion of his universal history.

After a long interval of about a century, we have Mirkhond, who in his Rauzatu-s Safa has given us the first detailed account of the history of the Ghaznivides. It is founded in the early portion upon the Yam'ini, but in later reigns rests upon some other authorities which are not quoted. Those which are mentioned, as the Ndsir'i and Guzida, are too meagre to have furnished the fuller information found in the Rauzatu-s Safa. This portion has been translated by F. Wilken into Latin, and published with the original text at Berlin in 1832, under the title of Historia Gasnevidarum. He has added in footnotes passages from Firishta and Haidar Bazi, where the details are more complete than in the Rauzatu-s Safa. Haidar Bazi, however, is no original authority. I have found all the passages, except two, quoted by Wilken to be word for word the same as the Tdr'ikh-i Alfi, even where other authorities are quoted, as Ibn Asir, Ibn Kasfr, and Hafiz Abru. The chief omission to be noted in. Mirkhond’s account is that of the expeditions to India intervening between those of Kanauj and Somnat, and the attack upon the Jats of Jud after Mahmud’s return from Somnat.

Mirkhond is followed by his nephew Khondamir in the Khulasatu-l Aklibdr and the Hab'ibu-s Siyar. The former has been translated by Price with additions from Firishta, and from the latter a trans- lation will be found in a later volume of this work. He follows the Rauzata-s Safa closely, and has no new authorities, omitting some passages, hut dealing more copiously with the biographies of cotem- porary poets and ministers. Altogether, Mirkhond’s narrative is preferable, and in this, as well as in many other portions of his history Khondamir might have saved himself the trouble of attempt- ing to rival his uncle.

The next authority of any value is the Tdr'ikli-i Alf'i. Like as in other portions of that work, it is, in the history of the Ghaznivides, also somewhat deficient in connexion, and troublesome, from adopt- ing a new era ; but. altogether, it is copious and correct. ’Utbi and Mirkhond are the chief authorities of the Tar'ikh-i Alfi, but something is added from the less known histories, which have already been mentioned as being quoted at second hand by Haidar Bazi. It is to be regretted that Abu-1 Fazl Baihaki is not amongst them. Here also we have no detailed account of the Indian expeditions between those of Kanauj and Somnat, and that to Thanesar is not mentioned.

Nizamu-d din Ahmad, in his Tabakat-i Akbari, gives a succinct account of the history of the Ghaznivides, and is particular in mentioning his dates. He notices very cursorily the events in Turkistan, Sistan, and ’Irak, confining his attention principally to what related to India. In his work we, for the first time, find mention of several expeditions to India, which are passed over by his predecessors ; and it is, therefore, to be regretted that he does not signify on what authority he relates them. The only probable source, among those mentioned as his general authorities, is the Zainu-l Akhbdr. Nizamu-d din is followed closely by Firishta.

’Abdu-1 Kadir, in his Tar'ikh-i Baddun'i, follows Nizamu-d din im- plicitly ; but, in order to show the variations, he occasionally quotes the Nizdmu-t Tawdr'ikh, and the Lubbu-t Tawdr'ikh. He adds, also, some verses of poets who were contemporary with the Ghaznivides.

The Huntakhabu-t Tawdr'ikh of Khaki Shirazi is very brief, and scarcely deserves notice. It chiefly follows the Habibu-s Siyar.

We next come to the history of Firishta, which gives the most complete and detailed account which we have of the Ghaznivides. Dr. Bird complains of the author’s ignorance of the geography of Upper India ; but he has exhibited no more than his predecessors, and in one or two instances attempts corrections. His chief resource is the Tabakat-i Akbar'i, but he has also used the Tdrikh-i Yaminl, the Tar'ikh-i Guzida, the Rauzatu-s Safa, and the Habibu-s Siyar. Some of the other works which he quotes there is reason to believe he never saw. The translation by Briggs is generally correct and faithful in this portion, and there are no omissions in it of any great consequence.

The Khulasatu-t TaivariJch discusses this history in a peculiar fashion of its own. It omits all notice of transactions on the frontiers of Persia and Turkistan, and confines itself solely to India, insomuch that it leaves out whole reigns in which the sovereign had no connection with India : and, in consequence, preposterously con- fines the whole number of reigns to seven only. There is no other novelty in this chapter, except that it substitutes two new readings of places, which if they are derived from the history of Mahmud by ’Unsuri, which is quoted in the preface, may be considered authentic.

These are all the authorities which it seems necessary to notice, as all the subsequent ones follow in the wake of Firishta. Abu-1 Fida, Ibn Shuhna, Ibn Asir, Ibn Kasir, Nikbi, and Lari, have had all that is valuable in them extracted by the diligence of Euro- pean authors, who have translated, abridged, or commented on the reigns of the Ghaznivides. The Turkish histories of the period, such as the Ndkhbatu-t Tawar'ikh, and the work of Munajjim Bashi, we may fairly presume to have been exhausted by the industry of Hammer-Purgstall amongst the fourteen different histories which he quotes as authorities upon Mahmud’s reign — so that the only hope now left us for ascertaining any new fact with respect to the history of the Ghaznivides is in the recovery of the missing volumes of Memoirs, which we know to have been written by contemporary writers, and to have been in existence less than two centuries ago — such as those of Abu-1 Fazl Baihaki, Abu Nasr Mishkani, and Mulla Muhammad Ghaznawi. The Makamat of Abu Nazr Mishkati 1 (Mishkani) is mentioned by Firishta (Briggs I. 32 and 97), and the same author is referred to in Wilken (Gasnevidarum, p. 189). Firishta quotes from him the anecdote about Mas’ud, which has been given from the Talalcdt-i Nasir'i {supra, p. 271), and which is there also attributed to Abu Nasr Mishkan. The Tdr'ikh-i Mulla Muhammad Ghaznaw'i is mentioned by ’Abdu-r Rahman, who wrote the Mir-dtu-l Asrar and Mir-dt-i Mas'ud'i, in Jahangir’s time. The author was contemporary with Sultan Mahmud, of whom his work is said to give an ample account.

1 [In Briggs’ translation, the name is written “Mukutty.”]

The times, places, and numbers of Mahmud’s expeditions to India have offered great difficulties to those who have dealt with the his- tory of that ferocious and insatiable conqueror. We look in vain for any enquiry on the subject from the native historians of this period, who, in their ignorance of Upper India, enter names and years without the scruples and hesitations which a better knowledge or a more critical spirit, would have induced.

It is only when European authors begin to discuss the matter that we are taught how many difficulties there are to solve, how many places to identify, how many names to restore. Those who have added most to our knowledge of this period, and have occasionally interspersed their narratives or notes with illustrative comments, and who will be quoted in the course of this Note, may be thus named in the order of their publications : — D’Herbelot,1 De Guigues,2 Hunt (?),3 Dow,4 De Sacy,6 Mill,® Wilson,7 Audiffret,8 Rampoldi,9 Briggs,’0 Wilkcn,11 Bitter,12 Bird, 13 Hammer-Purgstall,14 Elphinstone,15 and Beinaud.16 It is needless to mention Gibbon, Malcolm, Conder, Gleig, Murray, and others, whose works, however useful, are mere copies and abstracts of others, and add nothing to our previous information.

It has been usual to consider the number of Mahmud’s expeditions

1 Bibliotlieque Orientate , Art. “Mahmoud.” Paris, 1697.

2 Histoire Generate des Huns , Tom. II. Paris, 1756.

3 Modern Universal History, Yols. II. and III. London, 1766.

4 History of Eindoostan, Vol. I. London, 1768.

6 Notices et Extraits des Manuscripts , Tom. IV. Paris, 1798-9.

6 History of British India , Vol. II. London, 1818. 7 Hid , 1840.

8 Biographie Universelle, Art. “Mahmoud.” Tom. XXVI. Paris, 1820.

10 History of the Mahotn. Power in India , Vol. I. London, 1829.

11 Historia Gasnevidarum. Berolini, 1832.

12 Lie Erdkunde von Asien, Vol. IV. Part 1. Berlin, 1835.

14 Jahrbiicher der Literatur, No. 73. "Wien, and Gemaldesaal der Lebens- beschreibungen, Vol. IV. Leipsig, 1837.

is History of India, Vol. I. London, 1843.

16 Memoire sur Vlnde in the Memoires de I'Instilut, Tom. XVIII. Paris, 1849.

to India to be twelve. The first authority for this number is Nizamu-d din Ahmad in the Tababat-i Akbari ; and as Dow has also numbered them as twelve, most English authors following him as the standard, have entertained the same persuasion. But it is curious to observe that, while Nizamu-d din mentions that there were altogether twelve, in recording them seriatim, he enumerates no less than sixteen ; and Dow, while he marginally notes twelve, records no less than fifteen different invasions. Even Elphinstone, though he notes twelve, records more. The Khulasatu-t Tawar'ikh gives twelve, and confines itself to that number, or in reality only to eleven, as by some mistake an expedition to Kashmir and Kalinjar are placed in one year, and the tenth expedition is omitted. The Alchbar-i Muhabbat follows it in both errors. I will not attempt to maintain this established number of expeditions, but will consider them in the actual order of their occurrence.

First Expedition. — Frontier Towns, a.h. 390 (1000 a.d.) — Nizamu-d din Ahmad and Eirishta mention that about the year 390 h. Mahmud marched in the direction of India, and, after taking many forts and provinces, and establishing his own governors in them, he returned to Ghazni. This rests solely on the authority of these two authors, and is not supported by the Tarilch Yam'ini ; but there is no improba- bility in the statement.

It was to have been expected that Mahmud, after establishing himself on the throne of Ghazni, would have embraced the first opportunity of invading India ; for, while yet a prince, he had seen how easily the hardy warriors of Zabulistan had overcome the more effeminate sons of India. His father Subuktigin is described in the Yamin'i as making several attacks upon the country of Hind, inde- pendent of the three which are more specifically mentioned, the scene of which was Kusdar and Lamghan. Even during the fifteen years of Alptigin’s reign, Subuktigin is represented by Firishta in an untranslated passage to have made frequent attacks upon India, and even to have penetrated as far as Sodra on the Chinab, where he demolished idols in celebration of Mahmud’s birth, which, as it occurred on the date of the prophet’s birth, Subuktigin was anxious that it should be illustrated by an event similar to the destruction of the idols in the palace of the Persian king

by an earthquake, on the day of the prophet’s birth. In the words of the Bostan : —

Near the Lamglian valley two actions were fought, or more pro- bably in the valley of J alalabad, for as the plural, Lamghanat, is fre- quently used, there seems reason to believe that the valley to the south as well as the north of the Kabul river was included in that province. The first action fought in this neighbourhood was brought to a conclusion by the effect of the miraculous fountain or stream in the hill of Gliuzak, which emitted storms, thunder, and cold, when- ever some impurity was cast into it. A more particular account of this will be found in the extracts from the Yammi and the Jami'u-l Hikaydt.1

What could have given rise to this extraordinary story is not easy to conceive, and no one has attempted an explanation. The most probable solution seems to be that a snow-storm came on, and not only harassed but alarmed the Hindus, who had never witnessed such a thing before ; for it is quite compatible with probability that although the Lamghanat were then included in the country of Hind, yet that the soldiers, who, for the most part, came from the more eastern provinces, might never have seen a fall of snow. It is to be observed that the Tabakat-i Akbar'i expressly says that Jaipal and the Hindus were unaccustomed to the cold, and that was the reason why they suffered more than the Musulmans. It may fairly be surmised, then, that the snow and frost totally paralysed the Hindu warriors, and were felt as grievously by them as, nine centuries afterwards, by Indian and British troops combined, when they sus- tained the most grievous disaster that has ever befallen our nation. It is an extraordinary coincidence that the very scene of this first and last defeat of an Indian army was the same — what wonder if the cause also did not differ ?

The minds of the natives of India would naturally have tried to account for such a supernatural phenomenon as a fall of snow, and superstition was at hand to render her assistance.

There was a stone, celebrated amongst the Turkish nations, which had the peculiar property of causing rain, and hail, and snow, and excessive cold, and violent tempests, if the possessor, after repeating the name of God, and breathing upon it, threw it into the water. This stone is called the “ Yedeh,” or “ Jedeh.” The first stone of the kind was said to have been given to Japhet by Noah, to whom the secret was disclosed by Gabriel. The stone came into the possession of Turk, the eldest son of Japhet, and in an action which was fought between him and his nephew, for the possession of the stone, the latter was killed ; and, as he was the father of the Turko- mans, this stone is said to be the cause of the unceasing enmity between that tribe and the Turks. Subsequently, the art of using this stone was more generally disseminated, and occasioned magicians to be generally called “ yedehehis ; ” and we have frequent mention of its use in Mongol history for purposes similar to those for which we suppose it to have been applied on the present occasion. As early as the year 2634 before our era, we find the following statement in a quotation by M. Klaproth, to prove the antiquity of the compass among the Chinese : “ Tchi-yeou raised a thick fog, in order that by means of the darkness he might spread confusion in the enemy’s army. But Hiuan-yuan constructed a chariot for indicating the south, in order to distinguish the four cardinal points.”1

In an action between the Mongols and Chinese, with respect to the latter, Basliidu-d din says : “In consequence of the arts of the magician, the Chinese felt, in the middle of summer, a temperature which they had never experienced, even in winter, and were para- lysed.” Bergman says that the stone used at present among the nomadic nations is the Bezoar. Marco Polo, also, speaking of a country not far from the confines of India, says : — “ When the Carannas wish to overrun the country and rob it, they, by their enchantment and diabolical agency, cause the day to become dark, so that you can see to little or no distance.” In the mountains between Kashmir and Tibet, there is a lake, into which, if animal flesh is thrown, we are informed by Abu-1 Pazl, that a storm of snow or rain will arise. There is said to be a similar one at Damaghan, in

1 Lettre a M. A. Humboldt sur Vinvention de la Boussole. Paris, 1836 ; and Mr. Davies, in the British Annual for 1837.

Tabaristan, and Zakariya Kazwini mentions one near Ghazni, which is, no doubt, the one alluded to in Subuktigin’s battle with Jaipal. Altogether, we may consider Jaipal’s army to have been surprised and paralyzed by a snow-storm, and that superstition ascribed the unusual visitation to the “ Yedeh ” stone.1

Second Expedition. — Peshdwar — Waihind. a.h. 391-2. — Mahmud left Ghazni in Shawwal, 391 h., and a severe action took place on the 8th of Muharram, 392, at Peshawar, in which he was completely victorious, and Jaipal and fifteen of his principal chiefs and relations were taken prisoners, after the loss of 5000 men.

He is then represented by all the later authorities to have marched from Peshawar to Batinda, and invested it. Elphinstone observes that Batinda is beyond the Sutlej, “ and seems formerly to have been a place of more consequence than its situation in a sort of

■ desert would promise. It is said by Colonel Tod to have been the residence of the Baja of Lahore, alternately with the capital, from

■ which he took this name. As the battle of Peshawar was on the 27th of November, Mahmud would reach Batinda towards the end of the cold season, when the rivers of the Panjab, though not all fordable, would olfer little obstruction to cavalry.” Dr. Bird also speaks of Batinda as being in the most easterly and inaccessible part of the Panjab kingdom, and following the Tabakdt-i Alcbari and Firishta, says that Jaipal used to reside there. The latter indeed says he resided there for the convenience of opposing the Muham- madans— which is an absurdity, if we are to understand the most eastern city of his dominions. Bampoldi, with his usual confusion of names and places, makes his residence Multan.

All these difficulties about Mahmud’s movements are at once obviated by correcting the reading, and rejecting Batinda altogether. The real name is Bihand or Waihind, as is plainly indicated in the Yam'ini.2 It was a place of considerable importance, on the western

1 Kespecting this stone and these fountains, further information may be obtained by referring to Bergman, Nomadische Streifereien unter den Kalmiiken, Th. iii. p. 183. Miles, Shajrat ul Atrak, pp. 24, 26, 66. Gladwin’s Ayeen Akberee , Yol. II. p. 134. Marco Polo, Murray's Ed., p. 221. Modern Universal Bistory, Vol. IV. p. 417. D’Ohsson, Histoire des Mongols, Tom. II. p. 615. Khulasatu-t Tawarikh, Art. “HumayCin.” Mir-atu-l Istildh, Art. “Yedek.” Asdru-l Bildd and Bahru-l Bulddn , Art. “Ghazni.”

2 [Ibn Asir gives the name of the place correctly as “ 'W’aihand.”]

bank of tbe Indus, about fifteen miles above Attock, on the old high road from Lahore to Peshawar, and only three marches from the latter. It was the capital of Eastern Kandahar, and is noticed by Biruni, Baihakf, and Abu-1 Fida, from which latter author we leam that its foundation is attributed to Alexander the Great. The name is now Hund, and while I was in the neighbourhood I could not find that even any first syllable was ever added to it, either by natives or strangers.

By the capture of Waihind, Mahmud’s progress becomes easy and natural, and instead of having to cross and recross several foaming streams and marching through a hostile and difficult country, he has not yet crossed even the Indus.

Third Expedition. — Bhera ( Bhatia ). a.h. 395 (1004-5 a.d.) — After a rest of three years, during which attention was occupied by affairs in the west, we find Mahmud returning to India to take the city of Bhateea (Briggs), Battea (Dow), Bhatia (Elphinstone), Bhatnah (Bird), Bahadiyah ( TJniv. Hist.), Bhadiyah (Bampoldi), Bahatia (S. de Sacy), Hebath (D’Herbelot),1 Bihatia (Hammer-Purgstall). Briggs says he has failed in fixing the position of this place. Elphinstone says, “ a dependency of Lahore, at the southern side of Multan.” Bird says it is now called Bhatnfr, situated on the northern extremity of the Bikanfr desert. Reinaud says it is to the south-east of Multan, and in the middle of an arid country, apparently on the testimony of ’Utbi, but he makes no such assertion. Hammer-Purgstall conceives it to be the present Bahawalpur. But how could a dependency of Lahore be on the southern side of Multan, itself independent ? How could Mahmud advance over all the Panjab rivers to attack a city in a desert? Or Bahawalpur, leaving a country full of hostile and martial populations in his rear? How could Bijf Rai, deserting his fort, “ take post in a wood on the Indus,” as Firishta says, if Bhatia were on the other side of the Sutlej ? or how could he “ take refuge on the top of some hills,” as ’Utbi says, when there are no hills within a hundred and twenty miles from either place ?

Here again we must correct the reading, and all becomes explica- ble and easy. The real name of the place is Bhera. It lies on the

1 D'Herbelot in one part of his article on Mahmud speaks of his deriving immense plunder from Baarea, the strongest fort in India.

eft bank of the Jailam, under the Salt range. It bears evident marks of great antiquity, and has on the opposite side of the river the extensive ruins of Buraria, above Ahmadabad, which strike every beholder with astonishment. The only works which read Bhera are the Khuldsatu-t Tawar'ikh and its followers the Alchbar-i Muhabbat, etc. That Dow’s copy of Firishta must have been very near it, is evident, for, although Mahmud advances against the city of “ Battea,” he is made by a strange inadvertence to take the city of “ Tahera.” ’Utbi [and Ibn A sir] certainly read Bhatia, and A1 Birum mentions Bhatia and not Bhera, but his Bhatia scarcely seems the one we are dealing with.

Whether Bhatia is written by mistake, or whether Bhatia is an old name of Bhera, is difficult to say. The latter is very probable, for the Bhati or Bhatti Rajputs still point to this tract as the place of their residence before their advance to the eastward, and their name is still preserved in the large town of Pindi Bhattian, on the Chinab. It is worthy of remark, as observed by Mr. E. Thomas,1 that of the list of Hindu kings given by A1 Biruni, the four last beginning with Jaipal I. add the designation of Pal to that of Deva, borne by their Brahman predecessors. This would imply the suc- cession of a new tribe, which he considers to be Bhatti Rajput. There is no improbability in this, for there is no authority except that of Firishta for declaring Jaipal to be a Brahman, and Bhatia therefore may have been the local title of the capital of the tribe. Firishta2 makes the Raja of Bhatia to be a different personage from the Raja of Lahore ; but he afterwards tells us that the Lahore dominions extended from Kashmir to Multan — which, as has been shown, includes Bhatia.

It is to be observed, moreover, that Mahmud does not pass through Multan, or the province of Multan, to get there, but passes “ by the borders of Multan,” as Firishta says, or “ crosses the Indus in the neighbourhood of Multan,” as ’Utbi says. Now, as Multan must have extended, as it always has, even down to the days of Mulraj, nearly up to the Salt range, it is probable that Mahmud came from Ghazni by the valley of Banu, and following the course of the

Khuram, crossed tlie Indus near I'sakhel and the old town of Rorf, and so, passing the Sind-Sagar Doab through Mitta Tiwana, reached Bhera by way of Khushab and Shahpur.

A subsequent campaign also indicates the position of Bhera, as will be noticed more particularly hereafter. Meanwhile it is to be observed that Mahmud annexed Bhera to his dominions, which, had it been any place trans-Sutlej, would have been out of the question.

Fourth Expedition. — Multan, a.h. 396. — [Ibn Asir and] the Hab'ibu-s Siyar place the expedition to Bhatia and Multan in the same year, but it is quite evident from the Yam'mi that special preparations were made for this new campaign. Dr. Bird considers that Firishta has misplaced this campaign, and that it should be deferred till after the defeat of flak Khan. I see no reason whatever to doubt that it is correctly ascribed to the year 396 n., and that it has nothing what- ever to do with the invasion which took place after I'lak Khan’s defeat.

We find the governor or ruler of Multan with a Muhammadan name, “Abi-1 Futuh, or “Abu-1 Fath,” and he is not an infidel but a heretic, one “ who introduced his neologies into religion.” There can be little doubt, therefore, that he was a follower of the Karmatian heresy, which we know, from A1 Birum, to have pre- vailed extensively at Multan, and for a long period previous to this invasion. “ He says : “ When the Karmatians became masters of Multan, their chief broke the idol in pieces, and massacred its ministers ; and the temple, which was built of brick, and situated on an elevated spot, became the grand mosque in place of the old one, which was closed on account of the hatred borne against the Ummayide Khalifas, under whose rule it was constructed. Sultan Mahmud, after subduing the Karmatians, reopened the old mosque, so that the old one was abandoned ; and now it is as a plain, destined to vulgar uses.”

The authors which treat of this period do not, — except in a few instances, as the Tabakat- i Akhar'i, and the Khuldsatu-t Tawarikh — • expressly say that Multan was held by Karmatians, but by “ Mula- hida,” a more generic term, which, though it might include Karma- tians, was more generally, at a subsequent period, used to designate

the Isma’ilians.1 For more on the subject of the occupation of Multan at this period, the passages mentioned in the note may be consulted.2

Abu-1 Fath Baud was the grandson of Shaikh Hamid Lodi, who is represented to have done homage to Subuktigin. The word “tribute,” used by Briggs, is not authorized. Elphinstone says that Hamid Khan had joined the enemies of his faith for a cession of the provinces of Multan and Laghman, and submitted to Subuktigin after his victory over the Hindus. This statement is made on the authority of Firishta.3 Daud invited the co-operation of Anandpal, who, being defeated at Peshawar, was pursued as far as Sodra,4 on the Chinab. From Sodra Mahmud goes, by way of Batinda, to Multan, which is so circuitous a route as to be absurd. Here, again, Bliera should be read, which is in the direct line between Sodra and Multan.

Ibn Asir, Mirkhond, and Haidar Eazi make Daud flee away to Sarandip, but ’Utbi says a fine was levied from the inhabitants of 20,000,000 dirhams. Firishta says an annual tribute was fixed on Daud of 20,000 golden dirhams, or dinars, with promise of implicit obedience and abstinence from heresy for the future.

The Biographie Universelle contains a curious statement, respecting this expedition : “ La revolte du gouverneur qu’il avait laisse a Moultan et le debordement des fleuves qui semblait la favoriser, obligerent Mahmoud de demander passage a Andbal. Sur son refus, il le poursuivit a travers le Candahar et le Kaboulistan jusqu’ a Kaschmyre.”5 What Kandahar and Kabulistan have to do with the pursuit is not easy to say. Authors agree in saying Mahmud wished to march through Anandpal’s territory, but it is very difficult to discern the reason of the request, as he had already crossed the

1 Defremery, Eistoire des Seldjoukides, pp. 59, 86, 136-9.

2 Reinaud, Fragments Arabes et Persons, p. 142. Ritter, Erdkunde von Asien Vol. V. p. 6. Renaudot, Anciennes Relations, p. 172. NCtru-1 Hakk, Zubdalu-t Tawdrifch, fol. 366. Mir Ma’sum, Turikh-i Sind, Ch. 2 and 3. Khuldsatu-t Tawdrlkh, v. “ Baber.” Mir-dtu-l Abrdr, v. “ Bahau-d din Muhammad Zakaviya."

Tuhfatu-l Kiram, Vol. III. v. “ Multan.” Hadikatu-l Akdlim, y. “ Dipalpur.”

4 Ilammer-Purgstall identifies Sodra with Weirabad (Wazirabad), but they are two different towns.

5 [This statement is generally supported by Ibn Asir. See supra, p. 24 .]

Indus, beyond the borders of bis territory, and by a route which would lead him more directly towards his object.

’Unsuri informs us that Mahmud took two hundred forts on his way to Multan.

Fifth Expedition .l — Defeat of Nawasa Shall, a.h. 398. — When Mahmud was called away from Multan by I'lak Khan’s invasion of his territory, he left his Indian possessions in charge of Sewakpal, or “ Sukhpal, a son of one of the Rajas of India,” 2 and who, having been formerly made a prisoner in Peshawar by Abu ’All Sanjari, had become a convert to Islam. Sukhpal was taken prisoner by Mahmud’s advance cavalry, and was compelled to pay the sum of 400,000 dirhams ; and being made over, as Firishta informs us, to Tigi'n the Treasurer, was kept in confinement during the rest of his life.3

Dr. Bird says that there was no sucli expedition as this, and that Firishta has confounded it with the previous expedition to Multan ; but as it is mentioned by ’Utbi, Mirkhond, and Khondamir, as well as by Firishta, there is no reason whatever to discredit it.

Dr. Bird adduces, as an additional proof of confusion, that the name Nawasa, “ a grandson,” belonged to Abu-1 Fath Daud, because he was a grandson of Sheikh Hamid Lodi ; but there is no ground for saying that Daud was so called, as the name might have belonged just as well to the grandson of Jaipal, as of Sheikh Hamid. He apostatised to idolatry, after being converted, whereas Daud could only have apostatised to the Karmatian heresy, and not to idolatry and plural worship. The designation of Nawasa is considered doubtful. His name was Sewakpal or Sukhpal ; Ritter says Samukkel. Dow reads “ Shoekpal, who, on conversion to Islam,

1 [Under the year 397 h. Ibn Asir gives the following brief account of this expe- dition : — “"When Yaim'nu-d daula had finished (his differences) with the Turks he went on a campaign to India. The cause of this was that one of the sons of the sovereign of India named Nawasa Shah had become a Musulman under the hands of Mahmdd, and had then been appointed ruler over part of Mahmud’s conquests in that country. After Mahmud had retired he apostatized from Islam and assisted the infidels and rebels. When Mahmud approached, the Hindu fled before him, so he again occupied the country, brought it once more under the rule of Islam, appointed one of his officers over it, and then returned to Ghazni.”]

2 [These are the words of Firishta according to the lithographed edition of the text.]

8 Haidar BAzi says that Mahmud came to Naubar, in pursuit of the rebel, who

fled to the remote parts of Hind, on learning his approach.

assumed the name Zab Sais.” D’Herbelot has “Nevescha;” S. de Sacy, “ Nawaschteh ; ” Wilken, “Nuvasch Shah.” The Tabakdt-i Akbar'i says, “ Sukpal, the grandson of the Raja of Hind.” The readings in Firishta are by no means uniform. They are Ab Sahara, A'bsar, Ab bashaer and Zab sa. The Tar'ikh-i Alfi, and some other authorities, make it Zab Sais or Zab Shah. Hammer-Purgstall says, “Ssabsa or Schiwekpal.” All these are changes rung upon the word “ nawasa,’ or “ grandson,” especially “ a daughter’s child.” Bird says, Price is mistaken in calling him Nawasa Shah ; but ’Utbi gives this name, and there is no reason why we should reject it. It may have been bestowed upon him by Mahmud as a mark of endearment, and Shah, “ king,” may have been added as a term of aggrandize- ment, or it might have been Sah, a common title of respect. But what is more probable than all is that he was the grandchild (by a daughter) of Jaipal, because, in ’Utbi’s account of the expedition to Kanauj, we find Bhim Pal, the great-grandson, complaining that his uncle had been forcibly converted to Islam. Sukh Pal, therefore, was the name, Nawasa the relationship to Jaipal, and Sah the honorific title. He was probably one of the relations of Jaipal, made over by him as hostage to Mahmud ; and that, perhaps, was the period of conversion.

The movement by which his seizure was effected was so rapid, and a new invasion of India was entered upon so soon after, that it is probable the scene of the transaction was the valley of Peshawar.

Sixth Expedition. — Waihind, Nagarkot.1 a.h. 399 (1008-9 a.d.). — It will be observed that the account of the commencement of this expedition is described very differently in the Yamim, the Ilabibu-s Siyar ami Firishta. I prefer, as on former occasions, the former, the river of Waihind, or the Indus, being a more probable place of action than Peshawar, which was then within the Muhammadan border. That the Gakkhars may have performed the part assigned to them is probable enough, whether the action was fought at one place or the

1 [Ibn Asir places this campaign in the year 368, and says that Mahmtid en- countered Brahman-pal on “ the banks of the river "Waihand (which is changed in some MSS. to Handmand). Many men were lost in the waters, and the Hindus were near gaining a victory, when God made the Musulmans to triumph. Mahmud pur- sued the foe to Bhim-nughur (Bhim-nagar) , which he took, and gained immense plunder.”]

other; but that the Gakkhars are the ancestors of the modem Jats, as Dr. Bird asserts, is altogether a mistake, and likely to lead to serious errors.

About the proceedings at Nagarkot all accounts agree, and that Nagarkot is the same as Kot Kangra can admit of no doubt, for the name of Nagarkot is still used. Its position is well described, and corresponds with present circumstances. The impassable waters which surround it are the Ban-ganga and the Biyak. The town of Bhim, which is about a mile from the fort, is now on the spot called Bliawan, which means a temple raised to a Sakti, or female deity, and Bhim is probably a mistake arising from its presumed founda- tion by the heroic Bhim. M. Beinaud considers that it was called Bhim-nagar from Sri Bhima deva, of the Kabul dynasty. The different forms which the name assumes in different authors are shown at p. 34. Elphinstone is mistaken in saying that Nagarkot derived peculiar sanctity from a natural flame which issued from the ground within its precincts. This flame is at Jwala-mukhi, fifteen miles distant, where carburetted hydrogen issues from the sandstone rocks, and fills the superstitious pilgrim with awe and veneration. These jets of gas are made to burn with increased vigour by the removal of plugs, whenever a distinguished visitor is likely to pay well for this recognition of his superior sanctity.

Dr. Bird, who has given a most critical examination of these invasions, says that the capture of Nagarkot and the previous action beyond the Indus occurred in two different years. He observes : “ If we might trust Firislita, Mahmud at this time (after the battle of Peshawar) marching into the mountains captured the celebrated fortress of Nagarkot. It was not, however, till the following year, a.h. 400, according to the Tabalcat-i Alcbari and Hab'ibu-s Siyar, that this expedition was undertaken ; and as the hostile armies prior to the last battle had consumed three or four months in operations west of the Indus, it is not probable that Mahmud could have marched into India at the commencement of the rainy season. The Hijra year 399 given for the march to Peshawar, or the previous year a.d. commenced the 5th September, a.d., 1008 ; and as the spring season, when he left Ghazni, would not commence till a.d. 1009, he must have spent the summer in Kabul, and set out for Hindustan about October.”

I cannot trace in the Tabalcat-i Akbar'i and the Siabibus Siyar the assertion attributed to them ; but let us leave these inferior autho- rities and refer to the Yam'ini. There we find that it is in pursuit (of the flying enemy) that Mahmud went as far as the fort called Blnmnagar.” The campaign, therefore, must have been continuous, and there was no break between the action trans-Indus and the capture of Nagarkot. He has already traversed the same road as far as Sodra on the Chinab, and he would only have had ten or twelve marches over a new line of country.

In these enquiries we must be very cautious how we deal with the word “ spring.” Both Bird and Elphinstone speak of the conquerors setting out in the spring of a Christian year, but the spring of a Ghaznivide invader is the autumn of the Christian year. It is the period when the breaking up of the rains admits of warlike opera- tions. It is the Dasahra of the Hindus, and the season of the com- mencement of their campaigns. So, in the first decisive action against Jaipal, we find Mahmud leaving Ghazni in August, and fiechting: the action at Peshawar in November. And so here we find him leaving Ghazni on the last day of Kabf’u-l akhir, or the end of December, which, though unusually late in the season — so late, indeed, as to render marching in the uplands almost impossible — would still have enabled him to fight his action on the Indus at the beginning of February. He might then have completed his operations at Kangra before the end of March, and have left India again before the severe heat commenced. The only difficulty about the whole campaign is his leaving Ghazni in the heart of winter ; but that the action on the Indus and the one at Nagarkot occurred in the fair weather of the same year, there is no sufficient reason to doubt.

The opening part of the expedition is mentioned in more detail by Firishta, than by ’Utbf and Khondamfr. His account is as follows : —

“ In the year 399 h., Mahmud having collected his forces, deter- mined again to invade Hindustan, and punish Anandpal, who had shewn much insolence during the late invasion of Multan. Anandpal hearing of his intentions, sent ambassadors on all sides, inviting the assistance of the other princes of Hindustan, who now considered the expulsion of the Muhammadans from India as a sacred duty.

Accordingly, tlie Kajas of Ujjain, Gwaliar, Kalinjar, Kanauj, Dehli, and Ajmiv entered into a confederacy, and, collecting their forces, advanced towards the Panjab with a greater army than had ever taken the field against Amir Snbuktigin. Anandpal himself took the command, and advanced to meet the invader. The Indians and Muhammadans arrived in sight of each other on the plain of Pesh- awar, where they remained encamped forty days, neither side shewing any eagerness to come to action. The troops of the idolaters daily increased in number, and aid came to them from all sides. The infidel Gakkhars also joined them in great strength, and made extra- ordinary exertions to resist the Musulmans. The Hindu females, on this occasion, sold their jewels, and sent the proceeds from distant parts to their husbands, so that they, being supplied with all neces- saries for the march, might be in earnest in the war. Those who were poor contributed from their earnings by spinning cotton, and other labour. The Sultan perceived that on this occasion the idolaters behaved most devotedly, and that it was necessary to be very cir- cumspect in striking the first blow. He therefore entrenched his camp, that the infidels might not be able to penetrate therein.

Mahmud, having thus secured himself, ordered six thousand archers to the front to attack, and endeavour to draw the enemy near to his entrenchments, where the Musulmans were prepared to receive them. In spite of the Sultan’s precautions, during the heat of the battle, 30,000 infidel Gakkhars, with their heads and feet bare, and armed with spears and other weapons, penetrated on two sides into the Muhammadan lines, and forcing their way into the midst of the cavalry, they cut down men and horse with their swords, daggers, and spears, so that, in a few minutes, they slaughtered three or four thousand Muhammadans. They carried their success so far that the Sultan, observing the fury of these Gakkhar footmen, withdrew himself from the thick of the fight, that he might stop the battle for that day. But it so happened that the elephant upon which Anandpal rode, becoming unruly from the effects of the naphtha- balls and the flights of arrows, turned and fled. The Hindus, deeming this to be the signal for flight on the part of their general, all gave way, and fled. ’Abdu-llah Taf, with five or six thousand Arab horse, and Arslan Jazib, with 10,000 Turks, Afghans,

and Kliiljis, pursued the enemy for two days and nights, so that 8,000 Hindus were killed in the retreat. Thirty elephants and enormous booty fell into the hands of the pursuers, with which they returned to the Sultan.” 1

Seventh Expedition. — N drain, a.h. 400. — The TalaMt-i Akbari and Firishta do not mention this expedition at all ; but it is recorded in the Yam'm'i, Rauzatu-s Safa and the HaMbu-s Siyar. The latter gives no name, but mentions an invasion of Hind in a.h. 400, between the transactions at Nagarkot and Ghor.

It is not easy to identify the place. ’Utbi speaks of it as in the middle of Hind, where chiefs were reduced who up to that time had obeyed no master. Mirkhond calls it “ Narin; ” S. de Sacy has “ Nardin,” which he thinks there is reason to believe was situated in a part of India to the west of the Indus. This would be probable enough had it not been declared by ’Utbi to be in the heart of India, and a country of hill and valley. Hammer-Purgstall speaks of the “ Maharaja of Nardin.” Eeinaud confounds the campaigns of Narain and Nardin.

On his return to Ghazni, after this expedition, Mahmud received an embassy from the ruler of Hind (Jaipal), offering an annual tribute of fifty elephants, laden with rarities, and an Indian force of two thousand men — a curious stipulation, proving how early Indians became mercenary soldiers, even under their most bitter persecutors. This shows that this particular expedition must have made a great impression on Jaipal, and induced him to sue for humiliating terms.

It is barely possible that the Narin,2 between Inderab and Kunduz, may be indicated. It is the same longitude as Kabul, which we know to have been then comprised in India ; and, with reference to Balkh and Ghazni, it might have been considered so far to the east- ward and so difficult of access, as to deserve being spoken of as in the heart of Hind. In Istakhrx’s map of Khurasan, the position is almost included within “ Bilad Hind,” and its neighbourhood to

1 [This and the other passages from Firishta, are taken from Briggs’ translation, but I have compared them with the text, and have made the translations more literal and exect. — Ed.]

2 This town is not mentioned by the Arab geographers, but it was passed by Lieut. Wood. See his Journey to the Oxus , p. 409.